摘要
工业产品的质量检验是智能制造领域发展的关键,但现有的工业产品质量检验方法普遍存在人力资源成本高、方法复杂、难以规范化监控、检验效率低等缺点.针对这些问题提出一种基于机器视觉的表面裂纹自动检测与质量检测技术,提出了一种视场环境标定方法,给出了基于机器视觉的工件形状特征识别和尺寸测量的具体方法,实现了工件特征缺陷和尺寸偏差等质量问题的监测.结合多尺度关注模块和基于高低混合特征图恢复图像像素位置的上采样模块构建工件裂纹提取网络,实现工件裂纹特征提取、裂纹类型分类和损伤程度划分,有效提高了检测技术精度.
Abstract
The quality inspection of industrial products is the key to the development of intelligent manufacturing.However,there are problems in the existing methods of industrial product quality inspection:the high cost of human resources,complicated methods,difficult standardized monitoring and low inspection efficiency,etc.To solve these problems,the study proposes an automatic surface crack detection and quality detection technology based on machine vision,a field environment calibration method,and a specific method of workpiece shape feature recognition and size measurement based on machine vision,so as to realize the monitoring of workpiece quality problems,such as feature defects and size deviations.Through combining with the multi-scale attention module and the up-sampling module based on the high-low mixed feature map to recover image pixel position,the study constructs the workpiece crack extraction network to realize the workpiece crack feature extraction,crack type classification and damage degree division,which effectively improves the detection accuracy.
维普
万方数据