:::

【2021 Application Example】 Optical industry AOI imports AI Great Leap Forward to completely solve the pain points of lens defect detection

The stay-at-home economy such as smartphones and remote working is booming, and the information and communication industry is booming, driving the optical industry to flourish. However, the defect detection of optical lenses is mostly carried out by human eyes, which is not only time-consuming and labor-intensive, but also limited by the fact that human eyes are prone to fatigue. The misjudgment rate is also a lingering pain point for the optical industry. Benefiting from the evolution of AI technology, Shangyang Optics introduced diffraction optical technology for shooting, used the images captured by the system as the data source, introduced AI model training, and integrated the camera system and image recognition into a production line workstation, greatly improving defect identification The rate is as high as over 90%.

Taiwan’s optical production value accounts for 10% of the world’s, and the application range of precision optics is expanding day by day

The optical industry is a mainstream product in consumer electronics. Even though Taiwan was affected by the Sino-US trade dispute in 2019, the output value of optoelectronics still reached US$46.3 billion, accounting for 10% of the world's total. Among them, the "precision optics" segment accounts for NT$87 billion (approximately US$2.9 billion) in output value. In view of the increase in the number of smartphone lenses, precision optics still maintains a sustained growth of 4% compared to the decline in other fields.

Since Sharp launched the world's first camera phone equipped with a rear 110,000-pixel lens in 2000, end consumers' requirements for smartphone camera performance have continued to increase, and with the wave of 5G high-speed Internet The advent of the technology has led to the activation of application markets such as augmented reality (AR) or virtual reality (VR). The innovation and application of its technology have added a lot of momentum to the optical industry, and the application fields have extended from smartphones to popularization. to the mass consumer markets such as automobiles and home entertainment.

Optical lenses are inseparable from the economic development of "precision optics". As semiconductor technology continues to mature and network speeds continue to increase, optical lenses are used not only in smartphones, tablets, traditional cameras, projectors, In the field of people's livelihood vehicles, the demand for engineering visual inspection and security applications in high-precision manufacturing processes continues to grow rapidly.

Optical lens defects Detection is mostly done manually.

▲ Optical lens defect detection is mostly done manually.

"Optical lenses" are essential components of the overall optical-mechanical system. The lens finish inspection after incoming materials and before shipment not only affects the overall production line efficiency development, but also has an impact on the quality commitment of end customers that cannot be underestimated. For a long time, the optical industry has mostly used human eye detection for defect inspection. As production volume continues to increase, not only labor costs continue to rise. As inspectors age, their eyesight gradually declines, and the misjudgment rate increases every year. In addition, manpower recruitment has been difficult in recent years. Even if they are lucky enough to be recruited, it is not easy to develop the inspection technology, and the training time is lengthy, making it impossible to respond to the production line manpower needs in a timely manner.

Introducing diffraction optical technology and AI training model to improve defect recognition rate to more than 90%

The current market is flooded with a large number of automated optical inspection systems, and there are many substantial cases of lens defects. However, after years of market exploration and evaluation by Shangyang Optics, this system still cannot solve the current manual inspection problem. The main reason is that the appearance of the optical lens is curved and transparent, and it is not easy to photograph various defects, and once the defects are around There is interference from other stray lights, making judgment more difficult. Moreover, different types of lenses need to be individually rotated and lit and adjusted according to the defect status before entering the judgment stage. The labor consumption ratio is still high, which is not in line with the efficiency and cost.

Through this, through the matchmaking of the AI ​​project execution team of the Industrial Bureau of the Ministry of Economic Affairs, Xiaoma Optics assisted Shangyang Optoelectronics in establishing an effective defect photography system. Pony Optics provides guidance on precision diffraction optics. Based on the characteristics of "light" fluctuations, lens defects can be obtained through a unified lens shooting method. Current photography systems on the market mostly use geometric optics. Geometric optics uses linear light and is not easy to capture defects such as missing coatings, tiny scratches, and liquid dirt. The cooperation plan introduces diffraction optical technology for shooting. Through precise imaging from all angles, it can achieve higher contrast and better noise reduction than ordinary geometric optical elements, so as to obtain the necessary defective images.

Image of scratches and defects on the optical lens.

▲Schematic diagram of optical lens scratches and defects.

In order to improve the more detailed defect detection and recognition rate in this case, Shangyang Optics used the image captured by the system as the data source, imported AI model training, and integrated the camera system and image recognition into a production line workstation, which not only improved the defect recognition rate Reaching more than 90%, it is more conducive to the subsequent development of automated production lines.

