摘要
根据多尺度对比度增强算法在增强图像对比度的同时,较好地抑制图像噪声的特性,文中将多尺度分解应用到边缘检测中。先对图像进行多尺度分解,分解后得到的不同尺度的高频分量分别集中了图像的边缘信息和噪声,然后对边缘信息进行增强,对噪声信息和伪边缘进行抑制,再对处理后的图像结合Canny算子检测图像的边缘。通过实验对比发现该算法比传统的边缘检测算更能准确地提取图像的边缘信息的效果,特别在图像噪声较大或对比度较低的情况下,该算法进行边缘检测的效果更加明显。
According to the characteristic that the multi-scale contrast enhancement algorithm can reduce image noise efficiently while en hancing image, in this paper, multi-scale contrast enhancement algorithm is applied to image edge detection. Firstly, the image is decom posed through multi-scale,after decomposition the high-frequency part of different scale gathered the image＇s edge information and noi ses separately, then the edge information was improved, and the noise and the false edge was reduced, finally the Canny edge detection al- golithm was applied to detect the edge of the processed images. Experiment shows that comparing with the commonly edge detection al- gorithm,this algorithm can detect the image edges more effectively,especially to those noise image and low-contrast image,better effects can be got by this algorithm.
出处
《计算机技术与发展》
2012年第12期228-232,共5页
Computer Technology and Development
基金
广东省2010年自然科学基金项目(10151064007000000)
2010年广东省高等学校高层次人才项目(粤教师函字[2010]79号)
广东省2009年社会发展重点科技计划项目(2009A030200016)
关键词
多尺度对比度增强算法
边缘检测
噪声
低对比度
multi-scale contrast enhancement algorithra
edge detection
noise
low-contrast
