论文部分内容阅读
在汽车制造行业中,曲面缺陷检测是品质控制的重要环节.本文模拟了在实际缺陷检测过程中工程师通过平行光的影像图案形态发现车身表面缺陷的过程,提出了Integral Reflection Map(简称IRM)检测算法.该方法基于光线的反射定律和光线的可逆性,通过比较标准模型和制造模型在反射光线时表现出来的差异来检测曲面缺陷.同时,通过对有效视点区域范围的积分,该方法避免了在不同视点位置下检测结果的不稳定性.并且针对于标准模型和制造模型的整体刚体对齐无法检测某些尺度下的曲面缺陷的问题,本文提出了标准模型和制造模型的局部对齐方法.实验结果表明,我们的方法自动化程度高,与视点无关,能检测出不同尺度的曲面缺陷.
In the automobile manufacturing industry, surface defect detection is an important part of quality control.This paper simulates the process of the engineer detecting the surface defects of the vehicle body through the image pattern of parallel light during the actual defect detection, and proposes the Integral Reflection Map (IRM) This method detects the surface defects based on the law of reflection of light and the reversibility of light, and compares the difference between the standard model and the model when the light is reflected. At the same time, this method avoids the problem by integrating the range of effective viewpoints And the instability of the detection results at different viewpoints.Aiming at the problem that the whole body alignment of the standard model and the manufacturing model can not detect the surface defects under some scales, this paper presents the local alignment method of the standard model and the manufacturing model. The results show that our method has a high degree of automation and has nothing to do with the viewpoint, and can detect the surface defects of different scales.