基于局部模糊增强的顺序形态学边缘检测算法  

Order morphology edge detection algorithm based on partial fuzzy enhancement

作　　者：黄海龙[1] 王宏[1] 纪俐[1] 

机构地区：[1]东北大学机械工程与自动化学院,沈阳110004

出　　处：《仪器仪表学报》2012年第11期2608-2614,共7页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61071057)资助项目

摘　　要：为了提高检测精度、节省运算时间,对现有的基于数学形态学边缘检测算法进行改进,提出了一种基于局部模糊增强的复合顺序形态学边缘检测算法。采用二维直方图斜分法定位图像的边缘区域,对边缘区域进行模糊增强处理,局部模糊增强既可以突出边缘特征,又可以减少计算量,改善算法的实时性;复合顺序形态学边缘检测使用不同方向的直线形结构元素和2种大小的百分位得到边缘子图像,通过计算信息熵自适应确定权重系数,对边缘子图像作融合处理,细化后得到最终的边缘。实验结果表明,该算法检测的边缘精细、连续、完整,其均方误差和峰值信噪比要好于传统的算法,对受到噪声污染的图像和不同格式的图像具备良好的鲁棒性,与全局增强算法相比可节省近一半的运算时间。In order to increase detection accuracy and save operation time, a new multiple order morphology edge detection algorithm based on partial fuzzy enhancement is proposed through improving the existing mathematical morphology edge detection algorithms. Two-dimensional histogram oblique segmentation method is adopted to locate the image edge region, and the edge region is processed with fuzzy enhancement. Partial fuzzy enhancement not only emphasizes edge features, but also reduces computation task, and improves the real-time capacity of the algorithm. The multiple order morphology edge detection algorithm uses different direction linear structuring elements and two kind percentiles to obtain edge sub-images, determines the weight factors adaptively through calculating the information entropy ,fuses the edge sub-images, and obtains the final edge after thinning. Experimental results show that the edge detected with the proposed method is exquisite, continuous and intact, and the MSE and PSNR are superior to those of traditional methods. The algorithm possesses good robustness for noised images and different format images, and can cut down the operation time by nearly half compared with global enhancement algorithm.

关 键 词：模糊增强 复合顺序形态学 边缘检测 二维直方图 结构元素 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TH74[自动化与计算机技术—计算机科学与技术]