节点文献
基于小波变换的近红外图像特征提取方法
An approach of near infrared image feature extraction based on wavelet transform
【摘要】 提出用二次样条小波对近红外数字图像进行特征边缘提取,与传统边缘检测算子进行比较,并在W indows平台的VC++6.0集成开发环境下编程实现.实验结果显示,当目标对比度下降到一定程度时,常规算子无法提取出弱目标,利用二次样条小波变换能够将近红外图像中弱目标的特征边缘提取出来,效果较好,且算法简单,效率较高.
【Abstract】 Considering that image edge extraction is a fundamental step in many images processing technology,this paper concentrates on proposing an edge extraction algorithm of near infrared image based on 2-spline wavelet transform.This algorithm is compared with Roberts,Sobel,Laplacian and Canny operator.The experimental results show that when target contrast is reduced to a certain level,traditional operator cannot identify faint target while 2-spine wavelet transform has a better target edge identification effect.This algorithm is simple and efficient.
【关键词】 二次样条小波;
滤波器;
边缘提取;
近红外图像;
【Key words】 2-spline wavelet; filter; edge extraction; near infrared image;
【Key words】 2-spline wavelet; filter; edge extraction; near infrared image;
【基金】 广东省科技计划资助项目(2004B16001210)
- 【文献出处】 深圳大学学报 ,Shenzhen University Journal , 编辑部邮箱 ,2006年02期
- 【分类号】TP391.41
- 【被引频次】10
- 【下载频次】597