logotipo ISR
Map-marker-alt Phone
Buscar
Cerrar este cuadro de búsqueda.
MENU
logotipo ISR
Map-marker-alt Phone
Buscar
Cerrar este cuadro de búsqueda.

Industry Production

The industrial landscape is fiercely competitive, a well-known fact in any sector. This competition drives companies to meticulously control time and cost, especially in the realm of mass production. Industrial production refers to the process by which raw materials are transformed through a series of controlled manufacturing processes into products that are produced in large volumes annually. The goal is to create a final product with added value for its target market, often requiring tens of thousands of parts or components each year.

From this concept, two fundamental ideas emerge that define the keys to the industrial environment:

  • Competitiveness: This refers to everything an industry or factory does to meet its demand, i.e., satisfy its customers’ needs efficiently and effectively.
  • Profitability: This is derived from the difference between production costs and the sale price of the product, driven by commercial activities.

Understanding these concepts is essential before analyzing the factors that influence the competitiveness of factories. The success and evolution of a factory hinge on these factors, which also affect its socio-economic environment. It is important to remember that any variables negatively impacting industrial competitiveness will directly influence profitability. This focus on sustained profitability is critical for the viability of any project, whether industrial or otherwise. So, what are the factors that influence the competitiveness of factories? Let’s delve into them!

Production vs Quality

The ideal title for this section could be «Quality within Production,» but the reality in mass production is often quite different. Achieving high standards of quality is a challenging task that every factory in the world strives to accomplish. However, it’s not always possible to achieve this in the short term, or sometimes even in the long term. Why does this happen? The primary reason is complexity—integrating advanced processes, coordinating with different machines and suppliers, and maintaining reduced cycle times.  concepts is essential before analyzing the factors that influence the competitiveness of factories. The success and evolution of a factory hinge on these factors, which also affect its socio-economic environment. It is important to remember that any variables negatively impacting industrial competitiveness will directly influence profitability. This focus on sustained profitability is critical for the viability of any project, whether industrial or otherwise. So, what are the factors that influence the competitiveness of factories? Let’s delve into them!

This complexity often leads to a scenario where production demands clash with quality standards, creating what feels like a war within the factory. Quality is not just a requirement; it is the greatest demand from customers. However, production is equally critical, sometimes leading to controversial decisions where the need to meet production quotas compromises quality. This tension is particularly evident in the early stages of mass and serial production. However, as processes become established, quality typically improves significantly. 

Automatic inspection in quality control ​

To navigate this conflict, factories implement rigorous quality control measures as the main strategy to mitigate the Production-Quality conflict. Quality control is not a new concept; it has been an integral part of manufacturing for centuries, evolving from traditional, manual operations to the fully automated solutions seen today.  

 

The shift towards automatic inspection in quality control has brought numerous advantages to the industry. Some of the key benefits include: 

  • Elimination of Non-Value-Added Tasks: Automating inspection processes removes mundane tasks that do not add value, allowing human workers to focus on more critical aspects of production. 
  • Digitalization of the Process: Automation facilitates the digitalization of quality control, making data more accessible and easier to manage. 
  • Improved Data Access and Control: With digital processes, companies can more effectively monitor and control data, leading to better decision-making. 
  • Reallocation of Human Resources: Automation allows for the relocation of personnel to areas where they can be more productive and impactful. 
  • Objective Results: Automated systems reduce the subjectivity in quality assessments, leading to more consistent and reliable results. 
  • Investment Payback: While the initial investment in automation can be significant, the long-term financial benefits, including reduced labor costs and improved efficiency, often result in a favorable payback period. 

OIT systems of ISR Specular Vision accomplish precisely these benefits, being particularly the focus of the company the experience of the user. As mentioned in the previous article, The influence of quality control in production optimization, the improved data access and control has also a strong relationship with the optimization of the whole process, looking for the highest standard in quality from its roots. 

What is the payback?

The financial aspect of adopting automated quality control systems cannot be overstated. Companies often look at the return on investment (ROI) when considering such technology. The payback period is the time it takes for the savings and increased profitability to cover the initial cost of the investment. This period can vary depending on the scale of implementation and the specific industry, but in many cases, the investment is recouped within a few years, if not sooner. The long-term benefits, including enhanced product quality, reduced waste, and greater customer satisfaction, often outweigh the costs. 

For ISR Specular Vision technology, normally the payback is between 1 and 2 years, depending on the country where the technology is exported. This short period of amortization is crucial for the decision on investment of purchase departments of the customers and is achieved thanks to the efforts of designers and efficiency on the manufacturing process of ISR. 

Summarizing the key points

In today’s article, we explored the intricate relationship between production and quality in the industrial sector, emphasizing the financial drivers behind them. To summarize: 

  1. Competitiveness and Profitability: These are the two pillars that support industrial production, with competitiveness focusing on meeting customer demand and profitability arising from the efficient balance between production costs and sales.
  2. Production vs. Quality: The tension between meeting production quotas and maintaining quality standards is a significant challenge in mass production, particularly in the early stages.
  3. Automatic Inspection in Quality Control: Automation offers numerous advantages, including improved efficiency, data control, and financial payback, which can help resolve the production-quality conflict.
  4. The Future: The future of industrial production will likely see an increased reliance on advanced technologies to further harmonize production efficiency and quality control, ensuring sustained profitability and competitiveness. 

 In conclusion, the journey towards balancing production and quality is ongoing, but with the right strategies and technologies, industries can achieve both high efficiency and superior quality, ultimately leading to greater profitability and customer satisfaction. 

In the realm of steel production, ensuring the quality of every inch of material is a critical task. The demands for precision in detecting surface defects, such as rust, oxide, or edge imperfections, are stringent. The OIT Steel Inspector automates this complex process, providing an advanced solution for surface inspection. This article delves into the key capabilities of the system, exploring the rapid image capture, data processing efficiency, and technological innovations that enable the flawless production of high-quality steel. 

Let’s set the scene: Imagine you’re overseeing a massive steel production line. Every inch of that gleaming metal rolling past you needs to be perfect—free from the slightest hint of rust, the smallest trace of residual oxide, or the tiniest edge defect. It sounds daunting, doesn’t it? This is where our OIT Steel inspector steps in, turning what could be an overwhelming task into a streamlined, automated process that ensures nothing short of perfection. 

In this article, I’ll guide you through the fascinating world of automatic surface inspection systems by exploring the key figures that define their cutting-edge capabilities. We’ll delve into the lightning-fast speeds at which images are captured, the impressive rates at which data is processed, and how these technologies guarantee that the steel you produce is not just up to standard, but absolutely flawless. So, whether you’re soaking up the summer sun or comfortably settled in your office, I invite you to relax and join me on this deep dive into the artistry and precision of steel surface inspection.  

Resolution: Seeing the Invisible

Let’s dive into one of the most crucial parameters of our steel inspection system: image resolution. Imagine trying to spot a small residual oxide stain on a vast sheet of steel—one that’s only 100 microns in diameter.
For our OIT Steel Inspector to spot these minute flaws reliably, the image needs to be detailed enough so that a defect with a diameter of 100 microns occupies at least 13 pixels—think of it as needing a picture to be clear enough to distinguish every tiny detail from simple surface roughness. 

This translates to a resolution of 25 microns per pixel. In simpler terms, that’s 40 pixels per millimeter. Let’s put this into perspective. Picture a grain of table salt. Each grain typically measures around 200 microns across. With our image resolution, this grain would be represented by around 50 pixels in area. So, if a speck of salt were to land on your steel sheet, our system could easily pick it up as a distinct, visible entity. 

Information: The key for defect spotting

Some of you might be wondering, «Does this resolution really mean that this automatic system can detect a single grain of salt on a steel coil that’s 1,000 meters long and 1.6 meters wide?» The answer is a clear and resounding yes. 

To achieve this, the system needs to capture and process approximately 160,000 images covering both sides of the entire steel coil. The total data acquired and processed for just one coil amounts to around 4.65 terabytes. 

