What is defect detection?
Defect detection is an important part of the product quality management system in various industries, and it is also the last barrier before the product is officially put into the market. Due to the various quality problems that may occur in products and there is no unified measurement standard, product quality inspection has always been done manually. It can be said that the final delivery quality of the product largely depends on the work experience of the quality inspector. However, relying entirely on manual work is faced with the problems of low efficiency and increasing costs. How to improve the efficiency of quality inspection and reduce costs on the basis of ensuring product quality is one of the long-term goals of every manufacturing company. Thanks to the continuous development and maturity of machine vision, more and more manufacturing companies are trying to introduce machine vision inspection technology into product defect detection. At present, defect detection technology based on machine vision has been widely used in the defect detection of textiles, auto parts, semiconductors, photovoltaic modules and other products, which greatly improves the quality inspection efficiency of the manufacturing industry. The prospect of machine vision in industrial defect detection is unquestionable, and factors such as the diversity of industrial manufacturing, the complexity of production environments, and the non-standard nature of product defects have brought many challenges to the practical application of machine vision in defect detection. .
Early machine vision was mainly based on 2D image processing. As the manufacturing process became more and more complex, the requirements for the accuracy and stability of detection became higher and higher, and the limitations of 2D were gradually exposed. When the device float height, tilt angle, tombstone and other indicators, only based on RGB information can not be stably detected, can not measure the climbing angle and size of solder joints. With the help of the 3D imaging system, the 3D information can be used to complete the detection more stably, and the 2D detection items can also be fully compatible. Judging from the current application trend, the application scope of 3D is becoming more and more extensive.
how to learn
Defect detection is widely used in various manufacturing industries. However, due to the complexity of knowledge points, it is difficult for universities or enterprises to form a complete set of tutorials. Many students who are just getting started can only rely on scattered materials on the Internet to learn, and it is difficult to grow and progress efficiently. To this end, 3D Vision Workshop has launched the first domestic defect detection course for industrial-level practical combat. The teacher is a senior 3D algorithm engineer of a head intelligent manufacturing company, and has rich experience in defect detection algorithm design and implementation. The course will be launched from three parts: environment configuration , basic algorithm analysis , and practical explanation . It can really be aimed at novice and algorithm engineers with certain work experience. The syllabus is as follows:
Instructors
Lin Zixiang, currently a 3D senior algorithm engineer for a head intelligent manufacturing company, holds a master's degree in automation from Northeastern University. He has worked in the field of robotics and AOI inspection for many years, and has led the research and development of more than 10 3D products. And as the technical leader to promote the implementation of research and development results. Extensive working experience in computer vision algorithms.
object oriented
Undergraduate, master and doctoral students in defect detection related fields;
Algorithm engineers engaged in defect detection;
Algorithm team members of industrial vision;
After school harvest
1. In-depth understanding of the application of 3D vision in the field of defect detection, and the use of related algorithms and tools;
2. Familiar with post-processing algorithms commonly used in 3D vision, including filtering, segmentation, model fitting and other modules;
3. From understanding requirements, data preparation, algorithm design, to algorithm performance testing, master the complete development pipeline of 3D visual inspection;
Schedule
Course purchase
Course consultation
QQ consultation group: 910070197
▲Course consultation QQ group to learn more
▲Long press to add customer service WeChat to consult more