基于图像边缘增强与弱化的边缘检测  

Edge detection based on image edge enhancement and weakening

作　　者：张晗 钱育蓉[1] 王跃飞[1] 陈人和 田宸玮 ZHANG Han;QIAN Yu-rong;WANG Yue-fei;CHEN Ren-he;TIAN Chen-wei(School of Software,Xinjiang University,Urumqi 830008,China)

机构地区：[1]新疆大学软件学院

出　　处：《计算机工程与设计》2019年第11期3106-3110,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(61562086、61462079);新疆维吾尔自治区创新团队基金项目(XJEDU2017T002)

摘　　要：为提高边缘算子检测结果的清晰度,在扩充清晰边缘面积上,提出深度探测法以增强邻域边缘;在消除模糊纹理上,用跨越步长Δx和Δy作为模糊纹理和清晰边缘的分界,达到弱化模糊纹理的目的,凸显模糊区域局部对比度;在算法执行效率上,空间复杂度不高,但时间复杂度为O(4n3)。经测试,该算法可实现扩充清晰边界、弱化模糊纹理以及凸显模糊纹理区域对比度的目标,在小于1600*1600分辨率图像上处理速度较佳,但在更高分辨率上计算较久。该方法凸显了边缘分界,可用在其它领域算法预处理阶段。To improve the sharpness of the edge operator detection results,a depth detection method was proposed to enhance the neighborhood edge for extending clear edge area.For eliminating the fuzzy texture,the step sizeΔxandΔy were used as the blurred texture and the sharp edge.The boundary was to achieve the purpose of weakening the fuzzy texture and highlighting the local contrast of the fuzzy region.For the algorithm execution efficiency,the space complexity was not high,but the time complexity was O(4 n3).After testing,the proposed algorithm can expand clear boundaries,weaken the fuzzy texture and highlight the contrast of the blurred texture region.The processing speed is higher on images with less than 1600*1600 resolutions,but lower on the ones with higher resolution.This method highlights the edge demarcation and can be used in algorithm preprocessing stage in other areas.

关 键 词：边缘检测 PREWITT算子 边缘强化 边缘弱化 深度探测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]