基于改进分离阈值法的农作物遥感精细分类特征选择

doi:10.3969/j.issn.0255-8297.2019.01.006

应用科学学报 ›› 2019, Vol. 37 ›› Issue (1): 51-63.doi: 10.3969/j.issn.0255-8297.2019.01.006

基于改进分离阈值法的农作物遥感精细分类特征选择

杨惠雯^1,2, 方俊永¹, 赵冬¹

1. 中国科学院遥感与数字地球研究所人居环境遥感应用技术研究室, 北京 100101;
2. 中国科学院大学, 北京 100049

收稿日期:2018-02-28 修回日期:2018-04-16 出版日期:2019-01-31 发布日期:2019-01-31
通信作者: 方俊永,副研究员,研究方向:成像光谱技术、高光谱遥感,E-mail:fangjy@radi.ac.cn E-mail:fangjy@radi.ac.cn
基金资助:
国家重点专项课题基金（No.2016YFC0803003）资助

Feature Selection in Precise Crop Classification Using Remote Sensing Data Based on Improved SEaTH Algorithm

YANG Hui-wen^1,2, FANG Jun-yong¹, ZHAO Dong¹

1. Institute of Remote Sensing and Digital Earth, Human Settlement Division, Chinese Academy of Sciences, Beijing 100101, China;
2. University of Chinese Academy of Sciences, Beijing 100049, China

Received:2018-02-28 Revised:2018-04-16 Online:2019-01-31 Published:2019-01-31

摘要/Abstract

摘要：

为综合利用高空间分辨率可见光和红外遥感影像特征，提高分类处理效率，在分离阈值法的基础上提出一种结合图像二维熵的特征选择方法，解决了分离阈值法在特征选择时未考虑特征信息量的问题.利用随机森林分类器对选取后的特征子集进行农作物分类，并在同样的分类方法下与改进自适应波段选择方法和基于图像熵的密度峰值聚类波段选择方法进行分类精度对比.结果表明，所提出的方法既可提高分类精度和分类效率，又能降低特征选择的数量，还能进一步量化分析所选取的特征对各地物识别的效果.

关键词: 特征选择, 航空热红外遥感影像, 图像熵, 随机森林, 分离阈值, 精细分类

Abstract:

In this paper, we take full advantage of visible and infrared remote sensing images with high spatial resolution to improve classification processing efficiency. With these aims, we propose a method of feature bands selection combined with image entropy based on SEparability and THresholds (SEaTH) method. The proposed method can solve the problem that SEaTH takes no feature information into account. We apply random forest (RF) classifier in the selected features to finely classify crops of our experimental site. The proposed method is compared with those from the modified adaptive band selection (MABS) and density peaks clustering feature selection based on image entropy (IE-DPC). The comparisons illustrate that the proposed method can improve classification accuracy effectively and decrease the number of features. Meanwhile, our method can analyze quantitatively how the selected features affect identification of various land-covers.

Key words: aerial thermal infrared remote sensing image, SEparability and THresholds, random forest, image entropy, feature selection, precise classification

中图分类号:

P237

杨惠雯, 方俊永, 赵冬. 基于改进分离阈值法的农作物遥感精细分类特征选择[J]. 应用科学学报, 2019, 37(1): 51-63.

YANG Hui-wen, FANG Jun-yong, ZHAO Dong. Feature Selection in Precise Crop Classification Using Remote Sensing Data Based on Improved SEaTH Algorithm[J]. Journal of Applied Sciences, 2019, 37(1): 51-63.

