| 引用本文: | 刘俊义,王润生.一种快速检测纹理斑块的方法.[J].国防科技大学学报,2001,23(2):90-93.[点击复制] |
| LIN Junyi,WANG Runsheng.An Efficient Algorithm for Texture Fault Detection[J].Journal of National University of Defense Technology,2001,23(2):90-93[点击复制] |
|
| |
|
|
| 本文已被：浏览 6459次 下载 6399次 |
| 一种快速检测纹理斑块的方法 |
| 刘俊义, 王润生 |
|
(国防科技大学 ATR国家重点实验室， 湖南 长沙 410073)
|
| 摘要: |
| 提出一种检测纹理图像中的斑块的快速算法。算法利用图像中每个位置的代表块与纹理原型的K－S距离来进行纹理分类。研究了K－S距离的基本性质，给出了纹理原型的估计方法和层次快速搜索技术。对合成图像和自然图像的实验结果表明了该方法的有效性。 |
| 关键词: 纹理 纹理分类 Kolmogorov-Smirnov距离 |
| DOI： |
| 投稿日期：2000-09-28 |
| 基金项目: |
|
| An Efficient Algorithm for Texture Fault Detection |
| LIN Junyi, WANG Runsheng |
|
(ATR State Key Lab, National Univ. of Defense Technology, Changsha 410073, China)
|
| Abstract: |
| A method for fast detection of texture faults in texture images is presented. Texture discrimination is implemented based on the K-S distance of the representative block of each site in the image and the texture prototype. The basic properties of the K-S distance are studied here. A texture prototype estimation method and a hierarchical efficient searching method are presented. The experimental results for both natural and synthetical images have proved the efficiency of this algorithm. |
| Keywords: Texture Texture discrimination Kolmogorov-Smirnov distance |
|
|
