复杂光照条件下的图像边缘提取算法的改进  

Improved Edge Extraction Algorithm for Images under Complex Lighting Conditions

作　　者：郭健平[1] 张俊轩 GUO Jian-ping;ZHANG Jun-xuan(School of Information and Optoelectronic Science and Engineering,South China Normal University,Guangzhou 510006,China)

机构地区：[1]华南师范大学信息光电子科技学院

出　　处：《光电技术应用》2020年第1期43-46,55,共5页Electro-Optic Technology Application

基　　金：国家自然科学基金项目（61774062）;华南师范大学2017年教学研究及改革项目

摘　　要：边缘是数字图像中的一个重要的局部特征。在复杂光照条件的情况下,图像分辨率较低,传统Sobel算子利用卷积计算出来的灰度梯度较为平缓,边缘不能有效的识别。通过改进Sobel算子的模板,平衡图像的亮边缘与暗边缘,尽可能地避免暗边缘信息的丢失。与此同时,通过分析数据建立相应的数学模型,优化子区域大小,把图像划分为多个子区域,在子区域内利用大津法进行边缘的检测与提取,最后使用Hilditch细化算法将图像的边缘提取出来。通过MATLAB仿真结果表明,与传统So bel算子的边缘提取算法相比,文中算法简单易实现,对于复杂光照条件下的图像,边缘信息提取更完整,自适应能力更强。The edge is an important feature for digital image.In the case of complex lighting conditions,the im⁃age resolution is low while the gray gradient calculated by the traditional Sobel operator is relatively smooth,and the edges cannot be effectively identified.An improved Sobel operator template is proposed in which the bright and dark edges are balanced to avoid missing the information of the dark edges.At the same time,the data are analyzed to establish a corresponding mathematical model and optimize the size of the sub-region.By dividing the image into multiple sub regions,the Otsu method and Hilditch thinning algorithm are used to detect and extract edges of the im⁃ages.The MATLAB simulation results show that compared with the traditional Sobel operator edge extraction algo⁃rithm,the algorithm is simple and easy to implement.For images under complex lighting conditions,the edge infor⁃mation extraction is more complete and the adaptive ability is stronger.

关 键 词：改进Sobel算子 边缘提取 自适应阈值 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]