参考文献

[1] 许青云,杨贵军,龙慧灵,王崇倡,李鑫川,黄登成. 基于MODIS NDVI多年时序数据的农作物种植识别[J]. 农业工程学报,2014, 30(11):134-144. Xu Q Y, Yang G J, Long H L, Wang C C, Li X C, Huang D C. Crop information identification based on MODIS NDVI time-series data[J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(11):134-144. (in Chinese)
[2] Murakami T, Ogawa S, Ishitsuka N, Kumagai K, Saito G. Crop discrimination with multitemporal SPOT/HRV data in the Saga Plains, Japan[J]. International Journal of Remote Sensing, 2001, 22(7):1335-1348.
[3] 郭交,尉鹏亮,周正舒,苏宝峰. 基于时变特征的多时相PolSAR农作物分类方法[J]. 农业机械学报,2017, 48(12):174-182. Guo J, Wei P L, Zhou Z S, Su B F. Crop classification method with differential characteristics based on multi-temporal PolSAR images[J]. Transactions of The Chinese Society of Agricultural Machinery, 2017, 48(12):174-182. (in Chinese)
[4] 余凌翔,朱勇,鲁韦坤,黎小清,赵泽,李春丽. 基于HJ-1CCD遥感影像的西双版纳橡胶种植区提取[J]. 中国农业气象,2013, 34(4):493-497. Yu L X, Zhu Y, Lu W K, Li X Q, Zhao Z, Li C L. Rubber planting area extraction in Xishuangbanna region based on HJ-1CCD remote sesnsing image[J]. Chinese Journal of Agrometeorology, 2013, 34(4):493-497. (in Chinese)
[5] Tatsumi K, Yamashiki Y, Torres M A C, Taipe C L R. Crop classification of upland fields using random forest of time-series Landsat 7 ETM+ data[J]. Computers & Electronics in Agriculture, 2015, 115:171-179.
[6] 郭鹏,武法东,戴建国,王海江,徐丽萍,张国顺. 基于无人机可见光影像的农田作物分类方法比较[J]. 农业工程学报,2017, 33(13):112-119. Guo P, Wu F D, Dai J G, Wang H J, Xu L P, Zhang G S. Comparison of farmland crop classification methods based on visible light images of unmanned aerial vehicles[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(13):112-119. (in Chinese)
[7] 田振坤, 傅莺莺, 刘素红, 刘峰. 基于无人机低空遥感的农作物快速分类方法[J]. 农业工程学报,2013, 29(7):109-116. Tian Z K, Fu Y Y, Liu S H, Liu F. Rapid crops classification based on UAV low-altitude remote sensing[J]. Transactions of The Chinese Society of Agricultural Engineering, 2013, 29(7):109-116. (in Chinese)
[8] 贾坤,李强子. 农作物遥感分类特征变量选择研究现状与展望[J]. 资源科学,2013, 35(12):2507-2516. Jia K, Li Q Z. Review of features selection in crop classification using remote sensing data[J]. Resources Science, 2013, 35(12):2507-2516. (in Chinese)
[9] Grace K, Husak G J, Harrison L, Diego P, Michaelsen J. Using high resolution satellite imagery to estimate cropped area in Guatemala and Haiti[J]. Applied Geography, 2012, 32(2):433-440.
[10] 李颖,刘荣花,郑东东. 基于多源数据和决策树估算夏玉米种植面积[J]. 中国农业气象, 2014, 35(3):344-348. Li Y, Liu R H, Zheng D D. Summer maize planting area estimation based on multi-source data and decision tree[J]. Chinese Journal of Agrometeorology, 2014, 35(3):344-348. (in Chinese)
[11] 平跃鹏,臧淑英. 基于MODIS时间序列及物候特征的农作物分类[J]. 自然资源学报,2016, 31(3):503-513. Ping Y P, Zang S Y. Crop identification based on MODIS NDVI time-series data and phenological characteristics[J]. Journal of Natural Resources, 2016, 31(3):503-513. (in Chinese)
[12] 韩文霆,李广,苑梦婵,张立元,师志强. 基于无人机遥感技术的玉米种植信息提取方法研究[J]. 农业机械学报,2017, 48(1):139-147. Han W T, Li G, Yuan M C, Zhang L Y, Shi Z Q. Extraction method of maize planting information based on UAV remote sensing technology[J]. Transactions of the Chinese Society of Agricultural Machinery, 2017, 48(1):139-147. (in Chinese)