To truly appreciate the enormity of 4.65 terabytes of information, let’s put this figure into a more relatable context. Imagine you have a collection of high-resolution photographs from your last vacation, each one about 10 megabytes in size. With 4.65 terabytes, you could store approximately 465,000 of these vacation snapshots. That’s enough to fill dozens of photo albums with every sunset, landmark, and candid moment you captured! 

Now, consider a more everyday example. Think about streaming your favorite TV show in high definition. Assuming an HD movie is around 3 gigabytes in size, 4.65 terabytes would allow you to stream about 1,550 movies. That’s over four years’ worth of non-stop movie watching—definitely enough to keep you entertained on a long, rainy day! 

Speed of Inspection: Capturing Steel at Lightning Speed

Let’s now get focused on the steel moving speed. To get some round figures and considering average sector speeds, we will think at the material moving at 120 m/min. This means that each second the material moves 2 meters. As the linear cameras acquire a row of pixels each 25 microns, the final acquisition speed will be around 80 000 acquisitions per second, which means that an acquisition is performed each 12.5 microseconds. Bees are known for their incredibly rapid wing beats, typically flapping their wings at a rate of about 200 to 300 beats per second. Let’s use an average of 250 wing beats per second for this comparison. This means a single wing beat takes about 4 milliseconds (4,000 microseconds). So, in the time it takes for a bee to complete a single wing beat, our system could perform 320 different acquisitions. This incredible speed allows the system to monitor and analyze the steel surface with exceptional detail and precision, much faster than the blink of an eye (350 000 microseconds), or even the rapid movement of a bee’s wings. 

Processing Power: Handling Massive Data with Cutting-Edge Technology

Given the resolution of 25 microns per pixel, a speed of 120 meters per minute, and the inspection of both sides of the steel coil, our system processes an astonishing 572 gigabytes of information every minute. After capturing this immense amount of detailed data, the next challenge is processing it with speed and precision. 

To manage this info rate, our system relies on the simultaneous operation of eight state-of-the-art processors. These processors work in parallel, ensuring that the data is not only handled efficiently but also analyzed in real-time. This parallel processing approach allows the system to quickly detect and identify any imperfections, such as residual oxide stains or edge defects, without slowing down the production line. 

Just the processors consume 3000W electrical power for performing the calculations for processing this info rate. Consider a powerful hairdryer, which might use around 1,800 to 2,000 watts. At 2,800 watts, you’re looking at a device with significantly more power, capable of drying your hair in record time. 

This incredible processing power is what allows our system to maintain its high-speed inspection capabilities while ensuring that every detail is checked. It’s the backbone of our technology, enabling the flawless production of steel that meets the highest standards of quality. 

Additional Key Figures: Precision and Protection

Beyond its impressive speed and resolution, the OIT Residual Oxide Detector stands out with several critical features that highlight its advanced engineering. 

For protection, the system is shielded by 1,200 kilograms of steel tubes to withstand potential impacts from the moving material, ensuring reliable operation even in harsh conditions.  

The cameras, with a depth of field of just one millimeter, require precise positioning. This is achieved through active positioning technology using electric actuators and ultrasonic sensors, which continuously adjust the camera’s distance from the steel surface to maintain image quality. 

Previously, inspection performed by quality inspectors was limited to just five points on the coil, covering about one meter of material—only 0.5% of the entire surface. With our system, however, 100% of the steel surface is inspected, ensuring that every inch meets the highest quality standards. 

As a testament to the system’s capacity, it inspects an average of 30,200 tons of steel every day. These figures underscore the system’s precision and reliability, making it an essential tool in modern steel production. 

Conclusion

As we’ve explored, the OIT Steel Inspector represents a significant leap forward in the world of steel production. With its resolution, lightning-fast image capture, and immense processing power, it ensures that every inch of steel is inspected with meticulous precision. No longer are we limited to sampling small portions of the coil—now, 100% of the surface is scrutinized, guaranteeing that each sheet meets the highest standards of quality. 

The integration of cutting-edge technology, from robust steel shielding to sophisticated camera positioning systems, underscores the level of innovation at ISR Specular Vision. This system is not just about detecting defects; it’s about redefining the standards of steel inspection. By inspecting an average of 30,200 tons of steel daily, we’re ensuring that the materials used in countless industries are free from flaws, enhancing the safety, durability, and aesthetic appeal of the final products. 

In an industry where precision is paramount, the OIT Steel Inspector stands as a beacon of excellence, embodying the future of steel inspection—where nothing is left to chance, and perfection is the new norm. 

This article defines the concept of Specular Zero, developed by ISR Specular Vision. This concept is born since ISR can automate the aesthetic inspection of components and parts with complex geometries, such as those produced by TIER 1 companies, improving both the digitization of quality control and the quality of the data generated. 

This level of quality control automation, together with the improvement in the quality of the collected data, allows the use of AI along with Big Data techniques to compare production quality with the main parameters of the production process, material, and environment, ultimately seeking the optimization of production parameters to achieve a better quality, energy efficiency, reduce the carbon footprint of the final product, and increase the associated production benefits. 

Specular Zero is based on pre-trained AI software that can make automatic recommendations for production optimization. For this purpose, quality data collected in the OIT Systems OIT Systems , developed by ISR specifically for inspecting components and parts surfaces, are cross-referenced with data from the Autonomous Production Units (commonly known as UAP). Production data, in these cases, includes all information generated in the injection molding machines, the ovens, the coating systems and any other processes involved in the manufacturing lines. Material data is marked by the manufacturer, environmental data is collected through various sensors located in the UAP or directly taken from an internet source, and injected geometry data (geometry complexity, surface concavity and convexity, thickness at various critical points, etc.) is taken from the part’s CAD. 

How Specular Zero works?

The value of Specular Zero in the manufacturing sector is significant. The ability to detect and correct aesthetic defects in manufactured parts in real-time not only improves the quality of the final product but also reduces waste and reprocessing costs, which leads towards a sustainable production. 

The optimization of the production process aims to reduce energy consumption and the carbon footprint, which is crucial for worldwide companies committed to sustainability and innovation. The implementation of Specular Zero can translate into substantial economic benefits and a significant competitive advantage for customers, enhancing the experience of production and reinforcing its leadership in markets such as the automotive components, high value metal surfaces or glass. 
In summary, the application of the Specular Zero concept offers an advanced and efficient solution to improve product quality and optimize production processes, aligning with sustainability and efficiency trends in the industry. This approach ensures that the company’s specificities and needs are addressed precisely and effectively. 

Inputs 

  • Inspection Standard: The criteria or guidelines used to inspect and measure the quality of parts. 
  • Material: The raw materials used in the production process. 
  • Productive Process Parameters: The settings and conditions under which the production process operates.

Defects Prediction 

  • A central component that takes inputs from the inspection standard, material, and productive process parameters to predict potential defects in parts. 
  • This prediction helps in improving the OK Parts Ratio, which represents the proportion of parts that meet quality standards. 

Energy + Material Expenses 

  • The costs associated with the energy consumption and materials used in the production process. 
  • These expenses are influenced by the productive process parameters. 

Cycle Time 

  • The duration required to complete one production cycle. 
  • Like energy and material expenses, the cycle time is also influenced by the productive process parameters. 

Outputs 

  • OK Parts Ratio: The outcome of the defect prediction process, indicating the percentage of parts that are free of defects and meet quality standards. 
  • Profit (k€/h): The financial gain per hour, which is positively impacted by a higher OK Parts Ratio and optimized energy and material expenses. 
  • Pollution (g CO2 / part): The environmental impact, measured in grams of CO2 emissions per part. This is reduced by efficient energy and material usage and optimized cycle time. 

Flow and Impact 

  • The Inspection Standard, Material, and Productive Process Parameters serve as inputs to the Defects Prediction process. 
  • Defects Prediction improves the OK Parts Ratio, which in turn increases Profit and decreases Pollution. 
  • Energy + Material Expenses and Cycle Time are directly influenced by the Productive Process Parameters. 
  • Both Energy + Material Expenses and Cycle Time have a direct impact on Profit and Pollution. 

