基于图像灰度变换的OTSU阈值优化算法  

OTSU threshold optimization algorithm based on image gray level transformation

作　　者：娄联堂[1] 何慧玲 LOU Liantang;HE Huiling(College of Mathematics and Statistics, South-Central University for Nationalities, Wuhan 430074, China)

机构地区：[1]中南民族大学数学与统计学学院,武汉430074

出　　处：《中南民族大学学报（自然科学版）》2021年第3期325-330,共6页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：国家自然科学基金资助项目(60975011);中南民族大学研究生学术创新基金资助项目(3212020sycxjj303)。

摘　　要：针对X射线图像裂纹缺陷的分割问题,研究了基于图像灰度变换的OTSU阈值优化算法.通过引入灰度变换函数,建立图像灰度变换与最大类间方差之间的关系,解决由灰度分布不均匀产生的图像分割问题.通过OTSU阈值优化问题的设计,在图像灰度变换与图像分割之间寻求一个最佳策略,找到使类间方差达到最大的阈值及对应的灰度变换函数,由此得到了一种将分段灰度变换和OTSU阈值优化相结合的适合于X射线图像裂纹缺陷分割的方法.Aiming at the segmentation problem of X-ray image crack defect,the OTSU threshold optimization algorithm based on image gray level transformation is studied.By introducing the gray level transformation function,the relationship between the image gray level transformation and the maximum between-class variance is established,and the image segmentation problem caused by the uneven gray distribution is solved.Through the design of OTSU threshold optimization problem,an optimal strategy is found between image gray level transformation and image segmentation,and the threshold value and corresponding gray level transformation function to maximize the between-class variance are found.Thus,a method suitable for X-ray image crack defect segmentation is obtained by combining segmented gray level transformation with OTSU threshold optimization.

关 键 词：阈值优化 灰度变换 类间方差 图像分割 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]