Research on Defect Detection Technology for Powder Metallurgy Gears Based on Grayscale Operation
A powder metallurgy gear defect detection method based on grayscale operation was studied and designed.This method obtains grayscale images of powder metallurgy gears through image acquisition equipment,and extracts defect area features through grayscale processing,median filtering noise reduction,threshold segmentation,morphological analysis and other steps.Then,particle detection method is used,combined with area size and grayscale subtraction operation,to judge defects such as broken teeth,scratches,and stains.Finally,the VGG16 network model is used for automatic detection of gear defects.The results showed that the accuracy of the validation and testing sets of the research method was 98.24％and 98.35％,respectively,and the detection accuracy of gear defects was 98.96％.This indicates that the research method can achieve high-precision detection and evaluation of powder metallurgy gear defects,and improve the quality control and efficiency of the production process.
grayscale operationpowder metallurgy gearsdefect detectionimage processing technologynetwork model
NETL
NSTL
万方数据