The model in Figure 1 showcases the interdependent nature of various production factors and their collective impact on the efficiency, profitability, and environmental footprint of the manufacturing process within the Specular Zero concept. 
                                                                                                            Figure 1:

How AI is used in Specular Zero?

The implementation of Specular Zero, developed by ISR, is based on the powerful combination of Artificial Intelligence (AI), particularly Machine Learning (ML) techniques, and Big Data. This section explores how these technologies contribute to the optimization of the manufacturing processes, achieving an improvement in product quality, ultimately leading to Zero Defect Manufacturing (ZDM). The use of big data and AI to help guide manufacturing decisions is a central aspect of ZDM. 

Specular Zero concepts, leveraging machine learning (ML) techniques and big data analytics, is anticipated to yield significant improvements across various facets of the production process. Key outcomes expected from this concept include: 

  • Enhanced Product Quality: By analyzing vast datasets from production processes, ML algorithms will identify patterns and insights that contribute to reducing defects and achieving higher product quality standards. This improvement aligns with the overarching goal of Zero-Defect Manufacturing (ZDM), ultimately enhancing customer satisfaction and operational efficiency. 
  • Real-Time Decision Support: By deploying ML models on the Specular Zero PC within the factory, real-time quality warnings and production recommendations will be provided. This capability empowers operators with actionable insights, enabling them to make informed decisions swiftly and effectively. 
  • Continuous Improvement and Adaptability: Post-deployment, continuous monitoring and active learning techniques will ensure that the ML models remain accurate and adaptable to evolving production dynamics. This iterative refinement process aims to sustain high performance and relevance over time, aligning with customer’s commitment to innovation and operational excellence. 
  • Alignment with Business Objectives: Throughout the project lifecycle, close collaboration with customers ensures that technical objectives are aligned with overarching business goals. This alignment fosters a strategic approach to leveraging AI and big data, driving tangible business outcomes and reinforcing customer’s competitive position in the automotive industry. 

This concept represents a transformative initiative aimed at harnessing the power of AI and big data to achieve operational excellence and innovation in manufacturing processes. The expected results outlined above underscore the project’s potential to redefine production efficiency, product quality, and decision-making capabilities at manufacturing companies, paving the way for sustained growth and competitive advantage in a rapidly evolving market landscape. 

Summarizing the key points

In today’s article we have seen the potential of Specular Zero technology, summarized in the following points: 

  • Specular Zero project by ISR Specular Vision represents a significant advancement in the application of artificial intelligence (AI) for optimizing processes in the production of industrial processes, focused on enhancing aesthetic quality, energy efficiency, and sustainability at TIER 1 and TIER 2 companies.  
  • We have explored how AI will be utilized from data preparation to the implementation of advanced models, offering innovative and precise solutions for industry-specific challenges. 
  • Specular Zero exemplifies how artificial intelligence can transform traditional industrial processes, improving product quality, optimizing operational efficiency, and promoting sustainability.  
  • This advanced approach not only benefits ISR as a company, but also sets new standards for the future industry, driving innovation and excellence at every stage of the production cycle. 

Importance of quality control in steel industry

Imagine you are stepping into the newly renovated Santiago Bernabéu stadium for the first time. The grandeur and modern design leave you in awe. As you take in the sleek, shiny stainless-steel panels that adorn the exterior, adding a touch of elegance and modernity to the iconic stadium, you cannot help but marvel at the craftsmanship. These panels are not just for structural integrity; they play a crucial role in the stadium’s aesthetic appeal, reflecting light and creating a visually stunning effect. 

However, as your eyes wander across the surface, you notice glaring defect on one of the panels. It is a tiny imperfection, perhaps a scratch or a dent, but it stands out against the otherwise flawless expanse of steel. This single flaw disrupts the seamless beauty, drawing your attention away from the overall magnificence of the stadium. It is a reminder of how even the smallest defect can impact the visual and structural integrity of such a large project. 

This is an unreal scenario since all the approximately six thousand stainless steels sheets have been inspected using ISR Specular Vision Technology installed at Acerinox Europe plant. In any case, it highlights the critical importance of advanced inspection technologies at the stainless-steel sector, used to provide high aesthetic surfaces. Considering this or any other example, such as the stainless-steel panel that covers the door refrigerator you are buying to renew the aesthetics of your home, ensuring every steel panel is perfect is paramount.  

ISR Specular Vision linear inspection systems for steel are designed precisely for this purpose. These systems use cutting-edge technology to meticulously scan and detect even the smallest imperfections, as will be exposed below. By identifying and addressing these defects early in the manufacturing process, these inspection systems ensure that only the highest quality materials are produced, maintaining the aesthetic standards of any project that may arise. Also, by reducing scrap material and material reprocessing requirements, an eco-friendlier productive process is achieved. Apart from its noticeable benefits, this point can be a key factor for a steel productor considering both the perception that the general consumer has of your brand and the restrictive anti-pollution regulations in some countries. 

How does OIT linear inspection systems work?

At ISR Specular Vision, we have been at the forefront of developing advanced inspection technology for the steel sector since 2020. Our journey began with a mission to revolutionize the way defects are detected in the steel industry, partnering with leading companies across Europe and the United States. By leveraging our experience and deep understanding of the industry’s needs, we have created a robust and reliable inspection solution that meets the highest standards of quality and precision. 

Our technology centers around linear image capture systems, a method that scans materials line by line, providing a detailed and comprehensive view of the steel surfaces. This type of imaging technology is particularly effective because it captures continuous, high-resolution images that allow for the detection of even the smallest imperfections. Linear imaging is akin to having a magnifying glass that meticulously examines every inch of the material, ensuring that no defect goes unnoticed. Further info about the linear technology can be found at previous entries of ISR blog. 

Our systems are designed to be adaptable and scalable, making them suitable for a wide range of industrial environments at various applications.  

But let us get real for a moment. Beyond all the theory and technical jargon, ISR Specular Vision adds significant value by customizing these systems for each specific location. Our systems are designed to withstand a variety of challenges, including the hardest impacts, extreme temperatures from annealing ovens, metallic dust from oxide grinders, fluctuating lighting conditions, and a host of other external factors that do their best to disrupt proper functioning.  

As can be seen at the previous images, to withstand all these external factors, all our inspection systems intended for steel sector include the following elements: 

  • Metallic structures for precise positioning of main vision elements. We must keep in mind that for the most precise systems, the Depth of Focus (DOF) of the vision assemblies can be lower than a millimeter. For that reason, cameras positioning during the inspection is going to be a key factor to achieve a precise inspection. 
  • Protection structures for avoiding impacts. Let us consider a continuous line where the end of a 12 mm thickness coil is soldered to the beginning of the following one to be processed. In case the beam is not correctly performed, you will get a horrendous blade moving directly towards your inspection system. Luckily, the protective structure independent from the inspection system will be able to manage this situation (amongst others).
  • Specifically designed linear lighting systems will be able to provide enough lighting for a correct image acquisition (after all, how else could we ensure a flawless surface inspection!?). These systems use precisely calibrated LED arrays to create uniform illumination across the entire surfaces. The controlled lighting enhances the contrast of the images, making it easier for the inspection software to detect even the smallest imperfections. Additionally, the linear lighting systems are adjustable, allowing for fine-tuning of the light intensity and angle to suit different inspection environments and material properties. 
  • All sensitive components, including cameras, controllers, processing units and PLC, are housed within sealed enclosures equipped with positive pressure and air conditioning systems. These measures ensure protection against metallic dust, maintaining the optimal performance and longevity of the equipment in harsh industrial environments. 
  • Linear guides for allowing a maintenance retraction, intended to allow maintenance works over the inspection system or the production line considering short preparing times, which also is conceived as a Key Performance Indicator (KPI) of these inspection systems.  

