高光谱影像的BDT-SVM地物分类算法与应用

doi:10.11873/j.issn.1004-0323.2016.1.0177

鎽樿/Abstract

鎽樿锛�

闈㈠娴烽噺鏁版嵁鐨勭壒寰佺┖闂撮珮缁存�у強璁粌鏍锋湰鐨勬湁闄愭��,楂樺厜璋遍仴鎰熷奖鍍忚嫢閲囩敤甯歌缁熻妯″紡鐨勫垎绫绘柟娉曢毦浠ヨ幏寰楄緝濂界殑鍒嗙被缁撴灉銆傚洜姝ゆ帰璁ㄦ敮鎸佸悜閲忔満(SVM)鍒嗙被鍣ㄧ殑鍩烘湰鍘熺悊,閽堝EO-1 Hyperion楂樺厜璋卞奖鍍忕殑鍒嗙被鐗圭偣鍙婄幇鏈夊绫籗VM绠楁硶鎵�瀛樺湪鐨勮缁冩椂闂撮暱鍙婂垎绫荤簿搴︿綆绛夐棶棰�,寮曞叆浜屽弶鍐崇瓥鏍慡VM(BDT-SVM)鍒嗙被绠楁硶,骞舵彁鍑轰竴绉嶆柊鐨勭被闂村垎绂诲害瀹氫箟鏂规硶鍙婄浉搴旂殑瀹㈣纭畾浜屽弶鏍戠粨鏋勭殑绛栫暐,鐢辨鐢熸垚鏀硅繘鐨凚DT\|SVM绠楁硶銆傚疄楠岀粨鏋滆〃鏄�:涓庡叾浠栧绫诲垎绫绘柟娉曠浉姣�,鍩轰簬鏀硅繘鐨凚DT-SVM绠楁硶鐨勯珮鍏夎氨褰卞儚鍦扮墿鍒嗙被鏁堟灉鏇村ソ,鎬讳綋绮惧害杈惧埌90.96%,Kappa绯绘暟涓�0.89,璇ョ畻娉曡繕瑙ｅ喅浜嗙粡鍏窼VM澶氱被鍒嗙被鍙兘瀛樺湪鐨勪笉鍙垎鍖哄煙闂銆�

鍏抽敭璇�: 楂樺厜璋卞奖鍍�, 鏀寔鍚戦噺鏈�(SVM), 浜屽弶鍐崇瓥鏍�(BDT), 鍒嗙被绠楁硶

Abstract:

Hyperspectral remote sensing is a cutting edge field in remote sensing.It offers the fine detection of objects by its spectral response characteristics in various spectral bands,and has superiority to multispectral remote sensing in fine extraction.However,due to high\|dimensional feature space and limited training samples of the huge data of hyperspectral images,it is difficult for conventional statistical pattern identification methods to classify hyperspectral images.Thus this paper explores the basic principle of support vector machine classifier and employs Binary Decision Tree Support Vector Machine (BDT\|SVM) classification algorithm based on EO\|1 Hyperion hyperspectral imagery.And this study proposes a new definition of the class separation against the long training time and low classification efficiency of existing multi\|class SVM algorithm and generates a modified BDT\|SVM algorithm.On the basis of theoretical analysis,this paper completes the object classification experiments on Hyperion hyperspectral imagery of the test area and verifies the high classification accuracy of the method.Experimental results show that the effect of hyperspectral image classification based on the modified BDT\|SVM algorithm is apparently better than other multi\|class classification methods,which total classification accuracy is up to 90.96% and kappa coefficient is 0.89.The algorithm also solves the problem of non\|separable region,which may be present in the classic SVM multi\|class classification methods.

Key words: Hyperspectral imagery, Support Vector Machine(SVM), Binary Decision Tree(BDT), Classification algorithm

涓浘鍒嗙被鍙�:

TP 751.1

鏋楀織鍨�,鏅忚矾鏄�. 楂樺厜璋卞奖鍍忕殑BDT-SVM鍦扮墿鍒嗙被绠楁硶涓庡簲鐢�僛J]. 閬ユ劅鎶�鏈笌搴旂敤, 2016, 31(1): 177-185.

Lin Zhilei,Yan Luming. The Object Classification Algorithm and Application for Hyperspectral Imagery based on BDT-SVM[J]. Remote Sensing Technology and Application, 2016, 31(1): 177-185.