The AI ​​model training for this cooperation project is provided by Yirui Technology. Currently, most manufacturers have introduced AOI systems for production line defect inspection. Most of them use OCR (optical character recognition), which refers to the analysis and recognition processing of image files of text data. , the process of obtaining text and layout information) technology needs to be 100% accurate, and there is no room for error, resulting in accidental killings often occurring.

After adding the AI ​​training model, optical lens defects The recognition rate is greatly improved
.

▲After adding the AI ​​training model, the optical lens defect recognition rate is greatly improved.

AI+AOI solves the two major pain points of insufficient manpower and high misjudgment rate

This time, Yirui Technology and Xiaoma Optics cooperated to install Yirui's AI system in the optical inspection instruments developed by Xiaoma Optics, adding AI algorithms to the optical detection of defects, and training based on the data and needs provided by customers. AI model identification can greatly improve the accuracy of identification of defects, improve yield rate, and increase production line efficiency. Through the tripartite cooperation between Shangyang Optics, Xiaoma Optics and Yirui Technology, the optical industry AOI is introduced into AI, hoping to completely solve the pain points of industrial lens defect detection.

Since setting up the production line in 2019, Shangyang Optics hopes to introduce a smart production model. In view of the continuous growth of the company's operations and the continuous improvement of production volume, through the introduction and expansion of this achievement, the demand for manpower will be significantly reduced, and the impact of production scheduling can be reduced due to the high accuracy of the discrimination rate index, thereby improving production efficiency.

Shangyang Optics stated that as the development results are implemented, it will lead the technology to be promoted to upstream and downstream players in the optical industry, such as upstream optical lens raw material suppliers to downstream finished product applications, including immersive gaming equipment and related curved glass products , people's livelihood vehicle and security camera devices, etc.

「Translated content is generated by ChatGPT and is for reference only. Translation date:2024-05-19」

Recommend Cases

【導入案例】赫銳特科技VCSEL封裝元件瑕疵導入AOI檢測 提升產能效率20
HRT Technology Improves Production Efficiency by 20% Through AOI Detection of Defects in VCSEL Packaging

In 2017, the launch of the iPhone X made 3D sensor technology used in Face ID highly popular, which drove the development of VCSEL, a core component in the 3D sensor module In the detection of defects in incoming packaged VCSEL, the use of AI inference models can solve the industry's issue with low yield and improve reliability to 95 VCSEL technology currently can be used in many applications and various end consumer markets, including robots, mobile devices, surveillance, drones, and ARVR VCSELs are a good solution in applications that require high-speed modulation capabilities, such as cameras and biometrics VCSEL technology has a wide range ofnbsp applications, including in drones Pictured Zoyi Technology's Agricultural Drone VCSEL technology has a wide range of applications, AI technology assists in defect detection HRT Technology stated that the packaged VCSEL market is also facing strong price competition from competitors, and needs to further reduce costs and enhance product competitiveness One of the key problems is the replacement of glass lens with epoxy resin lens The production of traditional glass lenses has high yield, but the cost is higher than that of epoxy resin lenses Due to the cutting process of epoxy resin, the side wall of cutting lines can easily have rough edges, causing it to be oversized The release of stress caused by heat during the mounting process will directly cause the optical lens to break HRT Technology pointed out that the incoming inspection of VCSEL epoxy resin lenses is very important Under the constraints of packaging space, the space for fitting the package and optical lens is limited Moreover, the optical lenses will be confined to a metal frame If the dimensional tolerances are properly controlled, stress release due to heat during mounting can easily cause the optical lens to break, resulting in a yield loss of up to 10 in the VCSEL package reliability verification, resulting in an increase in production costs In order to solve the problems above, HRT Technology hopes to use AI to monitor the size and appearance defects of epoxy resin components in the VCSEL epoxy resin lens incoming stage, verifying whether their dimensions meet specifications, whether the cutting edges are smooth, and whether there are any defects in their appearance Since traditional incoming material inspection requires a rough visual inspection by humans to distinguish the quality The problem of image collection needs to be solved first to successfully collect image data Therefore, HRT Technology first developed an Automated Optical Inspection AOI device, which includes X, Y, Z three-axis motion, high-resolution cameras, and related control software to automatically record images After collecting the image data, opencv aligns the test image and a normal image to determine differences between the two images, and then pixel mapping is used to compare the pixel area to complete initial screening Manual labeling is carried out according to the image classification above, including samples that are normal, have defects in appearance, or have different shape characteristics, and then algorithm training and verification is carried out Residual neural network ResNet or other related algorithms are used for deep learning to identify the quality of lenses Implementation of AOI inspection improves production efficiency by 20 and above Comparing the differences before and after the implementation of AI image inspection, the incoming VCSEL lens inspection before implementation only involved manual inspection of the appearance The lens is packaged on the VCSEL package that has completed die bonding After passing the general light up test, the final reliability test high temperature reflow is performed Failed samples go into the rework process However, after the implementation of AOI inspection, it can screen defective lenses sooner and reduce the cost of subsequent materials input, it can also reduce the need for rework due to failure, improving yield to 95 and above in the reliability verification This is expected to help companies reduce production costs by 10 and increase production efficiency by 20 and above The difference before and after implementing AI image detection HRT Technology pointed out that this technology is an AI application developed based on tiny images It uses deep learning algorithms to identify defects in the images The trained network automatically classifies image data to predetermined categories Defect categories can be determined through reference images, so cumbersome programming is not required In the industrial machine vision environment, deep learning is mainly used for classification tasks in applications, such as inspection of industrial products or identification of parts In the future, with the development of IoT wearable devices and the trend of energy saving, the size of optoelectronic components will continue to shrink This technology can be applied to the detection of defects in the appearance of other tiny optoelectronic components in the future

