Mechanical part defect detection method and system

Abstract

The invention discloses a mechanical part defect detection method. Firstly, evaluation of a detection result of an ultrasonic flaw detector is realized based on the ultrasonic flaw detector and a BP neural network model; collection of a top view, a top view, a front view, a rear view, a left view and a right view of a mechanical part is realized through an image collection module, and image denoising and image enhancement preprocessing are respectively performed by adopting bilateral filtering and partwise linear transformation algorithms; an iterative adaptive threshold segmentation method isadopted to carry out image binarization small-area clutter noise is removed based on morphological region features of the connected components; and realizing detection of defects in the image basedon a DSOD algorithm, outputting a defect recognition result based on an Inception V3 depth model, and finally realizing classification of the mechanical parts based on an LSSVM according to the mechanical part defect information and the defect recognition result. Quick and high-accuracy mechanical part defect detection can be realized by utilizing an ultrasonic flaw detection technology and an image recognition technology.

Description

technical field [0001] The invention relates to the field of detection of mechanical parts, in particular to a method and system for detecting defects of mechanical parts. Background technique [0002] With the advancement of production technology, the output of mechanical parts continues to increase, and most manufacturers still use manual inspection of the quality of mechanical parts. Manual inspection is easily affected by the physical condition of the inspector, so long-term visual inspection is prone to problems such as missed detection rate and false detection rate increase. How to realize automatic detection of mechanical parts has become a technical problem to be solved urgently. SUMMARY OF THE INVENTION [0003] In order to solve the above problems, the present invention provides a mechanical part defect detection method and system, which can realize fast and accurate mechanical part defect detection by using ultrasonic flaw detection technology and image recogni...

AI Technical Summary

Problems solved by technology

Manual inspection is easily affected by the physical condition of the inspector, so long-term visual inspection is prone to problems such as missed detection rate and false detection rate increase

Images