顾及局部特性的CNN在遥感影像分类的应用-计算机工程与应用.PDF

188 2018 ，54（7 ） Computer Engineering and Applications 计算机工程与应用 顾及局部特性的CNN 在遥感影像分类的应用 用 杨嘉树，梅天灿，仲思东 应 YANG Jiashu, MEI Tiancan, ZHONG Sidong 与 武汉大学 电子信息学院，武汉 430072程 School of Electronic Information, Wuhan University, Wuhan 430072, China g 工 r o YANG Jiashu, MEI Tiancan, ZHONG Sidong. Application of Convolution Neural Netowrk using region information . j to remote sensing image classification. Computer Engineering and Applications, 2018, 54 （7 ）：188-195. 机 a 算 e Abstract ：Deep learning shows excellent performance in image recognition and natural language processing. This paper c . presents a learning method which is based on Convolution Neural Network （CNN ）to classify remote sensing images. 计 Aiming at the problem of low efficiency caused by too many sampling windows in remote sensing image classification, this w paper proposes a novel method based on region to improve the classification efficiency. There are two steps in the classifi- w cation process. The first step of the proposed algorithm is to classify the input images by convolution neural networks. w Then, SVM is used to classify the results of convolution neural networks. The proposed algorithm is evaluated by being compared with the results of other existing methods. Experimental results show that proposed algorithm can achieve higher classification accuracy. Key words ：Convolution Neural Networks（CNN ）; image segmentation; image classificatio