膨胀卷积与金字塔表达的深度学习模型用于农村建筑物提取

doi:10.13474/j.cnki.11-2246.2022.0111

测绘通报 ›› 2022, Vol. 0 ›› Issue (4): 61-65.doi: 10.13474/j.cnki.11-2246.2022.0111

膨胀卷积与金字塔表达的深度学习模型用于农村建筑物提取

王雪¹, 梁珂², 隋立春³, 钟棉卿⁴, 朱剑锋³

1. 咸阳师范学院, 陕西咸阳 712000;
2. 自然资源部第一大地测量队, 陕西西安 710054;
3. 长安大学地质工程与测绘学院, 陕西西安 710054;
4. 陕西省地理空间信息工程技术研究中心, 陕西西安 710199

收稿日期:2021-05-10 出版日期:2022-04-25 发布日期:2022-04-26
作者简介:王雪(1983-),女,博士生,讲师,主要研究方向为遥感图像处理、激光雷达数据处理。E-mail:178475464@qq.com
基金资助:
陕西省科技厅青年项目（2021JQ-819）；咸阳师范学院科研基金项目（XSYK20013）

Rural buildings extraction based on deep learning model with dilated convolution and pyramid representation

WANG Xue¹, LIANG Ke², SUI Lichun³, ZHONG Mianqing⁴, ZHU Jianfeng³

1. Xianyang Normal University, Xianyang 712000, China;
2. The First Geodetic Surveying Brigade of Ministry of Natural Resources, Xi'an 710054, China;
3. School of Geological Engineering and Geomatics, Chang'an University, Xi'an 710054, China;
4. Shaanxi Geospatial Information Engineering Technology Research Center, Xi'an 710199, China

Received:2021-05-10 Online:2022-04-25 Published:2022-04-26

摘要/Abstract

摘要： 由于农村建筑物结构多样、空间分布复杂等特征，自动提取面临较多困难。针对该问题，本文提出采用膨胀卷积和金字塔池化表达的神经网络模型用于遥感影像中农村建筑物自动提取。在膨胀卷积神经网络模块中，通过改变孔尺寸的大小，获取不同感受野的特征信息；在金字塔表达方面，每个模块输入不同尺度的信息，且同时下采样的倍率也不同，获取多维的金字塔尺度特征；最终将提取的浅层及深层尺度特征信息进行融合，构建一个改进的适用于农村建筑物目标自动提取的深度学习模型。试验结果表明，与FCN-8s和DeepLab模型提取的结果相比，本文方法在农村建筑物提取中表现较好的性能，提取精度明显提高，且更好保留了目标边界细节信息，减少了噪声。

关键词: 深度学习, 膨胀卷积, 金字塔表达, 农村建筑物提取, 遥感影像

Abstract: Automatic extraction rural buildings faces many difficulties because of the diverse structure and complex spatial distribution, etc. Aiming at this problem, this paper proposes a neural network model with dilated convolution and pyramid representation for automatic extracting rural buildings in remote sensing images. In the dilated convolution neural network module, the feature information with different receptive fields is obtained by changing the size of the dilated hole. In the pyramid representation, each module inputs different scale information, and the rate of down sampling is also different, which obtains multi-dimensional pyramid scale features. Finally, the model is fused shallow feature and deep feature to construct improved deep learning model for rural building automatic extraction. Compared with FCN-8s and DeepLab models, the experiment results show that this method performs better in the extraction of rural buildings. The accuracy of the extraction is obviously improved, the boundary details are better retention and the noise is little.

Key words: deep learning, dilated convolution, pyramid representation, rural buildings extraction, remote sensing images

中图分类号:

P237

王雪, 梁珂, 隋立春, 钟棉卿, 朱剑锋. 膨胀卷积与金字塔表达的深度学习模型用于农村建筑物提取[J]. 测绘通报, 2022, 0(4): 61-65.

WANG Xue, LIANG Ke, SUI Lichun, ZHONG Mianqing, ZHU Jianfeng. Rural buildings extraction based on deep learning model with dilated convolution and pyramid representation[J]. Bulletin of Surveying and Mapping, 2022, 0(4): 61-65.

