阈值分解下的冷轧极薄带钢表面缺陷分割

摘要

为分割具有低对比度和噪声复杂的冷轧极薄带钢缺陷,在二维Otsu分割算法的基础上进行了改进.为了避免在一个很大的二维空间上搜索阈值,分解二维Otsu分割算法的二维直方图,分别在图像像素灰度直方图和像素邻域灰度直方图上进行阈值求解,并将获得的阈值作为最佳分割阈值.为改善分解过程中忽略的边缘和噪声影响,将平衡因子加入到像素邻域灰度分割阈值求解中.最后通过大量实验对二维Otsu分割算法、量子粒子群双阈值分割算法和改进分割算法的分割结果进行对比分析.实验表明,该算法能够快速高效地分割出冷轧极薄带钢表面的各种缺陷.%Two-dimensional Otsu algorithm was improved in order to segment the defects of low contrast and complex noise in cold rolling of thin strip.In order to avoid searching threshold in a large two-dimensional space,two-dimensional Otsu algorithm of two-dimensional histogram has decomposed.The algorithm solves threshold in the image pixel gray histogram and gray histogram neighborhood pixels respectively.The balance factor to the pixel neighborhood grayscale was added in solving the segmentation threshold and using the threshold as the best image segmentation threshold.Finally,the segmentation result was analyzed through a large number of experiments,comparing with two-dimensional Otsu algorithm,Quantum-behaved particle swarm optimization (QPSO) and improved algorithm.The results show that the method can quickly and efficiently split out various defects on the surface of cold roiled thin strip in computer vision.