基于机器视觉的管纱污渍检测算法研究

针对管纱表面污渍的质量问题,提出基于机器视觉的管纱表面污渍检测算法。先对图像进行预处理以及Sobel算法处理,消除光照不均以及背景纹理的影响,再用双高斯算法增强目标与背景的差异,然后用灰度矩形法削弱背景的噪声,最后采用动态阈值的方法提取目标。实验结果表明,该算法对管纱表面污渍的检测有一定的准确性与实时性。

The Research on Bobbin Yarn Stain Detection Based on Machine Vision

In view of the quality problem of surface stains of fiberglass bobbin yarn,an algorithm based on machine vision is proposed to detect the stains on the surface of bobbin yarn. Firstly,the image preprocessing and Sobel al-gorithm are used to eliminate the influence of uneven illumination and background texture. Secondly,Gaussian al-gorithm is used to enhance the difference between the target and the background. Then the background noise is weakened by using gray rectangle method. Finally,the dynamic threshold is used to extract aims. The experimental results show that the algorithm has certain accuracy and real-time performance for the detection of bobbin yarn stains.