参考文献 15

[1]	贺晓璐, 刘振华, 胡月明. 基于面向对象的建筑物信息提取方法研究[J]. 河南理工大学学报(自然科学版), 2020, 39(2):53-61.
[2]	陶超, 谭毅华, 蔡华杰, 等. 面向对象的高分辨率遥感影像城区建筑物分级提取方法[J]. 测绘学报, 2010, 39(1):39-45.
[3]	DRĂGUŢ L, CSILLIK O, EISANK C, et al. Automated parameterisation for multi-scale image segmentation on multiple layers[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 88:119-127.
[4]	JUNG C R, SCHRAMM R. Rectangle detection based on a windowed Hough transform[C]//Proceedings of the 17th Brazilian Symposium on Computer Graphics and Image Processing. Curitiba:IEEE, 2004:113-120.
[5]	崔有祯, 吴露露, 辛星, 等. 基于改进Harris算法的高分辨率遥感影像建筑物角点检测研究[J]. 测绘通报, 2013(9):24-26.
[6]	赵翠晓, 陈曦, 杨辽, 等. 建筑物边界约束的航空影像DSM优化[J]. 遥感信息, 2016, 31(4):16-21.
[7]	SPORTOUCHE H, TUPIN F, DENISE L. Building extraction and 3D reconstruction in urban areas from high-resolution optical and SAR imagery[C]//Proceedings of 2009 Joint Urban Remote Sensing Event. Shanghai:IEEE, 2009:1-11.
[8]	王宏涛, 王变利. 融合LiDAR和多光谱影像的建筑物自动检测[J]. 测绘科学, 2015, 40(8):64-67.
[9]	程国旗, 张继贤, 李阳春, 等. 高分影像建筑物阴影优化提取与高度估算[J]. 测绘科学, 2020, 45(8):103-109.
[10]	SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651.
[11]	BITTNER K, CUI S, REINARTZ P. Building extraction from remote sensing data using fully convolutional networks[J]. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, XLII-1/W1:481-486.
[12]	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.
[13]	SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2021-04-20]. https://arxiv.org/abs/1409.1556.
[14]	HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[C]//Proceedings of 2014 IEEE Transactions on Pattern Analysis and Machine Intelligence.[S.l.]:IEEE,2014:1904-1916.
[15]	YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[EB/OL].[2021-04-20].https://arxiv.org/pdf/1511.07122.pdf.

膨胀卷积与金字塔表达的深度学习模型用于农村建筑物提取

Rural buildings extraction based on deep learning model with dilated convolution and pyramid representation

可视化

摘要/Abstract

引用本文

使用本文

参考文献 15

相关文章 15

编辑推荐

Metrics

本文评价

[1]	刘立, 董先敏, 刘娟, 文学虎. 人机融合智能的遥感解译生产新方法[J]. 测绘通报, 2022, 0(7): 118-123,137.
[2]	蔡柔丹. 一种基于用户异步轨迹的身份识别智能方法[J]. 测绘通报, 2022, 0(7): 158-162,167.
[3]	滕文鑫, 张建辰, 邵杰. 集成时空邻近与卷积网络车道级高精度定位算法[J]. 测绘通报, 2022, 0(6): 1-5,61.
[4]	李茂林, 闫庆武, 仲晓雅. 一种复合型校正植被信息的方法[J]. 测绘通报, 2022, 0(6): 49-54142.
[5]	张龚泉, 岳建平, 刘胜男, 王海龙, 杨智翔, 周航宇. 基于遥感影像的鄱阳湖地形提取方法[J]. 测绘通报, 2022, 0(5): 62-66,73.
[6]	张晋赫, 秦育罗, 张在岩, 宋伟东, 朱洪波. 复杂场景下农村道路裂缝分割方法[J]. 测绘通报, 2022, 0(5): 74-78,88.
[7]	潘建平, 付占宝, 程卫华, 陆世安, 李鑫, 孙博文, 黄桂萍. 基于地表类型的山地海绵城市生态建设评估[J]. 测绘通报, 2022, 0(5): 84-88.
[8]	陈智朗, 付振华, 朱紫阳, 王慧慧, 刘沁雯, 杨钰灵, 许耿然. 基于HRNet的高分辨率遥感影像建筑物变化信息提取[J]. 测绘通报, 2022, 0(5): 126-132.
[9]	刘小玉, 刘扬, 杜明义, 张敏, 贾竞珏, 杨恒. 基于DeeplabV3+的建筑垃圾堆放点识别[J]. 测绘通报, 2022, 0(4): 16-19,43.
[10]	师芸, 石龙龙, 牛敏杰, 赵侃. 基于单阶段实例分割网络的黄土滑坡多任务自动识别[J]. 测绘通报, 2022, 0(4): 26-31.
[11]	高鸣, 周鑫鑫, 刘琦, 杨光迪, 吴长彬. 深度学习网络支持下的农房侵占耕地自动化监测[J]. 测绘通报, 2022, 0(3): 47-53.
[12]	王胜利, 朱寿红, 蒋毅. 基于CNN模型迁移的OLI影像光伏电池板场景识别[J]. 测绘通报, 2022, 0(2): 5-9.
[13]	孙琴琴, 蔡国印, 杨柳忠, 张宁, 杜明义. 城市尺度典型地表要素综合提取方法研究[J]. 测绘通报, 2022, 0(2): 62-66.
[14]	张自强, 刘涛, 杜萍, 锁旭宏, 杨国林. 图残差神经网络支持下的建筑物群组模式分类[J]. 测绘通报, 2022, 0(1): 1-7.
[15]	车子杰, 高飞, 吴兆福, 李振轩. 基于改进U-Net网络的多源遥感影像洪涝灾害信息提取与变化分析[J]. 测绘通报, 2022, 0(1): 26-32.