摘要
对于行为识别的重要前提就是图像的边缘检测,针对传统图像边缘检测算法在微弱轮廓提取精度低的问题,对传统的边缘特征提取方法进行了改进.进一步提高了目标图像的边缘检测精度.首先对图像进行二值化预处理,去除光线、噪音等干扰,其次使用EDGE函数对图像进行边缘提取,最后提出边缘检测算子和梯度算子融合算法对图像边缘进行提取,实验结果表明该算法能有效地提高边缘检测的精度.改进后的方法比之传统检测方法精确度得到了明显的提高.
A target contour detection algorithm based on image enhancement and edge detection was designed to enhance the low precision of the target contour detection algorithm in the weak contour extraction of edge features.Firstly,the target image was binarized to reduce the interference of light and noise on the image edge detection,by which the image had a preliminary processing.Then,the binarized target image was further processed using the EDGE function.Finally,a fusion algorithm of edge detection operator and gradient operator was proposed to extract image edges.Experimental results show that this algorithm can improve the accuracy of edge detection effectively.The accuracy by using the improved method was obviously enhanced compared with that of the traditional method.
作者
李建军
李轲赛
LI Jianjun;LI Kesai(Information Engineering School,Inner Mongolia University of Science and Technology,Baotou 014010,China)
出处
《内蒙古科技大学学报》
CAS
2020年第1期78-81,共4页
Journal of Inner Mongolia University of Science and Technology
基金
内蒙古自治区自然科学基金资助项目(2018MS06018)
内蒙古自治区杰出青年培育基金资助项目(2018JQ02).
关键词
二值化
边缘检测
梯度特征
图像增强
binarization
edge detection
gradient features
Image enhancement