基于卷积神经网络的遥感沙漠绿地提取方法

doi:10.11873/j.issn.1004-0323.2018.1.0151

閬ユ劅鎶�鏈笌搴旂敤

鈥� 閬ユ劅搴旂敤鈥� 涓婁竴绡� 涓嬩竴绡�

鍩轰簬鍗风Н绁炵粡缃戠粶鐨勯仴鎰熸矙婕犵豢鍦版彁鍙栨柟娉曨��

鐢板痉瀹�^1锛�2锛屽紶鑰�鍗�^1锛�3 锛岃档鍥借緣^{1锛�2锛�3}锛岄煩绔嬮挦^{1锛�2顎�}

锛�1.涓浗绉戝闄㈣タ鍖楃敓鎬佺幆澧冭祫婧愮爺绌堕櫌锛岀敇鑲� 鍏板窞 730000锛涱��
2.涓浗绉戝闄㈠ぇ瀛︼紝鍖椾含 100049锛涱��
3.涓浗绉戝闄㈣秴绾ц绠楀叞宸炲垎涓績锛岀敇鑲� 鍏板窞 730000锛�

鏀剁鏃ユ湡:2016-12-17 鍑虹増鏃ユ湡:2018-02-20 鍙戝竷鏃ユ湡:2018-03-16
浣滆�呯畝浠�:鐢板痉瀹�(1993-)锛岀敺锛屽唴钂欏彜鍥涘瓙鐜嬫棗浜猴紝纭曞＋鐮旂┒鐢燂紝涓昏浠庝簨閬ユ劅鍥惧儚淇℃伅鎻愬彇銆佸瘨鍖哄湴璐ㄧ伨瀹宠瘎浠风爺绌躲�侲-mail锛歵iandy@lzb.ac.cn銆�
鍩洪噾璧勫姪:
鍥藉鑷劧绉戝閲嶇偣鍩洪噾椤圭洰锛�91125005/D011004锛夛紝涓浗绉戝闄俊鎭寲閲嶇偣椤圭洰锛圛NFO-115-D01锛夎祫鍔┿��

Convolutional Neural Network for Remote顎僑ensing Plant Cover Extracting

Tian Deyu^1锛�2锛孼hang Yaonan^1锛�3锛孼hao Guohui^{1锛�2锛�3}锛孒an Liqin^{1锛�2顎�}

锛�1.Northwest Institute of Eco-Environment and Resources锛孋hinese Academy of Sciences锛岊�僉anzhou 730000锛孋hina锛涱��
2.University of Chinese Academy of Sciences锛孊eijing 100049锛孋hina锛涱��
3.Lanzhou Supercomputing Center of Chinese Academy of Sciences锛孡anzhou 730000锛孋hina锛�

Received:2016-12-17 Online:2018-02-20 Published:2018-03-16

鎽樿/Abstract

鎽樿锛�

鏈�鍏堣繘鐨�(state-of-the-art)鏈哄櫒瀛︿範閬ユ劅淇℃伅鎻愬彇鏂规硶寰�寰�閫氳繃鍥惧儚鐨勬尝娈电粍鍚堛�佺汗鐞嗗垎鏋愭瀯寤虹壒寰佸悜閲忥紝浣嗘槸杩欑鏂规硶鍙�夌殑鐗瑰緛鏈夐檺涓旈渶瑕佽繃澶氫汉涓哄共棰勩�傞�氳繃寤虹珛鍗风Н绁炵粡缃戠粶鑷姩鑾峰彇澶氭尝娈甸仴鎰熷浘鍍忔繁灞傛鐨勭壒寰佽繘琛屽簱甯冮綈娌欐紶涓豢鍦版彁鍙栧疄楠屻�傝缁冨垎绫诲櫒骞惰繘琛岃秴鍙傛暟閫夋嫨锛岄�氳繃浜ゅ弶楠岃瘉鍜屽姣斿垎鏋愭潵妫�楠屾ā鍨嬬殑鎬ц兘銆傚疄楠岃〃鏄庯細寤虹珛鐨勬ā鍨嬮娴嬬簿搴﹂珮锛屾硾鍖栬兘鍔涘己锛屼负缁垮湴浠ュ強鏇村姞澶嶆潅鐨勫湴鐗╀俊鎭彁鍙栧紑杈熸柊鐨勬�濊矾銆�

鍏抽敭璇�: 鍗风Н绁炵粡缃戠粶, 鐗瑰緛鍚戦噺, 澶氭尝娈甸仴鎰�, 淇℃伅鎸栨帢, 搴撳竷榻愭矙婕�

Abstract:

The key point of the state-of-the-art machine learning method to extract land information is to construct the features-vector.The existing methods mainly use the spectral features锛宼exture features of remote sensing images to construct the features-vector锛宧owever锛宼his method can only get limited features and requires too much human intervention.In the face of the above problems锛宼his paper builds a convolutional neural network model for mining the deep-level features of multi-band remote sensing images and then extract the greenbelt in the Kubuqi Desert.The model was trained and hyperparameter selection was performed.The performance of the model was evaluated by cross validation and comparative analysis between methods.The experimental results show that the model is of high accuracy and good generalization ability.Finally锛宼he test data set was input into the model to predict land cover classes and to do visualization.The importance of this study is to inspire new thinking of better performance of the green land and even more complex information extraction from remote sensing images.

Key words: Convolutional Neural Network, Feature Vector, Multi-band remote sensing, Information mining, Kubuqi desert

涓浘鍒嗙被鍙�:

TP 79

鐢板痉瀹囷紝寮犺��鍗楋紝璧靛浗杈夛紝闊╃珛閽�. 鍩轰簬鍗风Н绁炵粡缃戠粶鐨勯仴鎰熸矙婕犵豢鍦版彁鍙栨柟娉曨�僛J]. 閬ユ劅鎶�鏈笌搴旂敤, DOI: 10.11873/j.issn.1004-0323.2018.1.0151.

Tian Deyu锛孼hang Yaonan锛孼hao Guohui锛孒an Liqin . Convolutional Neural Network for Remote顎僑ensing Plant Cover Extracting[J]. Remote Sensing Technology and Application, DOI: 10.11873/j.issn.1004-0323.2018.1.0151.

