基于未标签信息主动学习算法的高光谱影像分类

doi:10.11772/j.issn.1001-9081.2017.06.1768

计算机应用 ›› 2017, Vol. 37 ›› Issue (6): 1768-1771.DOI: 10.11772/j.issn.1001-9081.2017.06.1768

• 计算机视觉与虚拟现实 • 上一篇下一篇

基于未标签信息主动学习算法的高光谱影像分类

张良¹, 罗祎敏², 马洪超², 张帆¹, 胡川³

1. 湖北大学资源环境学院, 武汉 430062;
2. 武汉大学遥感信息工程学院, 武汉 430079;
3. 国网湖北省电力公司检修公司, 武汉 430077

收稿日期:2016-10-31 修回日期:2017-01-12 出版日期:2017-06-10 发布日期:2017-06-14
通讯作者: 张良
作者简介:张良(1986-),男,浙江绍兴人,讲师,博士,主要研究方向:机器学习、遥感影像智能分类、三维点云数据处理;罗祎敏(1993-),女,湖北武汉人,硕士研究生,主要研究方向:遥感影像智能分类;马洪超(1968-),男,浙江绍兴人,教授,博士,主要研究方向:机器学习、遥感影像智能分类、三维点云数据处理;张帆(1981-),男,湖北武汉人,讲师,博士,主要研究方向:机器学习、信号处理;胡川(1985-),男,湖北黄石人,硕士研究生,主要研究方向:机器学习、智能电网。
基金资助:
国家自然科学基金资助项目（41601504）。

Hyperspectral remote sensing image classification based on active learning algorithm with unlabeled information

ZHANG Liang¹, LUO Yimin², MA Hongchao², ZHANG Fan¹, HU Chuan³

1. Faculty of Resources and Environmental Science, Hubei University, Wuhan Hubei 430062, China;
2. School of Remote Sensing and Information Engineering, Wuhan University, Wuhan Hubei 430079, China;
3. Maintenance Company, State Grid Hubei Electric Power Company, Wuhan Hubei 430077, China

Received:2016-10-31 Revised:2017-01-12 Online:2017-06-10 Published:2017-06-14
Supported by:
This work is partially supported by the National Natural Science Foundation of China (41601504).

摘要/Abstract

摘要： 针对高光谱遥感影像分类中，传统的主动学习算法仅利用已标签数据训练样本，大量未标签数据被忽视的问题，提出一种结合未标签信息的主动学习算法。首先，通过K近邻一致性原则、前后预测一致性原则和主动学习算法信息量评估3重筛选得到预测标签可信度高并具备一定信息量的未标签样本；然后，将其预测标签当作真实标签加入到标签样本集中；最后，训练得到更优质的分类模型。实验结果表明，与被动学习算法和传统的主动学习算法相比，所提算法能够在同等标记的代价下获得更高的分类精度，同时具有更好的参数敏感性。

关键词: 高光谱遥感, 主动学习, 图像分类, 未标签信息。

Abstract: In hyperspectral remote sensing image classification, the traditional active learning algorithms only use labeled data for training sample, massive unlabeled data is ignored. In order to solve the problem, a new active learning algorithm combined with unlabeled information was proposed. Firstly, by realizing triple screening of K neighbor consistency principle,predict consistency principle, and information evaluation of active learning, the unlabeled sample with a certain amount of information and highly reliable prediction label was obtained. Then, the prediction label was added to the label sample set as real label. Finally, an optimized classification model was produced by training the sample. The experimental results show that, compared with the passive learning algorithms and the traditional active learning algorithms, the proposed algorithm can obtain higher classification accuracy under the precondition of the same manual labeling cost and get better parameter sensitivity.

Key words: hyperspectral remote sensing, active learning, image classification, unlabeled information

中图分类号:

P407.8

张良, 罗祎敏, 马洪超, 张帆, 胡川. 基于未标签信息主动学习算法的高光谱影像分类[J]. 计算机应用, 2017, 37(6): 1768-1771.

ZHANG Liang, LUO Yimin, MA Hongchao, ZHANG Fan, HU Chuan. Hyperspectral remote sensing image classification based on active learning algorithm with unlabeled information[J]. Journal of Computer Applications, 2017, 37(6): 1768-1771.

