基于改进DeepLabV<sup>3+</sup>网络的卫星遥感图像林地提取

孟芳芳; 许浩; 方薇; 张冬英; 张文涛

光学技术, 2023, 49 (6): 743, 网络出版: 2023-12-05

基于改进DeepLabV³⁺网络的卫星遥感图像林地提取

Forest land extraction from satellite remote sensing images based on improved DeepLabV³⁺ network

论文大纲

孟芳芳 ^1,*许浩 ¹方薇 ²张冬英 ²张文涛 ¹

作者单位

¹ 合肥学院先进制造工程学院, 安徽合肥 230601

² 中科院合肥物质科学研究院智能机械研究所, 安徽合肥 230031

遥感图像注意力机制语义分割 remote sensing image DeepLabV3+ DeepLabV3+ Transformer transformer attention mechanism semantic segmentation

摘要

针对普通卷积神经网络在遥感图像分割中林地边界区域识别不完整、小片林地分割精度低的问题,提出一种基于transformer与注意力机制的DeeplabV3+网络改进方法。在编码阶段引入transformer机制,将原池化金字塔部分中的空洞卷积操作替换为可获取更多上下文信息的transformer操作,从而提高网络对林地边界信息的提取能力; 将注意力机制引入到网络的解码部分,提升模型对小片林地的检测能力。实验表明,采用改进后的方法平均交并比(MIou)可达到81.83%,对比原DeepLabV3+网络模型提升了1.25%。该方法充分考虑了卫星遥感图像分割中林地边缘信息的提取以及对小目标的关注度,提出的改进方法可提升遥感图像中对林地提取的精度。

Abstract

Aiming at the problems of incomplete recognition of forest land boundary area and low accuracy of small forest land segmentation in remote sensing image segmentation by ordinary convolutional neural network, an improved method of DeeplabV3+ network based on transformer and attention mechanism is passively proposed. First, the transformer mechanism is introduced in the encoding stage, and the hole convolution operation in the original pooling pyramid is replaced by a transformer operation that can obtain more context information, thereby improving the network's ability to extract forest boundary information; then, the attention mechanism is introduced. Go to the decoding part of the network to improve the model's ability to detect small forests. Experiments show that the average intersection-over-union ratio (MIou) of the improved method can reach 81.83%, which is 1.25% higher than the original DeepLabV3+ network model. The method fully considers the extraction of forest edge information and the attention to small targets in satellite remote sensing image segmentation, and the improved method proposed can improve the accuracy of forest land extraction in remote sensing images.

参考文献

[1] 刘金香, 班伟, 陈宇, 等. 融合多维度CNN的高光谱遥感图像分类算法［J］. 中国激光,2021,48(16):1610003.

[2] 李树涛, 李聪妤, 康旭东. 多源遥感图像融合发展现状与未来展望［J］. 遥感学报,2021,25(01):148-166.

[3] Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation［C］∥Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention(MICCAI). Germany: Springer, 2015:234-241.

[4] Badrinarayanan V, Kendall A, Cipolla R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(12):2481-2495.

[5] Chen L C, Papandreou G, Kokkinos I, et al. Semantic image segmentation with deep convolutional nets and fully connected CRFs［J］.Computer Science,2014,1(4):357-361.

[6] Chen L C, Papandreou G, Kokkinos I, et al. DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(4):834-848.

[7] 蒋林, 刘奇, 雷斌, 等. 激光与视觉融合识别并构建语义地图改善定位性能［J］. 中国激光,2022,49(18):140-154.

[8] Chen L C, Papandreou G, Schroff F, et al. Rethinking atrous convolution for semantic image segmentation［J/OL］. arXiv preprint arXiv:1706.05587, 2017.

[9] 云飞, 殷雁君, 张文轩, 等. 融合注意力机制的对抗式半监督语义分割［J］. 计算机工程与应用,2023,59(08):254-262.

[10] Chen L C, Zhu Y, Papandreou G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation［C］∥ European Conference on Computer Vision. Germany: Springer,2018:801818.

[11] 李娇娇, 刘志强, 宋锐, 等. 一种改进Unet网络的遥感影像分割算法［J］. 西安电子科技大学学报,2022,49(06):67-75+128.

[12] 徐长友, 樊绍胜, 朱航. 采用通道域注意力机制Deeplabv3+算法的遥感影像语义分割［J］. 控制工程,2023,30(02):368-375.

[13] 欧阳雯思. 基于UNet++网络的城市高分辨率遥感图像植被信息提取研究［D］. 广西师范大学,2022.

[14] 苏志鹏, 李景文, 姜建武, 等. 基于改进DeepLabV3+遥感影像语义分割方法［J］. 激光与光电子学进展,2023,60(06):359-366.

[15] 李昭慧, 寇鸽子. 基于改进的Deeplabv3+的红外航拍图像架空导线识别算法［J］. 红外与激光工程,2022,51(11):181-189.

[16] Alexey D, Lucas B, Alexander K, et al. An image is worth 16x16 words: transformers for image recognition at scale［EB/OL］. (2021-6-3)［2022-2-24］.https:∥arxiv.org/abs/2010.11929.

[17] Liu Z, Lin Y, Cao Y, et al. Swin transformer: hierarchical vision transformer using shifted windows［C］∥International Conference on Computer Vision (ICCV). Montreal, Canada: IEEE,2021:10012-10022.

[18] 高家军, 张旭, 郭颖, 等. 融合Swin Transformer的虫害图像实例分割优化方法研究［J］. 南京林业大学学报(自然科学版),2023,47(03):1-10.

孟芳芳, 许浩, 方薇, 张冬英, 张文涛. 基于改进DeepLabV³⁺网络的卫星遥感图像林地提取[J]. 光学技术, 2023, 49(6): 743. MENG Fangfang, XU Hao, FANG Wei, ZHANG Dongying, ZHANG Wentao. Forest land extraction from satellite remote sensing images based on improved DeepLabV³⁺ network[J]. Optical Technique, 2023, 49(6): 743.

基于改进DeepLabV³⁺网络的卫星遥感图像林地提取

关于本站 Cookie 的使用提示

全站搜索

基于改进DeepLabV3+网络的卫星遥感图像林地提取

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索

基于改进DeepLabV³⁺网络的卫星遥感图像林地提取