1. 中国科学院 机构 国家空间科学中心 微波遥感技术重点实验室,北京,100190
2. 中国科学院 国家空间科学中心,北京,100190
3. 中国科学院大学 机构,北京,101408
纸质出版:2019
移动端阅览
贾立丽, 张升伟, 何杰颖, 李娜. 基于多任务联合稀疏表示的高光谱图像分类算法[J]. 微电子学与计算机, 2019, 36(10): 15-20.
JIA Li-li, ZHANG Sheng-wei, HE Jie-ying, et al. Joint sparse representation and multitask learning for hyperspectral lmage clasification[J]. 2019, 36(10): 15-20.
为了充分提取高光谱图像丰富的光谱信息
尽可能降低光谱冗余
同时保留较多有效判别信息.基于高光谱图像单波段的谱间相似性
提出了基于多任务学习和稀疏表示的分类算法.该方法将光谱间的冗余特性转化为有效信息加以利用
使用波段交叉分组策略构建子任务
并引入稀疏表示模型将所有分类任务进行联合表示
最终根据所有任务的累积残差确定测试样本的类别.实验对比分析了多任务联合表示分类和单任务分类的准确率.结果表明
基于多任务联合稀疏表示模型的分类性能优于单任务模型.
Hyperspectral images have abundant spectral information
and thus the adjacent single-band images usually contain similar and redundant information. To simultaneously reducing spectral redundancy and retaining more effective discriminant information
a novel classification algorithm based on multitask learning and sparse representation is proposed in this paper. In this method
the spectral redundancy is transformed into effective information and a band cross-grouping strategy is used to construct sub-tasks. Sparse representation model is introduced to jointly represent all classification tasks. Finally
the test samples are classified according to the cumulative residual of all tasks. An experiment is performed between multitask learning and the single task learning. The experimental results show that the classification performance of the former is better than the later.
ZHANG Y, DU B, ZHANG L, et al. A sparse representation based binary hypothesis model for target detection in hyperspectral images[J]. IEEE Trans. Geosci. Remote Sens, 2015, 53(3): 1346-1354. DOI: 1346
CHEN YI. Research on spatial-spectral resolution enhancement of hyperspectral image[D]. Xi'an: Northwestern Polytechnical University, 2016.
LI W, TRAMEL E W, PRASAD S, et al. Nearest regularized subspace for hyperspectral classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(1):477-489. DOI: 477
薛晓琴, 基于协同表示的高光谱图像亚像元制图技术研究[D].西安: 西北工业大学, 2017.
HUANG Z, SHI Z, YANG S, et al. Nonlocal similarity regularized sparsity model for hyperspectral target detection[J]. IEEE Geosci. Remote Sens. 2013, 10(6):1532-1536. DOI: 1532
余乐, 吴超, 吴静珠, 等.结合高光谱与CNN的小麦不完善粒识别方法[J].电子测量与仪器学报, 2017, 31(8):1297-1303. DOI: 1297
ZHANG Y, DU B, ZHANG L, et al. Joint Sparse Representation and Multitask Learning for Hyperspectral Target Detection[J]. IEEETransactionsonGeoscience % RemoteSensing, 2017, 55(2):894-906. DOI: 894
赵春晖, 李晓慧, 王玉磊, 等.高光谱图像异常目标检测研究进展[J].电子测量与仪器学报, 2014, 28(8): 803-811. DOI: 803
CHEN J, ZHOU J, YE J, et al. IntegratingLow-rank and Group-sparse Structures for Robust Multi-task Learning[C]// International Conference on Knowledge Discoveryand Data Mining. San Diego, USA, 2011: 281-288.
HUANG Z, SHI Z, QIN Z, et al. A dimension reduction model for sparse hyperspectral target detection with weighted l1-minimization[C]// IEEE 5th Int. Congr. Image Signal Process.Chongqing, China. 2012: 972-976.
0
浏览量
12
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010602201714号