文献类型：期刊文献

英文题名：An improved adaptive fusion edge detection algorithm for road images

作者：Xu, Guobao[1,3]; Zhou, Meijuan[1]; Xiong, Zenggang[2]; Yin, Yixin[3]

机构：[1] Lab of Ocean Remote Sensing and Information Technology, Guangdong Ocean University, Guangdong Province, China; [2] School of Computer and Information Science, Xiaogan University, Xiaogan City, Hubei Province, China; [3] School of Information Engineering, University of Sci. and Tech Beijing, Beijing City, China

年份：2012

卷号：4

期号：4

起止页码：129

外文期刊名：Advances in Information Sciences and Service Sciences

收录：EI(收录号：20121314903501)、Scopus(收录号：2-s2.0-84858779689)

语种：英文

外文关键词：Image enhancement - Robots - Median filters - Hough transforms - Image fusion - Signal detection - Roads and streets

外文摘要：As road images for visual navigation have many kinds of noise with different intensities, it is very difficult for a robot to detect the edge and profile of the actual road. Four improving measures were proposed in order to overcome the drawbacks of the existing morphological edge detection algorithm. These are: introducing the edge details on the corrosion-type edge detection operator, increasing the brightness normalization factor, choosing reasonable structuring elements, and using the optimized filter combining the median and multi-scale filtering algorithms. Based on the multi-scale edge detection and Hough transform, an improved adaptive multi-scale edge detection algorithm for fusion was presented. The experimental results showed that the proposed edge detection algorithm can not only accurately extract the edge and guide lines of the structure road, but also better extract the edge of the unstructured road from an image with noise.