本文選題：深度學(xué)習(xí)　切入點(diǎn)：地理國情普查　出處：《地球信息科學(xué)學(xué)報(bào)》2017年11期

【摘要】：地表覆蓋分類信息是反映自然、人工地表覆蓋要素的綜合體,包含植被、土壤、冰川、河流、湖泊、沼澤濕地及各類人工構(gòu)筑物等元素,側(cè)重描述地球表面的自然屬性,具有明確的時(shí)間及空間特性。地表覆蓋分類信息數(shù)據(jù)量大、現(xiàn)勢性強(qiáng)、人工評(píng)價(jià)費(fèi)時(shí),其自動(dòng)化評(píng)價(jià)長期以來存在許多技術(shù)難點(diǎn)。本文基于面向?qū)ο蟮膱D斑分類體系,引入深度卷積神經(jīng)網(wǎng)絡(luò)對(duì)現(xiàn)有地理國情普查-地表覆蓋分類數(shù)據(jù)進(jìn)行分類評(píng)價(jià),并通過試驗(yàn)利用AlexNet模型實(shí)現(xiàn)地表覆蓋分類評(píng)價(jià)驗(yàn)證。試驗(yàn)結(jié)果表明,該方法可有效判讀耕地、房屋2類圖斑,正確分類隸屬度優(yōu)于99%,而由于數(shù)據(jù)較少、訓(xùn)練不充分,林地、水體圖斑正確分類隸屬度不高,分別為62.73%和43.59%。使用本文方法,經(jīng)過大量數(shù)據(jù)充分微調(diào)的深度學(xué)習(xí)AlexNet可有效地計(jì)算圖斑的地類隸屬度,并實(shí)現(xiàn)自動(dòng)地表覆蓋分類圖斑量化評(píng)價(jià)。
[Abstract]:The classification information of surface cover is a complex that reflects the elements of nature and artificial surface cover, including vegetation, soil, glacier, river, lake, swamp wetland and all kinds of artificial structures, etc., which mainly describes the natural properties of the earth's surface.Has definite time and space characteristic.There are many technical difficulties in automatic evaluation of land cover classification, such as large amount of data, strong present situation and time-consuming manual evaluation.Based on the object oriented classification system of map spot, this paper introduces the deep convolution neural network to classify and evaluate the existing geographic situation census and land cover classification data, and realizes the ground cover classification evaluation and verification by using AlexNet model.The experimental results show that this method can effectively distinguish cultivated land, house two types of spots, the correct classification and membership degree is better than 99m, but due to less data, insufficient training, forest land and water spots, the correct classification and membership degree are 62.73% and 43.59%, respectively.By using the method of this paper, the subclass membership degree of map spot can be calculated effectively by depth learning AlexNet, which is fully fine-tuned by a large amount of data, and the quantification evaluation of surface cover classification map can be realized.
【作者單位】： 中國礦業(yè)大學(xué);國家測繪產(chǎn)品質(zhì)量檢驗(yàn)測試中心;中國測繪科學(xué)研究院;
【基金】：國家自然科學(xué)基金項(xiàng)目(41671440)
【分類號(hào)】：P237

【相似文獻(xiàn)】

相關(guān)期刊論文 前1條

1 李紅;宋尚萍;王廣福;顧劍華;;基于地理國情普查數(shù)據(jù)的綜合統(tǒng)計(jì)分析研究——以試點(diǎn)區(qū)域北安市農(nóng)業(yè)(耕地)變化統(tǒng)計(jì)分析為例[J];測繪與空間地理信息;2014年06期

，

本文編號(hào)：1717103