[13] 史飞飞,高小红,杨灵玉,何林华,贾伟. 基于HJ-1A高光谱遥感数据的湟水流域典型农作物分类研究[J]. 遥感技术与应用,2017, 32(2):206-217. Shi F F, Gao X H, Yang L Y, He L H, Jia W. Research on typical crop classification based on HJ-1A hyperspectral data in the Huangshui River Basin[J]. Remote Sensing Technology and Application, 2017, 32(2):206-217. (in Chinese)
[14] 董安国,龚文娟,韩雪. 基于线性表示的高光谱影像波段选择算法[J]. 国土资源遥感,2017, 29(4):39-42. Dong A G, Gong W J, Han X. Band selection method for hyperspectral image based on linear representation[J]. Remote Sensing for Land and Resources, 2017, 29(4):39-42. (in Chinese)
[15] 赵慧洁,李明康,李娜,丁昊,蔡辉. 一种基于改进子空间划分的波段选择方法[J]. 红外与激光工程,2015, 44(10):3155-3160. Zhao H J, Li M K, Li N, Ding H, Cai H. A band selection method based on improved subspace partition[J]. Infrared and Laser Engineering, 2015, 44(10):3155-3160. (in Chinese)
[16] Zhao J F, Zhou Q, Chen Y T, Feng H J, Xu Z H, Li Q. Fusion of visible and infrared images using saliency analysis and detail preserving based image decomposition[J]. Infrared Physics & Technology, 2013, 56(2):93-96.
[17] 张爱武,杜楠,康孝岩,郭超凡. 非线性变换和信息相关的高光谱自适应波段选择[J]. 红外与激光工程,2017, 46(5):213-221. Zhang A W, Du N, Kang X Y, Guo C F. Hyperspectral adaptive band selection method through nonlinear transform and information adjacency correlation[J]. Infrared and Laser Engineering, 2017, 46(5):213-221. (in Chinese)
[18] Jia S, Tang G, Zhue, Zhu J, Li Q. A novel ranking-based clustering approach for hyperspectral band selection[J]. IEEE Transactions on Geoscience & Remote sensing, 2015, 54(1):88-102.
[19] 赵海士, 路来君, 杨晨. 一种基于图像熵的密度峰值聚类波段选择法[J]. 吉林大学学报(理学版),2017, 55(2):376-378. Zhao H S, Lu L J, Yang C. A method for density peaks clustering band selection based on image entropy[J]. Journal of Jilin University (Science Edition), 2017, 55(2):376-378. (in Chinese)
[20] 余晓敏,湛飞并,廖明生,胡金星. 利用改进SEaTH算法的面向对象分类特征选择方法[J]. 武汉大学学报(信息科学版),2012, 37(8):921-924. Yu X M, Zhan F B, Liao M S, Hu J X. Object-oriented feature selection algorithms based on improved SEaTH algorithms[J]. Geomatics and Information Science of Wuhan University, 2012, 37(8):921-924. (in Chinese)
[21] Nussbaum S, Niemeyer I, Canty M J. SEaTH-a new tool for automated feature extraction in the context of object-based image analysis[C]//International Conference on Object-Based Image Analysis, 2006.
[22] Gao Y, Marpu P, Niemeyer I, Runfola D M, Mginer N, Hamill T, Pontius R G Jr. Object-based classification with features extracted by a semi-automatic feature extraction algorithm-SEaTH[J]. Geocarto International, 2011, 26(3):211-226.
[23] Wang H, Chen J S, Yu X M. Feature selection and its application in object-oriented classification[J]. Journal of Remote Sensing, 2013, 17(4):816-829.
[24] 向天烛,高熔溶,闫利,徐振亮. 一种顾及区域特征差异的热红外与可见光图像多尺度融合方法[J]. 武汉大学学报(信息科学版),2017, 42(7):911-917. Xiang T Z, Gao R R, Yan L, Xu Z L. Region feature based multi-scale fusion method for thermal infrared and visible image[J]. Geomatics and Information Science of Wuhan University. 2017, 42(7):911-917. (in Chinese)

基于改进分离阈值法的农作物遥感精细分类特征选择

Feature Selection in Precise Crop Classification Using Remote Sensing Data Based on Improved SEaTH Algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	范书瑞, 贾雅亭, 刘晶花. 基于三轴加速度传感器人体姿态识别的特征选择[J]. 应用科学学报, 2019, 37(3): 427-436.
[2]	孙物一, 冯国瑞. 基于高维特征正则化的JPEG图像隐写分析[J]. 应用科学学报, 2016, 34(5): 555-563.
[3]	胡昭华, 徐玉伟, 赵孝磊, 何军, 周游. 基于支持向量机的多特征选择目标跟踪[J]. 应用科学学报, 2015, 33(5): 502-517.
[4]	王进1;3，丁凌1，孙开伟1，李钟浩2. 过滤特征基因选择及演化硬件急性白血病分型[J]. 应用科学学报, 2012, 30(3): 287-293.