摘要: 针对磨粒识别中的图像分割问题,应用一种新的二值化算法将磨粒目标从复杂背景中分离。该算法以原始图像与分割后图像的相关系数最大为准则,选取最优阈值对磨粒图像进行二值化。应用实例表明了该方法的简洁有效性。
关键词:
图像分割,
相关系数,
阈值选取,
磨粒
Abstract: For the recognition of wear debris, a new thresholding method is applied to segment wear debris from its complex background. An optimum threshold is selected by maximizing the correlation between the original halftone image and the thresholded bi-level image. Experimental results show that the method being presented is brief and effective.
中图分类号:
王伟华,马艳艳,王成焘,叶俊. 一种基于相关准则的磨粒图像二值化算法[J]. , 2003, 29(18): 20-21.
WANG Weihua1, MA Yanyan1, WANG Chengtao1, YE Jun2. Wear Debris Image Thresholding Based on Correlation Criterion[J]. , 2003, 29(18): 20-21.