After diving into all this theory, you are probably eager to see some real-world examples of these solutions in action within the industry. Seeing how these advanced inspection systems operate in various industrial settings will provide a clearer picture of their capabilities and the tangible benefits they offer. So, let us take a look at some of these practical applications and see how ISR Specular Vision’s technology is making a difference. 

Examples of Optical Inspection Technology (OIT) for steel

Now, let us explore our highest precision system, capable of detecting pinpoint defects as small as one hundred microns, even smaller considering the latest updates requested by the system final user. Considering that every system is installed at a specific location, at the following images we can see the same system developed for different scenarios and ensuring its performance, reliability, and maintenance. 

The previous images, highlighting our inspection systems, were undoubtedly impressive from an integration standpoint. They demonstrated how seamlessly our technology can be incorporated into the steel production process. However, now that we have systems capable of withstanding the harsh conditions of steel manufacturing, it’s time to focus on the inspection results themselves. This is where the true value (and payback) lies, ensuring the highest quality standards in steel production. Let us delve into these inspection outcomes and understand the impact they have on the production process. 

Let us first focus on a common defect known as «scale,» which refers to small patches of residual rust on the surface of the material. Any steel, including stainless steel, tends to oxidize during the hot rolling process. To remove this superficial oxide layer, a process called pickling is performed using scraping brushes and acid. If this process is not conducted properly or production parameters are not correctly tuned, residual rust stains can remain. When these stains exceed one hundred microns, they are considered a defect that can significantly devalue the produced material. Detecting and addressing such defects is crucial to maintaining the quality and market value of the steel. 

In the industry, considering that production is a continuous process, detecting these types of defects as close as possible to where they originate is crucial for timely correction. This proximity prevents large sections of the steel coil from being classified as second-quality material. If a human performs the quality control, they cannot be placed near processes involving acid, metallic dust, or the elevated temperatures of furnaces due to occupational safety concerns. Consequently, the operator must be positioned far from the defect’s origin, typically detecting it an average of two hundred meters downstream. By this time, between 20 and 26 tons of material might be classified as second quality, thereby reducing its market value. Additionally, once the operator adjusts the production parameters according to the info being displayed at the User Interface, they must wait another two hundred meters to see how the new settings affect the material’s aesthetic properties. 

 

By reducing this 200-meter gap to just ten meters and equipping the system with the precision of automatic detection, much less material is wasted, and the production process can be adjusted far more efficiently. This proximity ensures that defects are identified almost immediately, allowing for rapid adjustments, and minimizing the amount of substandard material produced (again, a quick payback is achieved for these inspection systems). The result is a more streamlined, eco-friendly, and cost-effective production line, with higher quality output and reduced material loss. 

Summarizing the key points 

In conclusion, the following points can be highlighted: 

  1. At steel sector, precise visual inspection is a key factor considering both product quality standards, cost reductions and carbon footprint of produced material. 
  2. Technology integration, considering production line peculiarities and non-friendly environments, will be a key factor for ensuring inspection system performance.
  3. ISR Specular Vision has been able to develop a cutting-edge system able to fulfill steel industry highest requirements, being able to find a one hundred microns defect in a one-kilometer-long steel coil. 

Defectology, in the context of industrial production, refers to the systematic study, detection, and analysis of defects within manufacturing processes. It involves using advanced technologies and methodologies to identify and categorize these defects in different products. Our defects are not easy ones! The OIT systems characterize defects up to 0,1 mm. In this article we will focus on defects created by injection and coating processes in automotive sector. Let’s go for it! 

Importance of quality control in manufacturing

Quality control in manufacturing is a systematic process aimed at ensuring that products meet predefined quality standards and specifications. It involves various activities and procedures designed to detect and prevent defects, ensuring that the final products are safe, reliable, and meet customer expectations. 

In this regard, ISR Specular Vision operates in the field of computer vision, which is based on various disciplines, from computer science to artificial intelligence. Its main objective is to enable a system to interpret and understand the visual content present in images, like how a human does. 

Progress in this field has intensified in recent years due to various advancements, ranging from hardware topics like high-resolution cameras to software developments such as more complex image processing algorithms and machine learning techniques. This convergence between hardware and software development has allowed computer vision to reach a level of precision and efficiency in various industrial fields. 

Summarizing, quality control in automotive manufacturing is a discipline that is essential for ensuring compliance, and customer satisfaction. In an industry where the stakes are high, and competition is fierce, robust quality control processes are not just a necessity but a strategic advantage that can define a manufacturer’s success. 

For better understanding, it is necessary to briefly reminder the term OIT, the main principle in ISR solutions. OIT refers to Optical Inspection Technology, automatic inspection of the aesthetic quality of parts and components with specular surfaces (transparent and reflective) using artificial vision equipment and patented lighting systems. ISR markets these solutions to multinational clients in the industry sector with high aesthetic standards and stringent quality control. These systems consist of an integration of technologies that, after being designed by a team of engineers, form an industrial inspection system for a wide variety of parts and surfaces. Additionally, these systems are as comprehensive as the inspection and aesthetic quality control needs demanded, as well as the restrictions of the area where the system will be integrated. 

What kind of defects OIT can find?

The OIT (Optical Inspection Technology) systems developed by ISR can detect various defects originating in the production process of injected and coated manufactured parts. Distinguishing them according to their origin, the following defects are identified: 

Injection Machine:

Cold Slugs: 

  • Description: This defect is usually found near the gate where the melt enters the lens, but it can also be found further away from the gate. 
  • Main Cause: Cold slugs are caused when a small “slug” of cooler melt from the molding machine barrel, nozzle, or the hot manifold system, makes its way past the cold slug catch wells in the runner system and gets into the lens. 

Flow Marks: 

  • Description: Visible lines on the surface of the part caused by uneven flow of molten plastic in the mold. 
  • Main Cause: Uneven flow of molten plastic within the mold, usually due to incorrect injection temperature, poor mold design, or inadequate injection speed. 

Weld Lines: 

  • Description: Visible lines on the surface of the part where two fronts of molten plastic meet and solidify. 
  • Main Cause: Meeting of two fronts of molten plastic that do not properly fuse due to insufficient melting temperature or incorrect injection speed.

Scratches: 

  • Description: Surface damage to parts that affect the appearance and perceived quality of the product. 
  • Main Cause: Damage caused during demolding or subsequent handling, such as improper handling or inadequate demolding equipment. 

Warping or Distortions: 

  • Description: Parts that do not maintain their correct shape, affecting their functionality and appearance. 
  • Main Cause: Warping during cooling due to inadequate cooling times or temperatures. 

Foreign Material Inclusions: 

  • Description: Presence of foreign particles within the injected plastic, visible in transparent parts. 
  • Main Cause: Contamination of the plastic material with remnants of other materials or foreign particles during injection. 

Burn Marks or Discoloration: 

  • Description: Dark spots or discolorations on the final part, affecting its aesthetics and mechanical properties. 
  • Main Cause: Overheating of the plastic material during the injection process due to excessively high temperatures or prolonged cycle times. 

Lack of material: 

  • Description: Areas without material in the final part because the mold does not completely fill. 
  • Main Cause: Insufficient amount of plastic injected into the mold, low injection pressure, or a defective mold feeding system, such as poorly designed or blocked flow channels. 

Air Bubbles: 

  • Description: Visible air inclusions in the transparent part, which weaken the structure and affect the appearance. 
  • Main Cause: Air trapped in the plastic during injection, which can occur due to too high injection speed, inadequate mold venting system design, or moisture in the material. 

Coating:

Orange Peel: 

  • Description: Irregular surface texture resembling the skin of an orange, visible on varnished parts. Main Cause: Incorrect application of varnish, usually due to improper varnish viscosity, incorrect spray pressure, or inadequate distance and speed of the spray gun relative to the part. 

Varnish Drops: 

  • Description: Formation of varnish drops or runs on the surface of the part, affecting the uniformity and aesthetic of the finish. 
  • Main Cause: Excessive varnish application, excessively high spray pressure, or improper movement of the spray gun, as well as issues with varnish viscosity. 

