摘要
根据图象边缘灰度的梯度变化 ,构造图象灰度模糊矩阵和描述边缘点的隶属函数 ,利用遗传算法实现隶属函数各参数的寻优过程 ,并由输出隶属度判断提取图象边缘点 ,实现了图象的边缘检测。实验表明 ,该方法能有效地描述边缘的穿越过程 ,并可改善检测结果。
A fuzzy gradient method based on the character of gray gradient of the image edge is proposed. An image gray scale fuzzy matrix is constructed and a suitable membership function is selected. Genetic algorism is used to achieve the optimum parameters of the membership function. The edge points are determined and extracted according to the output of function. The edge detection is carried out. The experiment results show that this method has the ability to describe the edge crossing process and to improve the result of edge detection.
出处
《控制与决策》
EI
CSCD
北大核心
2001年第3期351-354,共4页
Control and Decision
基金
国家自然科学基金!项目 (396 70 6 0 7)
