计算机工程与应用2019，Vol.55Issue(24)：91-95,5.DOI:10.3778/j.issn.1002-8331.1904-0155
空洞卷积的多尺度语义分割网络
Multiscale Semantic Segmentation Network Based on Cavity Convolution
摘要
Abstract
The development of computer hardware has greatly promoted the development of computer vision. Convolu-tion neural network has made remarkable achievements in semantic segmentation. However, the stacking of multiple con-volutional layers inevitably result in the loss of detailed information in the boundary of objects. In order to preserve bound-ary information as far as possible and improve the accuracy of image segmentation, a multiscale atrous convolution neural network model is proposed. The proposed model utilizes multiscale pooling to adapt to different scale targets in images. Besides, atrous convolution layer is used to learn target features, thus the accuracy of detailed information is improved, better segmentation results are obtained. Experimental results on the ISPRS Vaihingen dataset show that the proposed mul-tiscale atrous convolution neural network is effective for target boundary fitting.关键词
深度学习/语义分割/空洞卷积/多尺度Key words
deep learning/semantic segmentation/cavity convolution/multiscale分类
信息技术与安全科学引用本文复制引用
曲长波,姜思瑶,吴德阳..空洞卷积的多尺度语义分割网络[J].计算机工程与应用,2019,55(24):91-95,5.基金项目
国家自然科学基金(No.71771111). （No.71771111）
