RGB空间下基于矢量梯度的彩色图像边缘检测算法改进  

Color Image Edge Detection Algorithm Based on Gradient Vector in RGB Space

作　　者：乔冬冬[1] 赵子涛[1] 

机构地区：[1]西安工程大学计算机科学学院,陕西西安710048

出　　处：《软件导刊》2016年第6期42-44,共3页Software Guide

摘　　要：由于彩色图像能够提供比灰度图像更为丰富的信息,因此彩色图像处理越来越受到人们的重视。为了充分利用彩色信息进行边缘检测,提出了基于多方向矢量梯度的彩色图像边缘检测方法。在分析传统基于矢量梯度的边缘检测方法存在边缘不连续、漏检等问题的基础上,通过引入多方向矢量梯度,综合考虑不同方向对图像梯度的响应,并借助信息融合技术融合多个方向下的检测结果,从而得到最终边缘检测结果。实验结果表明,该改进算法对边缘提取具有更好的连续性和定位准确性,而且对低照度下的边缘图像具有较好的响应。As color images can provide more abundant information than the gray image, color image processing has attracted more and more attention. In order to use the color information of the edge detection fully, put forward vector color image edge detection algorithm based on multi-direction gradient. First analyzes the traditional edge detection algorithm based on gradient vector edge discontinuity, undetected problems. Secondly, through the introduction of multi direction gradient vector, considering different directions of the gradient image response. The test results by information fusion technology integrating multiple direction, to get the final detection results of edge. The experimental results show that the continuity and positioning accuracy of the improved algorithm of edge extraction is better in this paper, and the edge image of low illumination has good response.

关 键 词：边缘检测 矢量梯度 信息融合技术 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]