[1]	瀛欑淮缁�, 鍒樻澃, 寮犺姵鑺�, 椹捣鑹�, 鐜嬫槍鏄�, 娼樿搐绔�. 鍩轰簬楂樺垎杈ㄧ巼閬ユ劅褰卞儚搴旂敤BASS-Net鏋勫缓鍖栧伐鍥尯鍏稿瀷鍦扮墿璇嗗埆妯″瀷[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2024, 39(3): 612-619.
[2]	寮犲濞�,璋竴鏁�,钁ｈ悕,瀛熷湥鍗�,鍙告捣骞�,鐜嬫檽骞�,椹柊鏄�. 鍩轰簬娣卞害瀛︿範鐨勯仴鎰熷奖鍍忔椂绌鸿瀺鍚堝強鍐皬楹︾妞嶉潰绉彁鍙�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2023, 38(3): 578-587.
[3]	瀹嬩匠,寰愭収绐�,楂樺皯鍗�,椹櫒鐕�,璇镐簯寮�. 杞婚噺鍖栧嵎绉缁忕綉缁滈仴鎰熷奖鍍忓缓绛戠墿鎻愬彇妯″瀷[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2023, 38(1): 190-199.
[4]	鏈卞崥,閽熸柟娲�,璧靛啗閿�. 鍩轰簬鍒嗗舰缁存暟鍜岃浜岄樁鐭╄緟鍔╃殑鍗风Н绁炵粡缃戠粶浜戦洩璇嗗埆鐮旂┒[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2022, 37(6): 1328-1338.
[5]	鍚曞啲姊�,椹帴,鏉庡崕鏈�. 鍩轰簬CNN鐨勫悏鏋椾竴鍙峰崼鏄熷煄甯傚湡鍦拌琚埗鍥炬綔鍔涜瘎浼�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2022, 37(2): 368-378.
[6]	浠樻兜,鑼冩箻娑�,涓ョ弽鐝�,鏉滃皬骞�. 鍩轰簬娣卞害瀛︿範鐨勯仴鎰熷浘鍍忕洰鏍囨娴嬫妧鏈爺绌惰繘灞�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2022, 37(2): 290-305.
[7]	閮戠,浣曠洿钂�,涓佹捣鍕�. 鍩轰簬ENVINet5鐨勯珮鍒嗚鲸鐜囬仴鎰熷奖鍍忕█鐤忓鏂欏ぇ妫氭彁鍙栫爺绌�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(4): 908-915.
[8]	灞堢個,琚佸崰鑹�,璧垫枃鏅�,闄堝娉�,闄堝闃�. 鍩轰簬澶氭椂搴忕壒寰佸拰鍗风Н绁炵粡缃戠粶鐨勫啘浣滅墿鍒嗙被[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(2): 304-313.
[9]	闄堝Ξ,搴斾赴,鐜嬮潤,鏉庡仴. 鍩轰簬U-Net鐨勯珮鍒嗚鲸鐜囬仴鎰熷浘鍍忓湡鍦板埄鐢ㄤ俊鎭彁鍙�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(2): 285-292.
[10]	鑳′箖鍕�,闄堟稕,鐢勫,鐗涚憺鍗�. 鍩轰簬鍗风Н绁炵粡缃戠粶鐨勯潰鍚戝璞￠湶澶╅噰鍦烘彁鍙�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(2): 265-274.
[11]	闄堟晱,娼樹匠濞�,鏉庢睙鏉�,寰愮拹,鍒樺姞鏁�,闊╁仴,闄堝浜�. 缁撳悎VGGNet涓嶮ask R-CNN鐨勯珮鍒嗚鲸鐜囬仴鎰熷奖鍍忓缓璁剧敤鍦版娴�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(2): 256-264.
[12]	鍒橀拪,璧垫,寤栨枑鍑�. 鍩轰簬鍗风Н绁炵粡缃戠粶棰勬祴缁撴灉鐨勭紳闅欎慨澶嶇畻娉曠爺绌�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2021, 36(2): 247-255.
[13]	寮犲嘲鏋�,鍚磋壋鍏�,濮氶洩涓�,姊佹辰姣�. 鍩轰簬鏀硅繘DenseNet缃戠粶鐨勫婧愰仴鎰熷奖鍍忛湶澶╁紑閲囧尯鏅鸿兘鎻愬彇鏂规硶[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2020, 35(3): 673-684.
[14]	閮戝畻鐢�,鑳℃櫒闆�,榛勫啲姊�,閭瑰浗鑹�,鍒樺厗鑽�,瀹嬪穽. 鍩轰簬杩佺Щ瀛︿範鍙婃皵璞″崼鏄熶簯鍥剧殑鍙伴绛夌骇鍒嗙被鐮旂┒[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2020, 35(1): 202-210.
[15]	宕斿垰,鍚撮噾鑳�,浜庨晣,鍛ㄧ幉. 娣卞害瀛︿範瀵逛笉鍚屽垎杈ㄧ巼褰卞儚鍐皬楹﹁瘑鍒殑閫傜敤鎬х爺绌�[J]. 閬ユ劅鎶�鏈笌搴旂敤, 2019, 34(4): 748-755.

鍩轰簬鍗风Н绁炵粡缃戠粶鐨勯仴鎰熸矙婕犵豢鍦版彁鍙栨柟娉曨��

Convolutional Neural Network for Remote顎僑ensing Plant Cover Extracting

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 15

缂栬緫鎺ㄨ崘

Metrics

鏈枃璇勪环