Industrial product defect detection method and industrial intelligent camera

Abstract

The invention relates to an industrial product defect detection method and an industrial intelligent camera. The industrial product defect detection method comprises the steps: S1, starting an industrial intelligent camera, and an onboard camera chip adjusting controllable light source illumination parameters according to imaging requirements; S2, photographing an industrial product by using the low-distortion lens, and transmitting a photographed product image to the embedded processor through the onboard camera chip; and S3, the embedded processor performing defect identification on the product image by using the deep learning defect detection model, and outputting an inference result to the outside. The industrial intelligent camera carries the deep learning defect detection model, defects with complex structures and complex production environments can be efficiently recognized, meanwhile, the deep learning defect detection model capable of adapting to different workpieces and environments can be loaded and updated through the server, and the industrial intelligent camera is used for collecting images of industrial products. Therefore, instant detection of product defects can berealized.

Description

technical field [0001] The invention relates to industrial product detection, in particular to an industrial product defect detection method and an industrial smart camera. Background technique [0002] At present, the machine vision system in the field of industrial defect detection is mainly divided into two parts: the image acquisition unit composed of traditional cameras, lenses, light sources, camera fixation and motion mechanisms, and the machine vision detection system composed of image processing units such as PC hosts and image acquisition cards. This kind of machine vision has high extensibility and plasticity. Custom hardware and custom software development can be done for different product defects. However, this traditional detection system has problems such as large hardware structure, complicated deployment and loading and unloading, and more professional and technical personnel required for system deployment. [0003] The industrial smart camera system appli...

AI Technical Summary

Problems solved by technology

[0005] (1) Image processing software uses solidified defect detection methods such as: dimension measurement, edge extraction, simple location search, etc. These methods cannot accurately and effectively locate defects with complex defect structures and complex production environments;

[0006] (2) It is impossible to carry out special treatment for different products and defects;

[0007] (3) It is impossible to memorize and learn the detected workpieces. Once the detection model is fixed, the accuracy is relatively fixed

[0010] However, the above problems are still not effectively solved, so there are deficiencies in the existing industrial smart camera detection field, which needs to be improved and improved.

Images