Dust Inclusion: 

  • Description: Presence of dust particles on the varnished surface, visible as white spots or imperfections on the part. 
  • Main Cause: Contaminated varnishing environment, lack of proper cleaning in the spray booth, or improper handling of parts before the varnish fully dries. 

Stripes and Scratches in the Varnish: 

  • Description: Surface damage to the varnish affecting the appearance and perceived quality of the product. 
  • Main Cause: Improper handling of parts after varnishing, or use of inadequate handling equipment causing stripes and scratches on the surface. 

Opacity or Haze in the Varnish: 

  • Description: Finish that is not completely transparent, presenting a cloudy or hazy appearance. 
  • Main Cause: Varnish contamination, improper mixing of varnish components, or inadequate drying conditions such as high humidity or incorrect temperature. 

Varnish Peeling: 

  • Description: Areas where the varnish peels or flakes off the part surface. 
  • Main Cause: Inadequate surface preparation before varnishing, such as insufficient cleaning or lack of proper adhesive, as well as issues with varnish quality or formulation. 

Uneven Gloss: 

  • Description: Variation in glossiness of the varnished finish, resulting in areas of the part that are more opaque or shiny than others. 
  • Main Cause: Uneven application of varnish, issues with varnish component mixing, or inconsistent drying due to variations in temperature or humidity in the drying environment. 

Summarizing key points

Summarizing, this is what we learned in this article: 

  • Quality Control: Ensures products meet quality standards, enhancing safety, reliability, and customer satisfaction. 
  • Optical Inspection Technology (OIT): ISR’s OIT uses artificial vision and patented lighting for inspecting aesthetic quality, particularly in transparent and reflective surfaces. 
  • Types of Defects Detected by OIT: 
  • Injection Machine Defects: Includes cold slugs, flow marks, air bubbles, cloudy swirls, weld lines, scratches, warping, foreign material inclusions, burn marks, and lack of material. 
  • Coating Defects: Includes orange peel, varnish drops, dust inclusion, stripes and scratches, opacity or haze, varnish peeling, and uneven gloss. 

Written by: Ángel Troyano

Linear and matrix systems

Anyone involved in the field of computer vision is well aware of the current trends and how the sector has evolved in recent years. Convolutional Neural Networks (CNNs) and Transfer Learning have become part of our daily conversations, allowing us to process images with hundreds of different approaches. Edge Computing is becoming indispensable in increasingly demanding applications with volumes of information to process that grow exponentially year after year. Big Data techniques applied to the data generated by automatic inspection systems enable us to draw conclusions that transform industries. However, we must never lose sight of the foundation of computer vision, which is image acquisition and the proper industrialization of image acquisition technology.

When a company asks for help considering introducing automatic inspection into its productive process, the key points are the system precision and its payback. Focusing on ISR’s automatic aesthetic inspection systems, a well-acquired image where defects are highly visible, and inspection technology adapted to the production will make it possible for to effectively help this company.

At this point, in our quest to align the inspection system with the production process, we will make an initial division of the production processes based on whether discrete parts are produced (e.g., any component used by various TIERs as well as OEMs to assemble a complete vehicle) or if a continuous material is produced (such as a coil of steel, plastic, fabric, or paper).

Differences bewtween linear and matrix systems

Image acquisition available technologies

When the public thinks about image acquisition, they imagine taking out their phone, opening the camera app, and snapping a selfie while enjoying a pleasant moment with family or friends. To immortalize the moment as accurately as possible, users seek a mobile device with a camera that has the highest number of megapixels, meaning a higher resolution. As an engineer who analyzes and appreciates units, I would like to pause on this aspect. We are talking about an image being composed of pixels, and a sensor capable of simultaneously measuring the values of tens of millions of these pixels, which are then organized into rows and columns to form that desired image (for simplicity, we will skip topics such as RGB channels and Bayer color compression).

This simultaneous capture of pixel intensity values across all the rows and columns that make up an image is what we associate with matrix technology today. Matrix sensors are those capable of simultaneously capturing all these pixels, resulting in a straightforward, complete capture.

As an example, this image capture process largely adapts to the production process of a discrete component, such as the transparent covers of headlights and taillights described in the latest blog entry. These pieces will be placed within an inspection system and analyzed by capturing the necessary images.


Although we have successfully implemented automatic inspection within the first type of production process listed above, deeper reflection leads us to the conclusion that this matrix technology will be challenging to integrate into a continuous production process due to its continuous nature—you cannot stop the material!


The sensor (in other words, the camera) needs a specific amount of time to acquire the information that will compose the image. This is known as «exposure time.» If the object or surface being captured moves during this time, it will result in a blurry final image. This is the reason why, when taking a photo of an athlete, arms and legs are not well defined. If this happens in an inspection system, it would fail the first premise we established previously: «defects must be clearly visible in the image».

Let us make some calculations in order to clarify this topic. The typical exposure time for a matrix image in industry ranges between 5 and 50 milliseconds (find further info about image acquisition speed at this link). Consider detecting defects of 100 µm on a steel sheet moving at 120 m/min; in 50 ms, the surface will have moved a total of 100 mm. Since this movement is three orders of magnitude greater than the defect size to be detected, this image acquisition technology is completely unsuitable. Attempting to reduce exposure time by adding additional lighting to the system could potentially lower exposure times significantly, but the associated cost of illuminating sufficiently to achieve this reduction would render the project unfeasible due to increased technical complexity and cost.

This is why optical sensor manufacturers develop linear sensors. Cameras using this technology also produce a matrix image as a result, but it is captured in a completely different manner. While matrix sensors are composed of a total, for example, of 2464×2056 pixels, a linear sensor will be composed of 2048×1 pixels. This linear sensor captures each row of the image individually as the object or surface to be inspected moves, and the camera combines all these rows to produce the complete image of the object.

How does Specular Vision use these Systems?

ISR Specular Vision has differentiated itself from its competitors through the industrialization of cutting-edge inspection technology in complex industrial situations. Therefore, within our portfolio of success stories, projects associated with both matrix and linear technology can be highlighted. In a previous blog entry, we described a successful case in the automotive sector using matrix technology, and in the coming weeks, we will publish another success story in a different sector where linear sensors have been employed. Different images considering automotive aesthetic defects detection can be seen.

Considering systems using linear cameras, it could be highlighted as an example the OIT Residual Oxide Detection system used in the metallurgical industry. This system is intended to inspect stainless steel surface looking for defects from 100 microns (same thickness as a human hair!) to larger visible defects. Images included below are taken using linear technology. As a reminder, the inspected surface is moving while a linear camera acquires a number of rows and puts all of them together in order to compose these images we can see below. There is no difference considering an image acquired using a matrix camera, but industrialization possibilities of linear cameras offer a wider range of possibilities for Optical Inspection Technology (OIT) systems.

In conclusion, the following points can be highlighted:

  1. The foundation of computer vision lies in image acquisition and the industrialization of inspection technology, making the selection of the most suitable image capture sensor necessary for each specific market scenario.

 

  1. Cameras integrating both technologies produce matrix images, although each operates with different timing mechanisms.

 

  1. Matrix cameras capture the entire image at once, whereas linear cameras capture individual rows of the image and combine them to form the final image.

 

  1. Matrix cameras are more suitable for production processes involving discrete parts, while linear cameras adapt better to continuous material production processes. Nevertheless, each case must be carefully evaluated to achieve a comprehensive technical solution for the inspection system.

Automotive Lighting sector

It is not a coincidence that ISR Specular Vision has specialized in automatic quality inspection in the automotive lighting sector. This sector is quite full of surfaces that need to be perfectly checked, which has been traditionally performed by manual inspection. As ISR Specular Vision’s mission is to be the eyes of the industry, automotive lighting’s mission is to be the eyes of the car and the main element to let us drive carefully on the road. 

