基于光照影响因子的动态Niblack算法研究及应用  

Research and application of dynamic Niblack algorithm based on illumination impact factor

作　　者：魏兴凯 蒋峥[2] 傅呈勋 刘斌[2] WEI Xing-kai;JIANG Zheng;FU Cheng-xun;LIU Bin(Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education,Wuhan University of Science and Technology,Wuhan 430080,China;College of Information Science and Engineering,Wuhan University of Science and Technology,Wuhan 430080,China;Intelligent Manufacturing Department,Hunan Xianggang Engineering Technology Limited Company,Xiangtan 411100,China)

机构地区：[1]武汉科技大学冶金自动化与检测技术教育部工程研究中心,湖北武汉430080 [2]武汉科技大学信息科学与工程学院,湖北武汉430080 [3]湖南湘钢工程技术有限公司智能制造部,湖南湘潭411100

出　　处：《计算机工程与设计》2022年第4期1066-1073,共8页Computer Engineering and Design

基　　金：国家自然科学基金项目(51805386)。

摘　　要：为解决钢板喷印字符图像存在的字符拖尾和光照不均匀导致的字符图像二值化效果差的问题,提出一种基于光照影响因子的动态Niblack算法。在Niblack算法的基础上引入光照强度项,降低光照不均匀带来的影响;定义灰度阈值中各项的动态系数,减少目标点和背景点的误判,获得良好的二值化效果;结合Otsu算法得到全局阈值,以全局阈值对算法的灰度阈值进行修正,加强二值化图的局部联系,提升二值化字符的连贯性。通过与Niblack、Sauvola等多种算法的对比,验证了该方法的可行性。To solve poor binarization effects of character image printed on steel plate due to character trailing and uneven illumination,a dynamic Niblack algorithm based on illumination influence factor was presented.Based on Niblack algorithm,the illumination intensity term was introduced to reduce the influence of illumination unevenness.The dynamic coefficients of each item in the gray threshold were defined,the misjudgment of the target point and the background point was reduced,and good binarization effects were obtained.The global threshold was obtained by combining Otsu algorithm,the gray threshold of the algorithm was modified with the global threshold,and the local connection of binarization image was strengthened and the coherence of binarization characters was improved.By comparing with other algorithms,the feasibility of the proposed method is verified.

关 键 词：光照影响因子 光照强度 动态系数 全局阈值 二值化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]