参考文献

[1] 郝泽东,余淞淞,关佶红.基于主动学习的高光谱图像分类方法[J].计算机应用,2013,33(12):3441-3443,3448.(HAO Z D, YU S S, GUAN J H. Hyperspectral image classification based on active learning[J]. Journal of Computer Applications, 2013, 33(12):3441-3443, 3448.)
[2] CHEN Y C, LIN Z H, ZHAO X, et al. Deep learning-based classification of hyperspectral data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(6):2094-2107.
[3] RAJAN S, GHOSH J. Exploiting class hierarchies for knowledge transfer in hyperspectral data[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(11):3408-3417.
[4] KIM W, CRAWFORD M M. Adaptive classification for hyperspectral image data using manifold regularization kernel machines[J]. IEEE Transaction on Geoscience and Remote Sensing, 2012, 48(11):4110-4121.
[5] 陈进.高光谱图像分类方法研究[D].长沙:国防科学技术大学,2010:11-12.(CHEN J. On classification method of hyperspectral images[D]. Changsha:National University of Defense Technology, 2010:11-12.)
[6] TUIA D, RATLE F, PACIFICI F, et al. Active learning methods for remote sensing image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2009, 47(7):2218-2232.
[7] 刘康.基于主动学习的高光谱图像分类技术研究[D].北京:中国矿业大学(北京),2014:12-13.(LIU K. Hyperspectral sensing image classification technology based on active learning[D]. Beijing:China University of Mining and Technology (Beijing), 2014:12-13.)
[8] WAN L J, TANG K, LI M Z, et al. Collaborative active and semisupervised learning for hyperspectral remote sensing image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015. 53(5):2384-2396.
[9] ANGLUIN D. Queries and concept learning[J]. Machine Learning, 1988, 2(4):319-342.
[10] TUIA D, RATLE F, PACIFICI F, et al. Active learning methods for remote sensing image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2009, 47(7):2218-2232.
[11] TUIA D, PASOLLI E, EMERY W J. Using active learning to adapt remote sensing image classifiers[J]. Remote Sensing of Environment, 2011, 115(9):2232-2242.
[12] DI W, CRAWFORD M M. View generation for multiview maximum disagreement based active learning for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2012, 50(5):1942-1954.
[13] DI W, CRAWFORD M M. Multi-view adaptive disagreement based active learning for hyperspectral image classification[C]//Proceedings of the 2010 IEEE International Geoscience and Remote Sensing Symposium. Piscataway, NJ:IEEE, 2010:1374-1377.
[14] CRAWFORD M M, TUIA D, YANG H L. Active learning:any value for classification of remotely sensed data[J]. Proceedings of the IEEE, 2013, 101(3):593-608.
[15] LI J, BIOUCAS-DIAS J M, PLAZA A. Semi-supervised hyperspectral image segmentation using multinomial logistic regression with active learning[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(11):4085-4098

基于未标签信息主动学习算法的高光谱影像分类

Hyperspectral remote sensing image classification based on active learning algorithm with unlabeled information

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	贾鹤鸣, 郎春博, 姜子超. 基于轻量级卷积神经网络的植物叶片病害识别方法[J]. 计算机应用, 2021, 41(6): 1812-1819.
[2]	许学斌, 张佳达, 刘伟, 路龙宾, 赵雨晴. 融合空间和通道特征的高精度乳腺癌分类方法[J]. 计算机应用, 2021, 41(10): 3025-3032.
[3]	郑延斌, 韩梦云, 樊文鑫. 基于二维主成分分析与卷积神经网络的手写体汉字识别[J]. 计算机应用, 2020, 40(8): 2465-2471.
[4]	易东义, 邓根强, 董超雄, 祝苗苗, 吕周平, 朱岁松. 基于图卷积神经网络的医保欺诈检测算法[J]. 计算机应用, 2020, 40(5): 1272-1277.
[5]	边小勇, 江沛龄, 赵敏, 丁胜, 张晓龙. 基于多分支神经网络模型的弱监督细粒度图像分类方法[J]. 计算机应用, 2020, 40(5): 1295-1300.
[6]	吴立人, 刘政浩, 张浩, 岑悦亮, 周维. 聚焦图像对抗攻击算法PS-MIFGSM[J]. 计算机应用, 2020, 40(5): 1348-1353.
[7]	张伍, 陈红梅. 基于多核模糊粗糙集与蝗虫优化算法的高光谱波段选择[J]. 计算机应用, 2020, 40(5): 1425-1430.
[8]	郭志强, 胡永武, 刘鹏, 杨杰. 基于特征融合的室外天气图像分类[J]. 计算机应用, 2020, 40(4): 1023-1029.
[9]	张凯琳, 阎庆, 夏懿, 章军, 丁云. 基于焦点损失的半监督高光谱图像分类[J]. 计算机应用, 2020, 40(4): 1030-1037.
[10]	徐戈, 肖永强, 汪涛, 陈开志, 廖祥文, 吴运兵. 基于视觉误差与语义属性的零样本图像分类[J]. 计算机应用, 2020, 40(4): 1016-1022.
[11]	张盼盼, 李其申, 杨词慧. 基于轻量级分组注意力模块的图像分类算法[J]. 计算机应用, 2020, 40(3): 645-650.
[12]	史梦安, 陆振宇. 基于YOLO-tiny-RFB模型的电站旋钮开关状态识别[J]. 计算机应用, 2020, 40(12): 3679-3686.
[13]	汪敏, 武禹伯, 闵帆. 基于多种聚类算法和多元线性回归的多分类主动学习算法[J]. 计算机应用, 2020, 40(12): 3437-3444.
[14]	武宽, 秦品乐, 柴锐, 曾建朝. 基于不同超声成像的甲状腺结节良恶性判别[J]. 计算机应用, 2020, 40(1): 77-82.
[15]	张伍, 陈红梅. 基于核模糊粗糙集的高光谱波段选择算法[J]. 计算机应用, 2020, 40(1): 258-263.