The market value of the lighting sector is approximately $35 Billion. This is divided into five groups in Europe: Valeo, Forvia Hella, Marelli Automotive Lighting, OPmobility and ZKW. Three main groups in America: Magna Lighting, Flex N Gate and North American Lighting. The other five groups in Asia: Koito Manufacturing, Stanley Electric, Hyundai Mobis, Hasco Vision and Xingyu. Maximum revenues yearly are between 5 and $6 Billions, and all the companies listed above have at least $1 Billion revenue. They are the valuable players, the TIER 1 companies following the automotive OEMs trends and creating the future of cars. This is clearly a great market to develop innovative technology and expand internationally, but why the lighting sector? 

Question to the reader, what is the main characteristic of the lighting manufactured parts? They need to be transparent to let the light pass through, and transparent is the kind of surface where ISR Specular Vision has its innovative applications in the industry. This is the reason it is not only not a coincidence, but a great competitive advantage and niche for Specular Vision technology. 

High-Quality modern vehicles

Lighting has been for a long time a critical component of vehicles on the roads. First for safety, figures from different resources had night vision as one of the first 3 causes for collisions in the US, and since the mid-80s as a focus for car designers to differentiate their brands. However, car manufacturers and suppliers are clearly entering a new lighting era, driven by several factors. 

FAROS DE COCHES

First, is fast electrification of new car sales, due to regulation and consumer demand in the light of decreasing battery costs (electric vehicles, a topic that will be written in future articles). Elimination of the radiator opens a large space to use lighting across the front grille to offer a specific signature to electric cars and illuminate the brand logo. Add to that, the integration of many sensors in the front of the vehicle, emerging new business model which provides the whole front as one integrated piece. These new signatures and design options, with slimmer shapes, will impact and enhance rear lighting too. 

 The direct consequence of the fact exposed above is the need of inspecting more surface all over the parts manufactured by the TIER 1 companies, headlamp or rearlamps. Manual inspection is no longer possible or feasible for inspection of parts up to 2 m long, in cycle times lower than 20/25 seconds.  

Secondly, the luxury car segment is a growing sector. Not only the exterior of the car involves lighting in the OEMs designs. With the development of the autonomous vehicle, automotive interior lighting is being reinvented  to reinforce the feeling of the comfort of being at home. Automotive interior lighting also has a role to play in the new communications between the increasingly automated vehicle and the driver/passenger, to create trust and reassure occupants. 

 Differentiation is achieved thanks to a dedicated visual identity, both outside and inside the vehicle. Onboard ambient lighting is a key element to this brand identity for a car. Current applications are what we call static ambient lighting. Whether monochromatic or color lights (RGB), direct or indirect, both premium and mainstream vehicles are now equipped with static ambient lighting solutions. All segments are concerned. 

 With the evolution towards a home-like cabin in the vehicle, one trend is to merge decorative components and lighting into one backlit system. Backlit decorations create a specific signature when activated, enabling a polymorphic interior, whether thanks to an “augmented” pattern (visible when off but highlighted when on) or reveal one that is hidden for a specific night signature. 

 Automotive interior lighting enables to change infinitely the design of the interior, enabling the owner or users to adapt the vehicle to their mood and taste and to deliver a personalized experience. 

 In the long term, the new developments and updates could deliver new “skin” packages or easter-eggs, offering an evolving experience throughout the lifetime of the vehicle, adding value to the travel experience. Beyond pure decoration, it also combines functional lighting with HMI (Human Machine Interface) projection, content, etc. 

Overview of OIT Headlamps & Rearlamps Systems

 OIT® Headlamps, OIT® Rearlamps  are inspection systems that detect aesthetic defects in headlamps and realamps cover lenses. The system is fully integrated in-line after the injection, coating, and oven processes. Normally the parts are automatically introduced in the OIT cabin with the same injection or coating robot, which handles in parallel several tasks in the line. The system could be integrated right after injection or right after coating, or even after both processes, depending on the strictness of the quality control applied. 

Thanks to their design, these solutions allow the inspection of different models with just a change of tooling (what is called Vision Jig), making it possible to inspect a wide variety of sizes and shapes. Per system and year, the average number of models inspected is 3. Every year, an average of 2 new models are integrated in the systems. 

The automatic inspection is performed following the control plan of the end customer, the OEM. In this document, the main defects to be detected are clearly specified, the different areas to be controlled and the procedures for an operator to perform manual inspection.  

This inspection cycle is divided into the following phases: 

  • 1st phase – Retro image. In this phase, images are obtained from the fixed cameras and the robot cameras to obtain a complete view of the part. The images from this phase are darker images which allow the inspection for black defects. Both fixed and robot cameras. 
  • 2nd phase – Aspect image. In this phase, the same images are obtained as in phase 1 with the difference that the images are brighter to improve the identification of white defects. Both fixed and robot cameras. 
  • 3rd phase – Catadioptric (in case of presence) using reflection. In this phase, images are obtained using robot cameras to analyze the catadioptric surfaces, to analyze white defects that cannot be detected in phase 2 due to the morphology of the catadioptric itself. Robot cameras are used. 

 

DEFECTS DETECTION
DEFECTS DETECTION

The cycle time varies between 12 and 25 seconds. When the images have been processed, the defects are detected in the images showing the following information: 

  • The red color shows the defects whose size exceeds the marginal defects maximum size and, therefore, makes the part NOK. 
  • The pink color shows the defects whose size is between the marginal defects’ minimum and maximum size thresholds. More than one (threshold quantity can be adjusted using the user interface) defect of this size must be detected for the part being classified as NOK. 
  • Green shows a contour which is not considered a defect since its size is smaller than the marginal defect minimum size. 
  • The numbers indicate the biggest dimension (mm) of the defect. 
  • The purple area of the image indicates the area that is out of the Region of Interest (ROI) and is not inspected. 

It must be pointed out that minimum and maximum size of marginal defects can be configured using the system user interface. In industrial conditions, typical values for limit defects are between 0.3 and 0.5 mm. 

 Nowadays, ISR Specular Vision systems inspecting headlamps and rearlamps are 10 all over the world, with 3 different customers. The main car models inspected by ISR are from the following OEMs: Volkswagen, BMW, and Renault. 

Benefits using OIT and comparison with traditional inspection

Reached this point, it is difficult to argue against the benefits and potential of the automatic inspection of Specular Vision technology in lighting cover lenses inspection. In this section, we will focus first on the typical defects found with OIT HL/RL Systems, and, secondly, briefly present the payback cases in the investment of this Capex equipment for factories. 

 Typical defects found in injection process: 

  • Flow Marks or Lines. 
  • Air Bubbles or entrapments. 
  • Black dots. 
  • Cloudy swirls, Opacity or Haze. 
  • Weld Lines. 
  • Scratches, Scuffs or Cold Slugs. 
  • Warping or Distortions. 
  • Foreign Material Inclusions. 
  • Burn Marks or Discoloration. 
  • Lack of material. 

Typical defects found in coating process: 

  • Orange Peel. 
  • Varnish Drops. 
  • Dust Inclusion. 
  • Stripes and Scratches in the Varnish. 
  • Opacity or Haze in the Varnish. 
  • Varnish Peeling. 
  • Uneven Gloss. 

And finally, the clear benefit of the OIT Headlamp and OIT Rearlamp inspection systems is the quick payback obtained in the investment. The traditional inspection costs of these production lines are between 3 and 6 workers per line. The ISR Specular Vision, instead, offers an automatic system whose payback is under the year in the best cases, and between 1,5 and 2 years for worst cases. 

As these two topics could be deeply treated, both will be further discussed in following articles. 

Summarizing the key points

In today’s article, we gave an overview of the lighting sector and introduced the OIT HL and RL systems. It can be summarized in the following points: 

  1. Specular Vision technology has a great competitive advantage and niche in the Automotive Lighting sector.  
  2. Manual inspection is no longer possible or feasible for inspection of parts up to 2 m long, in cycle times lower than 20/25 seconds. 
  3. OIT® Headlamps, OIT® Rearlamps are inspection systems that detect aesthetic defects in headlamps and realamps cover lenses.
  4. Nowadays, ISR Specular Vision systems inspecting headlamps and rearlamps are 10 all over the world, with 3 different customers. The main car models inspected by ISR are from the following OEMs: Volkswagen, BMW and Renault.
  5. The clear benefit of the OIT Headlamp and OIT Rearlamp inspection systems is the quick payback obtained in the investment. ISR Specular Vision offers an automatic system whose payback is under the year in the best cases. 