這是一張圖片。 This is a picture.
Testing Seat Contact Components AI Intelligent Flaw Detection

With rapid development in 5G, AIOT, automotive electronics, and other downstream sectors, the entire supply chain is expected to benefit from this consumer market As product demand momentum gradually increases, increasing production efficiency and reducing operational costs become the most important issues In order to meet the needs of customers for various packaging types, Yingwei Technology has been committed to developing highly customized test seats However, a resulting pain point is the inability to mass-produce and fully automate operations with machines some tasks still rely on manual execution In this project, the probe part of the test seat was outsourced in 2021, and under current and future large-scale demands, work hours, costs, supply, and quality are issues Yingwei faces The company achieves a defect detection rate of 9995, which seems high, but with an average inspector able to inspect 10,000 needles per day, there would still be 5 defective needles On a test seat that is only 3 cm wide with approximately 1,000 needles, just one defective needle could potentially lead to faulty testing at the customer end As the current operational mode relies on manual visual inspection, external factors such as fatigue or oversight of personnel, and subjective judgment by inspectors may lead to the outflow of defective products, which necessitates strict quality control of contact components We once sought to utilize optical inspections Rule-based for controlling the quality of appearances, but the metallic material of the contact components leads to light scattering, background noise interference, background scratches, and material issues that could result in misjudgments Therefore, we decided to look for AI technology service providers to solve our detection difficulties Developments of Dedicated AOI Line Scan Equipment To meet the needs for inspecting thousands to tens of thousands of probes within our company's IC test seats, traditional surface imaging and individual needle imaging would be too slow to achieve rapid inspection and labor-saving goals In response, the service provider proposed a trial with an AOI dedicated line scan module solution Utilizing a width of 63mm on the X-axis for reciprocal scanning of all probes on the test seat, the tests allowed for the simultaneous scanning of 8-9 probes, significantly enhancing the future detection efficiency of AOI machines This project will proceed with the aforementioned innovative Proof of Concept POC, focusing on the development of the line scanning equipment and performing imaging, learning, and training on both normal and abnormal probes provided by our company, with initial AI model training aimed at preliminary approval This project's customized line-scan imaging module Ideal future imaging result illustration A Single AI Technology Solution for MeasurementDetection Needs Unified use of AI DL CNN learning methods, instead of the current Rule-based system which necessitates defining each defect individually, to meet the needs for abrasion measurement and appearance defect detection of malfunctionsforeign objects When the same machine uses both measurement and detection technologies, not only does it increase costs, but it also affects the detection speed Hence, the service provider recommends the use of a line scan device for imaging Its resolution is sufficient for AI to simultaneously determine appearance defects and assess the condition of needle tip abrasion, as detailed below Line scan pixel imaging displaying needle tip abrasion conditions This AI detection technology meets both measurement and inspection needs for Yingwei, not only bringing more benefits to future probe testing but also introducing an innovative axis in AI technology Change the method of human inspection, enhance work efficiency and product quality After combining both hardware line scan and software AI model training approaches, we successfully ventured into new AOI detection applications Following the AI implementation POC, including the development and validation of a customized line scan module and an initial AI model, the plan is to officially develop the AOI machine next year and integrate it into the IC test seat production line Future Prospects Probe manufacturers upstream and downstream IC factory users both have needs for the AOI inspection machine upstream can ensure probe quality before leaving the factory, while downstream users can use this machine to regularly inspect the condition of numerous IC test seats in hand Given the future demands, the AOI machine is poised to have a significant positive impact on the IC testing industry in the foreseeable future 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-12-12」

