Snake模型初始轮廓选取的研究  

Study on Initial Contour Selection of Snake Model

作　　者：李丽勤[1] 高焕文[1] 周兴祥[1] 

机构地区：[1]中国农业大学工学院,北京100083

出　　处：《计算机工程与应用》2004年第11期43-45,共3页Computer Engineering and Applications

基　　金：国家"十五"科技攻关课题(编号:2001BA504B03)

摘　　要：针对Snake模型处理复杂背景图像时的初始轮廓自动选取问题,该文采用分水岭算法先对图像进行分割,并将得到的边界作为Snake模型的初始边缘轮廓。由于分水岭算法具有将目标物体从复杂背景中分割开来的优点,使得在应用Snake模型对复杂图像进行分割时减少了人工的干预。经过实验对比,采用分水岭算法对玉米秸秆图像进行边缘轮廓的提取能达到较好的效果,为自动进行Snake模型的计算提供了一种较好的初始轮廓处理方法。For automatic selection of ori gi nal object contour to snake model from complex background,the watershed algori thm is adopted to extract object contour in the paper,and then the extracted o bject contour is taken as initial contour of snake model for precise segmentat ion computation.The manual operation is reduced during the process of segmentin g complex image with snake model due to advantages of the watershed algorithm.T he result from comparative experiments shows that using watershed algorithm can segment images of maize stalk efficiently,and the described approach is prefe rable for automatic computation of snake model.

关 键 词：图像分割 SNAKE模型 分水岭算法 

分 类 号：TN911[电子电信—通信与信息系统]