| 融合低阶特征与全局特征的图像语义分割方法 |
| |
| 引用本文： | 董立红,李宇星,符立梅. 融合低阶特征与全局特征的图像语义分割方法[J]. 计算机工程与应用, 2021, 57(17): 210-216. DOI: 10.3778/j.issn.1002-8331.2005-0223 |
| |
| 作者姓名： | 董立红 李宇星 符立梅 |
| |
| 作者单位： | 西安科技大学 计算机科学与技术学院,西安 710600 |
| |
| 摘 要： | 目前,深度全卷积网络在图像语义分割领域已经取得了瞩目的成就,但特征图的细节信息在多次下采样过程中会大量损失,对分割精度造成影响.针对该问题设计了一个用于图像语义分割的深度全卷积网络.该网络采用编码器—解码器结构,在编码器后端引入空洞卷积以降低细节信息的损失,在解码过程中融合对应尺寸的低阶语义特征,并在解码器末端融入...
|
| 关 键 词： | 语义分割 特征融合 全卷积神经网络 空洞卷积 |
Image Semantic Segmentation by Fusion of Global and Low Order Features |
| |
| DONG Lihong,LI Yuxing,FU Limei. Image Semantic Segmentation by Fusion of Global and Low Order Features[J]. Computer Engineering and Applications, 2021, 57(17): 210-216. DOI: 10.3778/j.issn.1002-8331.2005-0223 |
| |
| Authors: | DONG Lihong LI Yuxing FU Limei |
| |
| Affiliation: | College of Computer Science and Technology, Xi’an University of Science and Technology, Xi’an 710600, China |
| |
| Abstract: | At present, deep full convolutional networks have made considerable achievements in the field of image semantic segmentation. However, with the repeated down-sampling operation, the detail information of the feature map is greatly lost, which affects the segmentation accuracy. In this paper, it designs a deep full convolutional network for semantic segmentation. The network structure consists of encoder and decoder. Some dilated convolution layers are introduced at the end of the encoder to reduce the loss of detailed information. In each stage of the decoding process, the low-order semantic features of corresponding dimensions are integrated, and the global features are integrated at the end of decoder to improve the segmentation accuracy. Finally, it uses the data-enhanced CamVid dataset to train and test the network. And the test results reach 90.14% PA and 71.95% mIoU. Experimental results show that the output of the network performs better in terms of regional smoothing. |
| |
| Keywords: | semantic segmentation feature fusion full convolutional neural network dilated convolution |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程与应用》浏览原始摘要信息 |
|
点击此处可从《计算机工程与应用》下载免费的PDF全文 |
|