這是一張圖片。 This is a picture.
CCTV Intelligent Video Search System

Search for a specific person, find someone with a suitcase entering the factory in Gao'an area Color features of the person and the object confirmed, person in blue and black top, suitcase in black color, throughCCTV the intelligent video search system, by setting object and color retrieval conditions, it can successfully locate three video clips containing the target subject This greatly aids operational staff in finding the target items, and through this system, search speed can far surpass manual effort6fold Pain Points The CSE-Kaohsiung Plant is densely equippedCCTVto monitor every corner of the plant area, but when an incidenthappens, it's impossible within a limited time throughCCTVvideo playback to find the incident, the implications and risks behind this are self-evident Many areas that are usually unmanned can easily become security blind spots Thus, how to monitor a vast plant area more intelligently and effectively is one of the crucial aspects of building a smart plant for the semiconductor industry The AES Plant in Kaohsiung covers a vast area, with many important sites requiring monitoring of personnel movements to ensure corporate secrets and employee safety 1 Automated production lines and warehouses In semiconductor enterprises’ automated production lines and warehouses, oftenAGV(Automated Guided VehicleAGVs automated guided vehicles travel at high speeds if plant personnel inadvertently enterAGVthe moving area and cannot issue a warning to the person, then the regrettable accidents that occur will be too late to reverse 2 Material and product storage areas Materials used in semiconductor-related processes are costly if areas storing materials or products are breached, there is a risk of loss of high-value materialsproducts 3 High-security areas Trade secrets relate to the core technological competitiveness of semiconductor-related enterprises if someone breaches the high-security areas, there is a risk of corporate secrets being leaked The safety of trade secrets has always been one of the most critical issues for semiconductor enterprises 4 Loading docks At AESLButthe dock area often has loading vehicles coming and going if someone intrudes into the dock area, there is a risk of vehicle collisions and accidents Additionally, goods awaiting shipment at the dock area could be stolen or potentially damaged from collisions, thus causing significant reputation and financial losses for the company, further leading to production and shipping inconvenience When an abnormal event occurs, how to quickly search for the relevant key footage from massive data Many important locations within the AES Kaohsiung Plant need to be equippedCCTVfor safety checks, butCCTVWith thousands to tens of thousands of cameras, manually searching through footage for an event requires laborious frame-by-frame review which is time-consuming and inefficient In light of advancements in computer vision, it's beneficial to utilizeAIto replace manual playback and searching Problem Scenario Object Detection The data source for object detection comprises two parts Open-source datasetsOIDv4and AES Kaohsiung PlantCCTVImage files For these files, search for usable data, specificallyOIDv4image files For these files, extract the defined nine major categories of objects for training data among them, two object categories, knives and gasoline barrels, were not found inOIDv4found usable data for knives and gasoline barrels, while the remaining seven categories of objects are available fromOIDv4useful training data found for the remaining seven categories of objects, all marked Regarding the Kaohsiung PlantCCTVimage files, select some frames Frame of the footage, and manually annotate the objects to be_detected for training and testing data Nine Major Objects Color Recognition The data source for color recognition is divided into two partsInternet image screenshots, and Kaohsiung PlantCCTVimage files Currently, no publicly available open-source datasets specifically for color recognition applications have been found, so images are collected from the web Search the web for images of the defined nine major object categories, save the images after separating the objects from the background, keeping only the object sections, and mark the images according to color Additionally, for the Kaohsiung PlantCCTVimage files, use the already-markedbounding boxextractCCTVimage files from variousFramesections of objects identified by color, and finally, visually identifiable images are marked according to color Each object category has its specific color definition, depending on the usual colors seen in these objects in real life Dynamic Ignore during Training FromOIDv4during the training of the object detection pilot model, since each image in this dataset is only marked for a single category, but the image may contain other desired detection categories unmarked For such cases, dynamic ignore techniques will be employed during training to avoid confusion Next, use the extracted training data from the Kaohsiung Plant toFine-Tuneenhance the detection rate of the object in specific designated areas Finally, select the model that computes the lowest loss value in the test set during the training process as the main object_detection model Dynamic Ignoring AIHelp You View CCTV The intelligent video search system primarily serves as an assistive system for searching surveillance footage, capable of speeding up the process of finding target events by setting search conditions for objects By simply defining the search conditions, you can quickly produce thumbnails of critical objects and playback for review, shortening the time required for manual case retrieval of the past The search time is quickly6doubled, allowing the front-end security unit to use this platform to strengthen the first line of risk management supervision and take timely preventive measures 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-12-12」