鍙傝�冩枃鐚�

锛�1锛絇ignatti S,Cavalli R M,Cuomo V,顎奺t al顎�.Evaluating Hyperion Capability for Land Cover Mapping in a Fragmented Ecosystem:Pollino National Park,Italy锛籎锛�.Remote Sensing of Environment,2009,113(3):622-634.顎�

锛�2锛紺hi M M,Feng R,Bruzzone L.Classification of Hyper-spectral Remote Sensing Data with Primal SVM for Small-sized Training Dataset Problem锛籎锛�.Advances in Space Research,2008,41(11):1793-1799.顎�

锛�3锛組ountrakis G,Im J,Ogole C.Support Vector Machines in Remote Sensing:A Review锛籎锛�.ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(3):247- 259.顎�

锛�4锛絑hou L G,Lai K K,Yu L.Least Squares Support Vector Machines Ensemble Models for Ccredit Scoring锛籎锛�.Expert System with Application,2010,37(1):127-133.顎�

锛�5锛絊engur A.Multiclass Least-squares Support Vector Machines for Analog Modulation Classification锛籎锛�.Expert Systems with Applications,2009,36(3):6681-6685.顎�

锛�6锛組elgani F,Bruzzone L.Classification of Hyperspectral Remote Sensing Images with Support Vector Machines锛籎锛�.IEEE Transactions on Geoscience and Remote Sensing,2004,42(8):1778-1790.顎�

锛�7锛紺amps-Valls G,Gomez-Chova L,Galpe-Maravilla J,顎奺t al顎�.Robust Support Vector Method for Hyperspectral Data Classification and Knowledge Discovery锛籎锛�.IEEE Transactions on Geoscience and Remote Sensing,2004,42(7):1530-1542.顎�

锛�8锛紻alponte M,Bruzzone L,Gianelle D.Fusion of Hyperspectral and LIDAR Remote Sensing Data for Classification of Complex Forest Areas锛籎锛�.IEEE Transactions on Geoscience and Remote Sensing,2008,46(5):1416-1427.顎�

锛�9锛紿u Gensheng,Chen Changchun,Liang Dong.Cloud Detection for Landsat Images by Combination of ACCA with WSVM锛籎锛�.Acta Geodaetica et Cartographica Sinica,2014,43(8):848-854.锛昏儭鏍圭敓,闄堥暱鏄�,姊佹爧.鑱斿悎浜戦噺鑷姩璇勪及鍜屽姞鏉冩敮鎸佸悜閲忔満鐨凩andsat褰卞儚浜戞娴嬶蓟J锛�.娴嬬粯瀛︽姤,2014,43(8):848-854.锛筋��

锛�10锛紿e Lingmin,Shen Zhangquan,Kong Fansheng,顎奺t al顎�.Study on Multi-source Remote Sensing Classification with SVM锛籎锛�.Journal of Image and Graphics,2007,12(4):648-654.锛讳綍鐏垫晱,娌堟帉娉�,瀛旂箒鑳�,绛�.SVM鍦ㄥ婧愰仴鎰熷奖鍍忓垎绫讳腑鐨勫簲鐢ㄧ爺绌讹蓟J锛�.涓浗鍥捐薄鍥惧舰瀛︽姤,2007,12(4):648-654.锛筋��

锛�11锛紽eng Xiao,Xiao Pengfeng,Li Qi,顎奺t al顎�.Hyperspectral Image Classification based on 3-D Gabor Filter and Support Vector Machines锛籎锛�.Spectrocopy and Spectral Analysis,2014,34(8):2218-2224.锛诲啹閫�,鑲栭箯宄�,鏉庣惁,绛�.涓夌淮Gabor婊ゆ尝鍣ㄤ笌鏀寔鍚戦噺鏈虹殑楂樺厜璋遍仴鎰熷浘鍍忓垎绫伙蓟J锛�.鍏夎氨瀛︿笌鍏夎氨鍒嗘瀽,2014,34(8):2218-2224.锛筋��

锛�12锛絃i Xiaorun,Zhu Jie顒媏r,Wang Jing,顎奺t al顎�.Hyperspectral Image Classification based on Compsite Kernels Support Vector Machine锛籎锛�.Journal of Zhejiang University (Engineering Science),2013,47(8):1403-1410.锛诲帀灏忔鼎,鏈辨磥灏�,鐜嬫櫠,绛�.缁勫悎鏍告敮鎸佸悜閲忔満楂樺厜璋卞浘鍍忓垎绫伙蓟J锛�.娴欐睙澶у瀛︽姤(宸ュ鐗�),2013,47(8):1403-1410.锛筋��

