Automatic detection system and operation method thereof

Abstract

The invention discloses an automatic detection system and an operation method of the automatic detection system. The automatic detection system comprises a machine vision detection device, a repair station connected with the machine vision detection device, an artificial intelligence module and a processing platform connected with the repair station and the artificial intelligence module. The machine vision detection device can obtain an image of a to-be-detected object and display the image through an operation picture of the restoration station, the artificial intelligence module captures the image of the to-be-detected object for classified inference, and the processing platform is used for automatically superposing result data of image classified inference of the to-be-detected object onto the operation picture of the restoration station. And intercepting manual re-judgment result data input by an operator to the repair station through the user interface, and then sending the result data to the artificial intelligence module to serve as data for subsequent automatic classification model re-training learning.

Description

technical field [0001] The present invention provides a system and method, and more particularly, relates to an automatic detection system and method of operation thereof. Background technique [0002] With the advancement of science and technology, the size of electronic parts is reduced and the precision is increasingly required. Therefore, in order to ensure the improvement of the qualified rate of products, in the modern process, it is necessary to carry out inspections at various important sites to facilitate early detection of defects or defects. Defective and substandard products, reduce the waste of process resources, improve the output and pass rate of the production line, but also increase the time consumed by the process with the multi-station inspection. Therefore, in order to speed up the inspection time, automatic visual inspection and Automated inspection technologies such as automatic optical inspection have become an indispensable role in the process. [00...

AI Technical Summary

Problems solved by technology

[0003] The general automatic inspection equipment is widely used in the appearance inspection of the circuit board assembly production line, and replaces the previous manual visual inspection. For example, it can be used to detect the soldering of parts on the circuit board and the quality status after assembly. Technology is used to compare whether the object under test is too different from the standard image to judge whether the object under test meets the standard. However, traditional automatic detection equipment often has a high false alarm rate due to factors such as color difference and height drop, which increases the Afterwards, manpower re-inspects abnormal detection and wastes processing costs

[0004] However, with the rise of modern Industry 4.0 and intelligent manufacturing, and in intelligent factories that focus on production efficiency and quality, quality inspection is one of the most important procedures to ensure the pass rate of production lines and product quality. In order to improve the accuracy of inspection operations In the past, automated testing equipment was often used to assist on-site production line personnel. However, as the application of artificial intelligence technology has gradually become mainstream, the gradually mature artificial intelligence technology has also begun to integrate with automated testing equipment. Through machine learning or deep learning algorithms , to make the defect detection of automated testing equipment more accurate, to reduce the false alarm rate of automated testing equipment and save a lot of labor costs, but the machine learning or deep learning used by the manufacturer needs to go through the training mode first, and then download the training results The biggest problem in establishing the training mode is that the proportion of good samples and defective samples is very different. The lack of defective samples will also make it difficult to effectively train and identify defects. model of

Images