| 基于Gabor滤波器和特征加权的红外图像识别 |
| |
| 引用本文： | 赵英男,刘正东,杨静宇. 基于Gabor滤波器和特征加权的红外图像识别[J]. 计算机工程与应用, 2004, 40(32): 22-24 |
| |
| 作者姓名： | 赵英男 刘正东 杨静宇 |
| |
| 作者单位： | 南京理工大学计算机系,南京,210094;南京理工大学计算机系,南京,210094;南京理工大学计算机系,南京,210094 |
| |
| 摘 要： | 针对红外图像边缘模糊、噪声较多的特点,文中提出一种新的基于Gabor滤波器和特征加权的红外图像识别方法。该方法利用Gabor滤波器的多尺度特性对红外图像进行特征抽取,然后根据Gabor特征矢量中邻近分量的离散程度对其自身进行加权。通过对三种不同类型的红外车辆进行识别测试以及与传统的方法进行比较,结果表明Gabor特征矢量经过加权处理后,能够有效降低红外图像识别的错误率,增强鲁棒性。
|
| 关 键 词： | Gabor滤波 识别 红外图像 特征提取 特征加权 鲁棒性 |
| 文章编号： | 1002-8331-(2004)32-0022-03 |
Recognition for Infrared Images Based on Gabor Filters and Feature Weighting |
| |
| Zhao Yingnan Liu Zhengdong Yang Jingyu. Recognition for Infrared Images Based on Gabor Filters and Feature Weighting[J]. Computer Engineering and Applications, 2004, 40(32): 22-24 |
| |
| Authors: | Zhao Yingnan Liu Zhengdong Yang Jingyu |
| |
| Abstract: | To weaken the effects of blur edges and much noise in infrared images,a novel approach on recognition for infrared images based on Gabor filters and feature weighting is presented in this paper.It consists of two main steps:feature extraction and feature weighting,which weights the raw features derived from2D Gabor filters according to their own degree of dispersion.Some classified experiments on three kinds of infrared vehicle images and the comparison with the conventional method are performed to test the performance of the new technique.And the experimental results show visible improvements both in diminishing error rate and robustness. |
| |
| Keywords: | Gabor filter recognition infrared image feature extraction feature weighting robustness |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