锛�13锛紺hen Wei,Yu Xuchu,Zhang Pengqiang,顎奺t al顎�.Object Recognition based on One-class Support Vector Machine in Hyperspectral Image锛籎锛�.Journal of Computer Applications,2011,31(8):2092-2100.锛婚檲浼�,浣欐棴鍒�,寮犻箯寮�,绛�.鍩轰簬涓�绫绘敮鎸佸悜閲忔満鐨勯珮鍏夎氨褰卞儚鍦扮墿璇嗗埆锛籎锛�.璁＄畻鏈哄簲鐢�,2011,31(8):2092-2100.锛筋��

锛�14锛紺ortes C,Vapnik V.Support-vector Networks锛籎锛�.Machine Learning,1995,20(3):273-297.顎�

锛�15锛紺ristianini顎� N,Shawe-Taylor J.An Introduction to Support Vector Machines and other Kenel-based Learning Methods锛籑锛�.Cambridge:Cambridge University Press,2000.锛�16锛絃iu Mengmeng,Liu Yalan,Sun Guoqing,顎奺t al.顎奡VM Land Cover Classification based on Spectral and Textural Features Using Seratified Samples锛籎锛�.Remote Sensing Technology and Application,2014,29(2):315-323.锛诲垬钀岃悓,鍒樹簹宀�,瀛欏浗搴�,绛�.缁撳悎绾圭悊鐗瑰緛鐨凷VM鏍锋湰鍒嗗眰鍦熷湴瑕嗙洊鍒嗙被锛籎锛�.閬ユ劅鎶�鏈笌搴旂敤,2014,29(2):315-323.锛筋��

锛�17锛絁ian Yanhuang,Zhao Qiangli.Machine Learning Techniques锛籑锛�.Beijing:Publishing House of Electronics Industry,2009.锛昏拫鑹冲嚢,璧靛己鍒�.鏈哄櫒瀛︿範鏂规硶锛籑锛�.鍖椾含:鐢靛瓙宸ヤ笟鍑虹増绀�,2009.锛筋��

锛�18锛紹ai Peng,Zhang Xi顒媌in,Zhang Bin,顎奺t al顎�.Support Vector Machines and Its Application in Mixed Gas Infrared Spectrum Analysis锛籑锛�.Xian:Xidian University Publisher,2008.锛荤櫧楣�,寮犲枩鏂�,寮犳枌,绛�.鏀寔鍚戦噺鏈虹悊璁哄強宸ョ▼搴旂敤瀹炰緥锛籑锛�.瑗垮畨:瑗垮畨鐢靛瓙绉戞妧澶у鍑虹増绀�,2008.锛筋��

锛�19锛絋ooke T R,Coops N C,Goodwin N R,顎奺t al顎�.Extracting Urban Vegetation Characteristics Using Spectral Mixture Analysis and Decision Tree Classifications锛籎锛�.Remote Sensing of Environment,2009,113(2):398-407.顎�

锛�20锛結oav顎� F,Robert E S.A Decision-theoretic Generalization of on-line Learning and an Application to Boosting锛籎锛�.Journal of Computer and System Sciences,1997,55(1):119-139.顎�

锛�21锛紽riedl M A,Brodley C E.Decision Tree Classification of Land Cover from Remotely Sensed Data锛籎锛�.Remote Sensing of Environment,1997,61(3):399-409.顎�

锛�22锛絏u顎� Ranran,Ju Haolin,Li Chaofeng.Non-balanced Binary Tree with Active Learning Support Vector Machine锛籎锛�.Microelectronics & Computer,2013,30(5):55-58.锛诲緪鍐夊唹,鐞氭槉闇�,鏉庢湞閿�.闈炲钩琛′簩鍙夋爲涓诲姩瀛︿範鏀寔鍚戦噺鏈猴蓟J锛�.寰數瀛愬涓庤绠楁満,2013,30(5):55-58.锛筋��

锛�23锛絎ang Xiaofeng,Qin Yuping.Research on SVM Multi-class Classification based on Binary Tree锛籎锛�.Journal of Hunan Institute of Engineering,2008,18(3):68-70.锛荤帇鏅撻攱,绉︾帀骞�.鍩轰簬浜屽弶鏍戠殑SVM澶氱被鍒嗙被绠楁硶鐮旂┒锛籎锛�.婀栧崡宸ョ▼瀛﹂櫌瀛︽姤,2008,18(3):68-70.锛筋��

锛�24锛紾oodenough D G,Dyk A,Niemann K O,顎奺t al顎�.Processing Hyperion and ALI for Forest Classification锛籎锛�.IEEE Transactions on Geoscience Remote Sensing,2003,41(6):1321-1331.

