基于随机特征字典的纹理分类方法
为解决稀疏表示在提取全局纹理特征时受维数限制的问题，提出一种基于随机特征字典的特征提取及分类方法。方法利用稀疏系数中非零系数的分布特点，统计各图像块在稀疏分解过程中字典原子的使用频率，得到能突出纹理在稀疏域类别信息的直方图特征，进而实现分类。为提高分类准确率，通过随机投影将多尺度多方向的小波特征进行融合，并对其训练得到纹理描述能力更强的小波随机特征字典。在分类实验中，其分类准确率达94．79％，并能在噪声、光照条件影响下获得较好的鲁棒性，在分析全局纹理特征方面具有高效、稳定的特点。...
Saved in:
| Published in | 计算机应用研究 Vol. 32; no. 1; pp. 303 - 306 |
|---|---|
| Main Author | |
| Format | Journal Article |
| Language | Chinese |
| Published |
合肥工业大学计算机与信息学院,合肥,230009
2015
|
| Subjects | |
| Online Access | Get full text |
| ISSN | 1001-3695 |
| DOI | 10.3969/j.issn.1001-3695.2015.01.071 |
Cover
Loading…
| Summary: | 为解决稀疏表示在提取全局纹理特征时受维数限制的问题，提出一种基于随机特征字典的特征提取及分类方法。方法利用稀疏系数中非零系数的分布特点，统计各图像块在稀疏分解过程中字典原子的使用频率，得到能突出纹理在稀疏域类别信息的直方图特征，进而实现分类。为提高分类准确率，通过随机投影将多尺度多方向的小波特征进行融合，并对其训练得到纹理描述能力更强的小波随机特征字典。在分类实验中，其分类准确率达94．79％，并能在噪声、光照条件影响下获得较好的鲁棒性，在分析全局纹理特征方面具有高效、稳定的特点。 |
|---|---|
| Bibliography: | 51-1196/TP sparse representation; dictionary learning; texture classification; global texture feature extraction Extracting global texture feature through sparse representation faced some problems, which mainly caused by high dimension. In order to solve those problems, this paper proposed a feature extraction and classification method based on ran- dom feature dictionary. The proposed method utilized the distribution of non-zero coefficients, which were computed by sparse decomposition, to generate a statistics histogram feature. The acquired histogram could reflect the dictionary atoms＇ using fre- quency in sparse decomposition, and was able to reflect the class information, Thus, the classification could be realized. For the sake of improving classification accuracy, it fused multi-scale and multi-direction wavelet features through random projec- tion, and then trained a more descriptive dictionary by those fused features. In the classification experiments, it achieved 94.79% classification accuracy. Further |
| ISSN: | 1001-3695 |
| DOI: | 10.3969/j.issn.1001-3695.2015.01.071 |