OIT: Proprietary Registered Technology

“Optical Inspection Technology”, are the words behind OIT. This is the denomination under which ISR Specular Vision has developed the mechatronic system that goes with the technology of Specular Vision. OIT technology is the commodity where the integration of systems take place, where the software finds the hardware to be involved in the final purpose of the automatic inspection. 

 This was not the name used in the beginning. It was first used in 2019, during the development and validation of the first prototype headlamp inspection system for our main customer at that time. The name is composed of three words: 

  • Optical: field of study, the branch of physic, that studies the fundamental properties related to the behavior of light in interaction with matter.  
  • Inspection: the act of looking at something carefully. In the technical or the industry context, to check potential failures. 
  • Technology: the joint of both previous concepts. 

 Why was it adopted? It explained well the devices that we were developing to follow our purpose: OIT systems of ISR Specular Vision are the eyes of the industry, and that is our leitmotiv. 

What is OIT

But what is an OIT system? What is the original and technological approach of the term? ISR was founded in 2016 under the name of (Spanish): Integración Sensorial y Robótica (ISR). The translation: Sensory and Robotic Integration. The OIT is just the integration of a huge sensor, served to industry with the orchestra of the robotic and the automatic sub-systems. The big sensor is, as you can already imagine, the camera system. The origin of the technological approach comes from the computer vision concepts of sensor and image acquisition. 

The importance of the term resides in the technological assets of ISR Specular Vision, particularly in two of them: 

  1. Integration in the production line: OIT systems are fully integrated and automated together with the process and the production lines. 
  1. Flexibility in inspection: we are full owners of our own systems design and development, which means that the requirements of inspection and integration can change completely the OIT system if needed. 

 The orchestra starts playing when the computers (a non-single quantity of them) begin to send orders. One channel is open with the PLCs (Program Logic Controllers), which subsequently send orders to the rest of the actuators, such as the lighting. Another channel is open with the cameras, ready to snap images when the time is right. And, if robots are present, also robots are commanded on what to do, where to move, and how to interact with the part inspected, to reach the established cycle time. This integration, all together, is what means Optical Inspection Technology. 

Advantages of OIT

In 2021, ISR decided to register the brand name of OIT: National brand registered. The reason for that was purely marketing. There are other names for the same technology, for example, EOL (End of Line Inspection), or AOI (Automated Optical Inspection). The key is not in the name but in the technology behind it. 

 In the market, what are the main advantages of the systems such as the OITs developed by ISR Specular Vision? Firstly, as mentioned, the customers enjoy the experience of deciding freely how their process is going to be. They are free of constraints when the definition of an OIT system is made. The customer chooses the best way to integrate the automatic inspection in the factory, which makes the work easier for the engineers involved. At the other end, the engineers and designers of ISR Specular Vision are working closely with customers to fulfill the expectations of industrialization. The result is a fine combination of work between the two parties interested in the automatic inspection, and an OIT system integrated into the production line doing its job: to inspect parts or surfaces 24/7. 

 Secondly, the OIT systems are not black boxes, at least not anymore. The team of software developers in ISR Specular Vision is focusing efforts on the robustness and openness of our systems. We know deeply the needs of our customers, and we try to align constantly with them. The use of AI is contaminating partially the view of the software systems that intersect with human morality or values. In this case, the decisions made by OIT systems on whether a manufactured part is OK or KO is not away from the debate, especially when AI is used in the decision tree. ISR Specular Vision technology has been opening the software since its conception, chasing the goal of having a back-frontend more reliable and friendly to users in the factory, 

  1. To let the customer have the autonomy to control the decisions over the quality-production KPIs. 
  1. To let the customer understand how the vision algorithms work. 
  1. To let them define new defects or integrate new models.  

And still, there is work to do. 

The last point to mention, ISR OITs can be adapted to any kind of manufacturing process, any kind of part or any kind of surface. This is why this technology shows great potential in different sectors. In the following articles, we will treat the process of designing an OIT, understanding how, first, is the detection of defects, and second is the industrialization of OIT systems. They are two related but different things. This path is the key to obtaining the flexibility that ISR Specular Vision can offer in the market. 

Summarizing the key points

In today’s article, we have introduced our view of the concept of OIT. It can be summarized in the following points: 

  1. OIT is the mechatronic system that goes with the technology of Specular Vision. 
  2. Two main assets:                                                                                                                                                                                                 a) Integration in the production line: OIT systems are fully integrated and automated in the process and the production lines.   b) Flexibility in inspection: the requirements of inspection and integration can change completely the OIT system if needed.
  3. OIT systems are not black boxes. Customers can access, understand, and modify software parameters to have autonomy over production and flexibility in new integrations. 
  4. OITs can adapt to any kind of manufacturing process, any kind of part, or any kind of surface. 

What is Specular Vision?

Specular Vision is the brand name created by the company ISR at the beginning of its technology developments (https://isr.es/en/specular-vision-en/). It refers to the property of the light in which a reflected ray of light emerges from the reflecting surface at the same angle to the surface normal as the incident ray. This is only theory (first described by Hero of Alexandria AD c. 10-70), but the surfaces that ISR inspect are not all strictly specular, but it reminds us of this effect of the light.

The importance of Specular Vision technology resides in solving efficiently the automatic aesthetic inspection of the surface that is transparent or reflective, even glossy. We inspect and detect defects in manufacturing parts with industrial cameras, as simple as that. And we make it for several industry cases: automotive, steel, glass, etc.

It is a hard task to inspect manually thousands of components/parts a day looking for tiny defects, with little rest, sleepy eyes, or after 8 standing hours under the heavy temperatures of summer. The added value of the technology is related to the change of vision for inspection tasks, together with the financial benefits, of course, that could report positively to factories and companies’ productivity.

Besides this, Specular Vision is intended not only to check quality isolated from the rest of the process but also to optimize the production. Data regarding quality checks, however, has not been traditionally easy to collect in the factories. Specular Vision comes to help with:

  1. Automatize the process.
  2. Detect and identify the defects.
  3. Collect the data for a better understanding of the production process and its quality.In this article we will try to help you understand how it works, its applications in industry, and examples of ISR Specular Vision. Let’s go for it!
Aestethic Inspection
Imagen1
IMG_2764

How Specular works?

Specular Vision technology is based on artificial vision intelligence. But just that? Not. The technology is not only about artificial vision; it doesn’t remain only at the software level. One of the most valuable keys of ISR Specular Vision is the deep knowledge acquired over the years of the mechatronics environment that surrounds the vision algorithm (A machine vision system for defect characterization on transparent parts with non-plane surfaces).

Why is this important? Well, imagine taking a picture of the Eiffel Tower, in reverse, dark, and blurred. Would you think this picture is something you can post on your social media? No, it’s not! So just the same thing applies to Specular Vision Technology: we are obsessed with image acquisition.

The images in our vision systems are acquired by a number of cameras that synchronize perfectly together to achieve a high-resolution final image of the part inspected. This final image is divided into areas, as many areas as the manufacturing part inspected requires (depending on its size). In each of these sub-areas, the local defects present on/in the surface are identified properly. With this information, our customers can sort the parts whether they are OK or KO (scrap parts) into different conveyor belts or outputs.

This is the base of the technology. But what happens inside our systems? Well, we need to make a difference between two types of vision systems:

1. Matrix systems: the parts inspected leave the production process (for a short period) and enter our cabin to be inspected.
2. Linear systems: the part/surface inspected passes over/under/through our system of cameras to be inspected.

In future articles, we’ll explain the differences between these two approaches.

