A deep residual network-based method for automatically and rapidly detecting micro-defects on the surface of an injection molded part

Abstract

The invention discloses a deep residual network-based method for automatically and rapidly detecting micro-defects on the surface of an injection molded part. The method comprises the following steps: S1, collecting micro-defects and defect-free simulation and real graphs on the surface of an injection molded part by utilizing injection molding simulation software and an injection molding industrial product; s2, carrying out micro-defect and defect-free simulation on the surface of the injection molding part, and carrying out normalization, noise reduction and cutting treatment on a real graph; s3, constructing an injection molding part surface microdefect mixed data set; s4, designing a deep residual network injection molding part microdefect identification framework based on deep learning; s5, training a deep residual network microdefect identification framework based on deep learning; s6, acquiring a deep residual network microdefect identification framework model based on deep learning; and S7, detecting the types of the micro-defects. According to the invention, the types of the micro-defects can be quickly identified only by inputting a micro-defect pattern on the surface of the injection molded part. The invention is high in microdefect identification speed and high in detection rate, and has wide practical value and application prospect.

Description

technical field [0001] The invention belongs to the technical field of detection and identification of micro-defects, and in particular relates to a method for automatic and rapid detection of micro-defects on the surface of injection molded parts based on a deep residual network. Background technique [0002] With the improvement of people's living standards, the requirements for the appearance quality and performance of injection molded parts are getting higher and higher. In the injection molding process, due to various factors, injection molded parts may have surface micro-defects such as short shots, flashes, weld lines, bubbles, cracks, etc. These micro-defects cannot be detected and identified by manual methods. Therefore, the realization of micro-defect detection is an important link to ensure the quality of injection molded parts. [0003] At present, surface defects are the main detection objects in the field of injection molded parts inspection. The detection an...

AI Technical Summary

Problems solved by technology

Although the above-mentioned non-manual detection methods have the advantages of non-contact and high degree of automation, they rely heavily on professionals to extract and analyze the defect areas in the collected images, especially when the images contain noise, and this feature analysis becomes more difficult.

[0004] In summary, some progress has been made in the detection and identification of surface defects of injection molded parts, but the detection and identification of micro defects on the surface of injection molded parts is still rare, and in-depth research is urgently needed to realize the automatic and rapid detection of micro defects on the surface of injection molded parts. detection

Images