[1]	寮犲織鍐�,鐜嬭尮,濮氭湀,閮芥垚濡�,鐢宠寽. 鍩轰簬璧勬簮涓�鍙�02D楂樺厜璋卞崼鏄熷奖鍍忕殑闈掓捣婀栨偓娴墿娴撳害鍙嶆紨鐮旂┒[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2023, 38(5): 1159-1166.
[2]	鑳″疁濞�,瀹夊,鑹炬辰澶�,閮戒紵鍐�. 鍩轰簬鏃犱汉鏈洪珮鍏夎氨褰卞儚鐨勪笁姹熸簮鑽夌绮剧粏璇嗗埆鐮旂┒[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(4): 926-935.
[3]	娈甸噾浜�,寮犵憺,鏉庡,搴炲娉�. 涓�绉嶅熀浜庡櫔澹版按骞充及璁＄殑鎵╁睍绾挎�у厜璋卞垎瑙ｇ畻娉�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(4): 820-826.
[4]	鏉庡噣,娓╂澗妤�. 鍩轰簬3绉嶆満鍣ㄥ涔犳硶鐨勫お闃宠緪灏勬ā鎷熺爺绌�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2020, 35(3): 615-622.
[5]	鑲栨槉锛岀帇鏉邦�嬵��. 鍩轰簬IDL鍜孧ATLAB娣峰悎缂栫▼鐨勪袱绉嶅厜璋辨贩鍚堝垎鏋愭柟娉曟瘮杈�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2017, 32(5): 858-865.
[6]	鏉庢亽鍑紝娆у浆锛屽垬闆ㄥ┓锛岄偙鐜夊疂顎嬵��. 鍩轰簬娣峰悎鍍忓厓鍒嗚В鐨勯珮鍏夎氨褰卞儚鏌戞璇嗗埆鏂规硶顎�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2017, 32(4): 743-750.
[7]	椴嶈晩,澶忎繆澹�,钖涙湞杈�,鏉滃煿鍐�,杞︾編鐞�. 鍩轰簬褰㈡�佸灞炴�у墫闈㈢殑楂樺厜璋卞奖鍍忛泦鎴愬垎绫活��[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2016, 31(4): 731-738.
[8]	鑳′繆,璋惃,鍚寸珛鏂�. 涓�绉嶅熀浜庡椤瑰紡閫昏緫鍥炲綊楂樺厜璋卞奖鍍忓垎绫绘柟娉曠殑鏀硅繘[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2015, 30(1): 135-139.
[9]	鏉庤寽妤�,鑻忕孩鍐�. 鍩轰簬钀ょ伀铏畻娉曠殑楂樺厜璋遍仴鎰熸尝娈甸�夋嫨鏂规硶顎�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2014, 29(5): 761-770.
[10]	鑻忎繆鑻�. 鍩轰簬鍏夎氨鏇茬嚎灏忔尝绯绘暟鍒嗙淮娴嬪害鐨勯珮鍏夎氨褰卞儚楂樻晥鍒嗗壊鏂规硶顎�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2012, 27(4): 555-559.
[11]	瀹嬬繝鐜�,鏉庡煿鍐�,鏉ㄩ攱鏉�. 鍩轰簬澶氬厓灞�閮ㄤ簩鍊兼ā寮忕殑閬ユ劅鍥惧儚绾圭悊鎻愬彇涓庡垎绫�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2011, 26(3): 322-327.
[12]	鑻忎繆鑻�,鑸掑畞. 涓�绉嶅熀浜庨潪绾挎�у鐩婂皬娉㈡护娉㈢殑楂樺厜璋卞奖鍍忓幓鍣妧鏈爺绌�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2008, 23(4): 434-439.
[13]	寮犻,濉旇タ鐢媺鎻�?鐗逛緷鎷�,涓佸缓涓�,渚濆姏浜氭柉姹�.鍔皵楹﹂害鎻�,鐢版簮. 鏂扮枂骞叉棻鍖虹豢娲插湡澹ょ洂娓嶅寲淇℃伅鎻愬彇瀵规瘮鐮旂┒[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2008, 23(4): 398-404.

楂樺厜璋卞奖鍍忕殑BDT-SVM鍦扮墿鍒嗙被绠楁硶涓庡簲鐢��

The Object Classification Algorithm and Application for Hyperspectral Imagery based on BDT-SVM

PDF (PC)

鍙鍖�

鎽樿/Abstract

寮曠敤鏈枃

浣跨敤鏈枃

鍙傝�冩枃鐚�

鐩稿叧鏂囩珷 13

缂栬緫鎺ㄨ崘

Metrics

鏈枃璇勪环