In general terms, each system has a matrix of fixed cameras (https://www.stemmer-imaging.com/) Their goal is to acquire a bunch of images of the parts or surface, playing with the lighting system. Yes, there are also several lighting systems (which are patented) to create the optimal conditions for the inspection cycle in the cabin. The fixed cameras are not alone, at least in matrix systems. They are supported by one, two, three, or even four robotic arms, each one, having a camera also integrated. The aim of having dynamic cameras is to gain flexibility and speediness in the process.

How much time does it need to inspect 100% of the parts? This concept is called cycle time. The cycle time is the time needed in each production cycle to inspect completely the part surface, regarding the acquisition plus processing phase. Normally ISR Specular technology needs an average of between 10 and 25 seconds to complete the inspection and detection of defects.

After this first approach, it’s important to talk about the differences between the results when the acquisition phase of the vision systems is affected or not by external conditions, such as dust, unstable production, defects on the mold, etc. As said before, acquiring a good image is essential for the processing of the defects, and the external conditions could lead, occasionally, to false positives. A false positive is everything detected in the image that is not really a defect, but the system has “confused” it. It’s a very important topic in machine vision systems. One of the benefits of Specular Vision technology is that this topic is attacked with AI algorithms, able to cope with different situations when facing a decision of OK vs KO parts.

Applications of Specular Vision

Made the introduction! Where are we? Let’s see in this section what are the main applications where Specular Vision technology is used and what benefits it brings:

  • Automotive sector.

ISR Specular Vision technology has been applied to automotive sector since 2016, starting with the surface inspection of lighting lenses. The benefits of Specular Vision have been also found in the same TIER1 companies, manufacturers of headlamps and rearlamps, with the inspection of their covers (https://isr.es/en/solutions/lighting-inspection/), as well as interior decorative parts. Normally, all these parts are injected in tech plastic, which is also a topic for another article.

  • Steel sector.

In 2018, Specular Vision was applied to steel processes. With the use of linear systems, the coils were laminated under the eyes of our systems.

  • Glass sector.

After some iterations with different customers, ISR Specular Vision entered the market of glass inspection (construction or automotive glass). A great market with a huge need for inspection because of its particular and quick payback coming from automatization.

ISR Specular keeps innovating in sectors such as aluminum, metalized parts for decorative purposes, thin plastic material and even EV batteries.  

Examples of Specular Vision in use

The success case that represents the best example of Specular Vision technology is the automatic inspection systems of headlamp covers. It is widely implemented in factories of TIER 1 companies. The system can adapt with highly flexibility to different models in the production process. This is made with what is called Vision Jigs, adaptative elements for the integration of the models. A system handles between 3 and 4 models, all of them interchangeable in serial production. The time needed to change a model varies between 5 and 10 minutes, and the integration development can be implemented in days!

Main features:

  • Automatic detection of defects in serial production.
  • Classification of defects in real-time.
  • Data analysis of your production performance.
  • High flexibility for new models’ integration.
  • Autonomy of the factory workers on production vs quality decisions.
  • Quick payback.

Summarizing the key points

In today’s article we have seen the potential of Specular Vision technology, summarized in the following points:

1. Specular Vision technology efficiently solves the automatic aesthetic inspection of transparent, reflective and glossy surfaces.

2. The benefits of Specular Vision mainly are,
a. Adding value to the inspection task in the industry.
b. Obtaining amortization in the short term.
c. Having data about your production quality in real-time.

3. Specular Vision technology combines industrial cameras, lighting systems, and mechatronics.

4. Markets and sectors of applications are a huge variety.

"INVESTIGACIÓN EN EL EMPLEO DE ESPECTROSCOPÍA Y TECNOLOGÍAS COMPLEMENTARIAS EN LA DETECCIÓN DE FITOSANITARIOS IN SITU EN ACEITUNA ENTERA”

Espectrolive

Objetivo General del proyecto:

Investigar la aplicación de tecnologías de espectroscopía y nariz electrónica para la detección de productos fitosanitarios en aceituna entera in situ con el fin de asegurar el cumplimiento de los límites reglamentados para estos compuestos en aceites de oliva y realizar una supervisión en tiempo real del proceso de producción que permita una rápida identificación de problemas y adopción de medidas inmediatas relacionadas con la presencia de contaminantes, ayudando a reducir el desperdicio y mejorando la eficiencia y el control de calidad de los aceites de oliva, a la vez que realizar una clasificación apropiada del fruto, antes de su molturación, para conseguir aceites con “residuo cero”, muy valorados por el consumidor actual.

Objetivos técnicos

Principales resultados

  • Selección de sensores MOS que tengan sensibilidad a la presencia de las sustancias contaminantes seleccionadas en el proyecto.
  • Diseño y desarrollo, a nivel electrónica y mecánico, de un prototipo de ENOSE que a escala de laboratorio permita poner en contacto los gases emitidos por los lotes de aceitunas con los sensores MOS seleccionados.
  • Realizar medidas con la ENOSE a partir de los lotes de aceitunas y hojas generados.
  • Diseño y desarrollo de una metodología software para el procesado del flujo de información procedente de la ENOSE desarrollada a escala de laboratorio, con el objetivo de calibrar el sistema ENOSE y obtener modelos de clasificación.
  • Configurar el sistema hiperespectral basado en el equipo Pika L 400 – 1000 nm de la casa Resonon y en cinta transportadora de laboratorio.
  • Realización de medidas tipo “pushbroom” sobre las muestras generadas artificialmente.
  • Calibración y validación de modelos de predicción basada en los cubos hiperespectrales adquiridos de las muestras seleccionadas.
  • Identificación de las bandas de frecuencia más discriminantes desde el punto de vista cualitativo.
  • Optimizar los parámetros de adquisición de espectros Raman para mejorar la resolución y precisión de los resultados (maximizar la relación entre señal y ruido).
  • Identificar y cuantificar la concentración de fitosanitarios específicos en las muestras.
  • Desarrollar modelos multivariantes para la clasificación de diferentes variedades de fitosanitarios y su evolución en función del tiempo.
  • Evaluar la sensibilidad y selectividad de la técnica en comparación con otras técnicas analíticas existentes.
  • Validar los resultados obtenidos mediante técnicas de validación cruzada y comparación con los valores de referencia.
  • Definir la configuración óptima para la adquisición de espectros de las aceitunas.
  • Adquisición de los espectros y desarrollo de modelo predictivo. Se tratará de un modelo de tipo cualitativo que detecte presencia/ausencia de fitosanitarios.
  • Como objetivo secundario se desarrollará de un modelo cuantitativo, sin embargo, debido a las bajas concentraciones a detectar las probabilidades de éxito son escasas.
  • Validación de los modelos desarrollados en un entorno de laboratorio.
  • Analizar las diferentes opciones de instalación del sensor NIR en una almazara para evaluar las necesidades de una posible integración en línea.
En tiempo real de aceituna en función 
de la presencia de contaminantes.
Saludables del aceite de oliva como argumento de venta de estos
productos.

Del aceite de oliva por el mejor cumplimiento de las restrictivas normativas sobre la presencia de residuos fitosanitarios en el aceite de oliva.

Del sector oleícola y la industria auxiliar asociada.

En tecnologías de sensórica alineadas con la Industria 4.0.

INOLEO (Coordinador)

SECPHO (SOUTHERN EUROPEAN CLUSTER IN PHOTONICS AND)

AOTECH (ADVANCED OPTICAL TECHNOLOGIES S.L.

ISR (INTEGRACIÓN SENSORIAL Y ROBOTICA SL)

UNIVERSIDAD DE JAÉN (UJA)

UNIVERSIDAD DEL PAÍS VASCO (UPV/EHU)

Ejecución
Plazo: 05/2023 al 04/2024
Presupuesto: 308.010 euros

05/2023 al 04/2024

Importe: 308.010 euros
Ayudas de apoyo a Agrupaciones Empresariales Innovadoras con objeto de mejorar la competitividad de las pequeñas y medianas empresas y se procede a la convocatoria correspondiente al año 2023, en el marco del Plan de Recuperación, Transformación y Resiliencia. Ministerio de Industria, Comercio y Turismo (MINCOTUR)