中国激光, 2010, 37 (6): 1599, 网络出版: 2010-07-07
激光主动成像图像边缘检测方法
Edge Detection Algorithm of Laser Active Imaging
成像系统 激光主动成像 边缘检测 多分辨分析 多尺度形态学 image system laser active imaging edge detection multi-resolution analysis multi-scale morphology
摘要
针对实际应用中激光主动成像图像的特点,提出了一种基于小波的非线性增强和多尺度形态学相结合的方法对激光主动成像图像进行边缘检测。从小波多分辨分析出发,区分噪声和图像的边缘细节,构造基于小波的非线性增强算法对图像的边缘细节适度增强。引入多尺度形态学对增强后的图像进行边缘提取,利用大尺度形态算子抑制噪声,小尺度形态算子提取目标边缘细节信息,获得了较清晰的图像边缘。实验结果表明,与传统的边缘检测方法相比,该方法能够检测到完整清晰的激光主动成像图像边缘,且抗噪性能良好。
Abstract
Laser active imaging system can obtain intensity image, range image and velocity image, and it provides more useful information than other imaging systems. According to the features of laser active imaging, a new edge detection algorithm of laser active image based on wavelet transform and multi-scale morphology is presented. The wavelet multi-resolution analysis is used to differentiate the noise and the edge detail of the image, and then a non-linear algorithm is constructed to strengthen the wake parts of the image. The concept of the multi-scale morphology is introduced to obtain the edge of the processed image, bigger morphology operator eliminating noise and smaller morphology operator is used to pick up image detail, and the better edge of the processed image is obtained. The experimental results indicate that compared with classical edge detection operator, this method has better noise immunity and performance on edge detection of laser active imaging with large noise and blurry edge.
李晓峰, 徐军, 张胜修, 罗积军, 朱桂芳, 徐利国. 激光主动成像图像边缘检测方法[J]. 中国激光, 2010, 37(6): 1599. Li Xiaofeng, Xu Jun, Zhang Shengxiu, Luo Jijun, Zhu Guifang, Xu Liguo. Edge Detection Algorithm of Laser Active Imaging[J]. Chinese Journal of Lasers, 2010, 37(6): 1599.