The accurately and efficiently extracting rural settlements from high resolution remote sensing image is of importantsignificance for rural government management. Due to the complex environment in rural region, the traditionalsupervised classification methods already could not satisfy the application requirements for automatically extractingrural settlements, and they can only obtain the results of low precision and incomplete extraction. In recent years, withthe rapid development of deep learning in computer vision, the deep learning method has been widely used to targetextraction based on high resolution remote sensing imagery. So, this paper proposed a rural settlements extractionmethod based on the deep learning using high-resolution remote sensing image. The Tensorflow deep learningframework was built up to train the Faster regional recommendation convolutional neural network model(FasterR-CNN). Image feature maps were extracted by the Convolutional Neural Network(CNN) firstly. The region proposalnetwork (RPN) was built to extract the regions that might contain rural settlements. And the region was identified andclassified by detection network. The method was tested and verified in the homemade datasets. This paper selected atypical area for testing. The experimental results show that the proposed method can extract the rural settlements areaswith higher accuracy compared with traditional rural extraction ways.
展开▼
