数字图像边缘检测方法的对比分析及优化

A Comparative Analysis and Optimization of Numerical Image Edge Detection Method

由于数字图像本身的复杂特性和各种噪声源的影响,使得图像边缘检测技术成为图像处理的一个难点.通过对几种常用边缘检测算法的介绍、归纳总结及仿真实验,对比分析了各自的优缺点,并在此基础上结合目前先进的小波理论和传统的微分法提出了一种新的图像边缘检测方法.新方法把用不同方法处理得到的边缘图像融合生成一幅新的边缘图像,先取适当阈值,过滤多余的信息.经Matlab仿真验证,在突出图像的边缘和局域细节信息方面具有良好效果,是一种有效的图像边缘检测方法.

Because of the complex nature of the digital image itself and the influence of various noise sources,the image edge detection technology as image processing becomes a difficult problem. Some commonly used algorithms for edge detection are introduced. A comparative analysis is made on the advantages and disadvantages of the summaries and the simulation experiments ~ on this basis,a new image edge detec- tion method is proposed, combining advanced wavelet theory with the traditional differential method. The method by different methods to deal with the a new edge image is generated before taking proper thresh- olds,and removing redundant information. Matlab simulation proves that with outstanding image edges and local area details information, it is an effective image edge detection method.