A method and a device for detect defects of a die casting part

Abstract

A method and apparatus for detect defects of die casting parts are disclosed, By collecting pictures of die castings with defect features, the defect image is denoised, enhanced processing, The depthconvolution network Alexnet is trained with the defect pictures to classify the defect pictures, establish the corresponding semantic tags and store them in the defect feature database, track the defects of the die castings in real time. When the defects of the die castings are detected, the defects are fed back to the inspection device and the inspection data are counted to generate the defect inspection report of the die castings. The invention improves the defect detection efficiency and precision of the die casting, and has the advantages of high accuracy, high speed and good real-time performance, and can quickly and accurately find the defective die casting.

Description

technical field [0001] The invention relates to the technical field of quality inspection, in particular to a method and device for detecting defects of die castings, a readable storage medium and a computer control system. Background technique [0002] Die casting quality inspection is an important part of the post-processing stage of die casting, and its inspection efficiency and quality are directly related to the quality of the final product. The traditional quality inspection method of die castings is to detect the surface defects of die castings, which is generally manual inspection. Manual inspection is a static local inspection method, which can only detect and measure specific points and positions. In order to improve the measurement accuracy and accuracy Degree, often requires multiple workers to measure the average value. During the inspection process, the workers are completely relied on to carry the castings, and the products are classified according to the res...

AI Technical Summary

Problems solved by technology

Moreover, the efficiency of manual detection is low, the detection environment is harsh, and it is easily affected by the subjective factors of workers.

With the development of industrial technology, automatic defect detection equipment based on machine vision has emerged as the times require. However, the current defect detection method based on vision is still in the initial stage of development, the detection technology is not yet mature, and the defect detection method system is not complete.

Especially in the die-casting defect detection industry, due to the various defect characteristics on the surface of die-casting parts, different defect characteristics such as oil stains, scratches, flashes, and burrs may appear. During the detection process, different defect characteristics may interfere with each other, resulting in errors. For example, if the defect feature to be detected is a relatively obvious burr, the die casting with oily characteristics may also be included in the test result

Images