模具技术Issue(6)：158-165,8.
扫描图像灰度分割下成型模具工作模面缺陷无损检测
Non destructive testing of defects on the working surface of forming molds under grayscale segmentation of scanned images
摘要
Abstract
In the field of non-destructive testing in industrial manufacturing,due to the presence of dust,particles,and other impurities in the workshop air,these impurities can adhere to the scanning head or mold surface during the scanning process,interfering with the normal movement of the scanning line and the reception of reflected light.This interference can cause degradation factors in the scanned image,affecting the detection effect.Therefore,a non-destructive detection method for defects on the working surface of the forming mold under scanning image grayscale segmentation is proposed.After obtaining the image of the working surface of the forming mold through a flatbed scanner,the degradation factors in the image are removed using filters and grayscale linear transformation methods to enhance the image quality.Based on the energy generalization function of grayscale and variance,the pixel grayscale value difference of the enhanced image is calculated,and the image is segmented into the mask part and the background part.Calculate the similarity between the grayscale segmented template image and the defect free template to achieve non-destructive testing.Performance tests have shown that the proposed method avoids damage to the mold caused by destructive testing,ensuring the integrity of the mold while efficiently and accurately detecting potential defects,providing a reliable basis for timely repair and maintenance of the mold.关键词
扫描图像/无损检测/灰度分割/模具工作模面/图像增强/能量泛化函数Key words
scanning images/non destructive testing/grayscale segmentation/mold working surface/image enhancement/energy generalization function分类
信息技术与安全科学引用本文复制引用
王佳祥,黄思齐,冯成,吕鹏辉,杨书,苑建坤..扫描图像灰度分割下成型模具工作模面缺陷无损检测[J].模具技术,2025,(6):158-165,8.基金项目
贵州省科技重大专项项目"智慧果园关键技术研究及装备研发"(编号:黔科合重大专项字[2024]002). （编号:黔科合重大专项字[2024]002）
