Correg-YOLOv3: a method for dense buildings detection in high-resolution remote sensing images

doi:10.11947/j.AGCS.2022.20210177

Abstract

Abstract: The exploration of building detection plays an important role in urban planning, smart city and military. Aiming at the problem of high overlapping ratio of detection frames for dense building detection in high-resolution remote sensing images, we present an effective YOLOv3 framework, corner regression-based YOLOv3 (Correg-YOLOv3), to localize dense building accurately. This improved YOLOv3 algorithm establishes a vertex regression mechanism and an additional loss item about building vertex offsets relative to the center point of bounding box. By extending output dimensions, the trained model is able to output the rectangular bounding boxes and the building vertices meanwhile. Finally, we evaluate the performance of the Correg-YOLOv3 on our self-produced data set and provide a comparative analysis qualitatively and quantitatively. The experimental results achieve high performance in precision (96.45%), recall rate (95.75%), F₁ score (96.10%) and average precision (98.05%), which were 2.73%, 5.4%, 4.1% and 4.73% higher than that of YOLOv3. Therefore, the proposed algorithm effectively tackles the problem of dense building detection in high-resolution images.

Key words: high resolution remote sensing image, Correg-YOLOv3, corner regression, dense buildings, object detection

CLC Number:

P237

CHEN Zhanlong, LI Shuangjiang, XU Yongyang, XU Daozhu, MA Chao, ZHAO Junli. Correg-YOLOv3: a method for dense buildings detection in high-resolution remote sensing images[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(12): 2531-2540.

References

[1] CHENG Gong, HAN Junwei.A survey on object detection in optical remote sensing images[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 117:11-28.
[2] 邓睿哲,陈启浩,陈奇,等.遥感影像船舶检测的特征金字塔网络建模方法[J].测绘学报,2020,49(6):787-797. DOI:10.11947/j.AGCS.2020.20190117. DENG Ruizhe, CHEN Qihao, CHEN Qi, et al. A deformable feature pyramid network for ship detection from remote sensing images[J].Acta Geodaetica et Cartographica Sinica,2020,49(6):787-797. DOI:10.11947/j.AGCS.2020.20190117.
[3] 张涛,丁乐乐,史芙蓉.高分辨率遥感影像城中村提取的景观语义指数方法[J].测绘学报,2021,50(1):97-104. DOI:10.11947/j.AGCS.2021.20190463. ZHANG Tao, DING Lele, SHI Furong. Urban villages extraction from high-resolution remote sensing imagery based on landscape semantic metrics[J].Acta Geodaetica et Cartographica Sinica, 2021, 50(1):97-104. DOI:10.11947/j.AGCS.2021.20190463.
[4] 张勇.遥感与GIS在厦门市违章建筑监测与管理中的应用[J].测绘与空间地理信息,2016,39(2):75-77. ZHANG Yong.Application of remote sensing and GIS in monitoring and management of illegal construction Xiamen city[J].Geomatics & Spatial Information Technology, 2016, 39(2):75-77.
[5] 孙长奎,刘善磊,王圣尧,等.浅谈无人机遥感技术在智慧城市建设中的应用[J].国土资源遥感,2018,30(4):8-12. SUN Changkui, LIU Shanlei, WANG Shengyao, et al. Application of UAV in construction of smart city[J]. Remote Sensing for Land & Resources, 2018, 30(4):8-12.
[6] 刘扬,付征叶,郑逢斌.高分辨率遥感影像目标分类与识别研究进展[J].地球信息科学学报,2015,17(9):1080-1091. LIU Yang, FU Zhengye, ZHENG Fengbin.Review on high-resolution remote sensing image classification and recognition[J].Journal of Geo-Information Science, 2015, 17(9):1080-1091.
[7] LIN Jingbo, JING Weipeng, SONG Houbing, et al. ESFNet:efficient network for building extraction from high-resolution aerial images[J]. IEEE Access, 2019(7):54285-54294.
[8] ZOU Weitao, JING Weipeng, CHEN Guangsheng, et al. A survey of big data analytics for smart forestry[J]. IEEE Access, 2019, 7:46621-46636.
[9] 冯丽英.基于深度学习技术的高分辨率遥感影像建设用地信息提取研究[D]. 杭州:浙江大学, 2017. FENG Liying.Research on construction land information extraction from high resolution images with deep learning technology[D]. Hangzhou:Zhejiang University, 2017.
[10] 马长辉,黄登山.纹理与几何特征信息在高空间分辨率遥感影像分类中的应用[J].测绘地理信息,2019,44(6):66-70, 92. MA Changhui, HUANG Dengshan.Application of texture features and geometric feature information in high spatial resolution remote sensing image classification[J].Journal of Geomatics, 2019, 44(6):66-70, 92.
[11] HU Lei, ZHENG Jin, GAO Feng. A building extraction method using shadow in high-resolution multispectral images[C]//Proceedings of 2011 IEEE International Geoscience and Remote Sensing Symposium. Las Vegas, NV, USA:IEEE, 2011:1862-1865.
[12] 施文灶,毛政元.基于图割与阴影邻接关系的高分辨率遥感影像建筑物提取方法[J].电子学报,2016,44(12):2849-2854. SHI Wenzao, MAO Zhengyuan.Building extraction from high-resolution remotely sensed imagery based on shadows and graph-cut segmentation[J]. Acta ElectronicaSinica, 2016, 44(12):2849-2854.
[13] KONSTANTINIDIS D, STATHAKI T, ARGYRIOU V, et al. Building detection using enhanced HOG-LBP features and region refinement processes[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016, 10(3):888-905.
[14] 吕凤华,舒宁,龚龑,等.利用多特征进行航空影像建筑物提取[J].武汉大学学报(信息科学版),2017,42(5):656-660. LV Fenghua, SHU Ning, GONG Cun, et al. Regular building extraction from high-resolution image based on multilevel-features[J].Geomatics and Information Science of Wuhan University, 2017, 42(5):656-660.
[15] CHAUDHURI D, KUSHWAHA N K, SAMAL A, et al. Automatic building detection from high-resolution satellite images based on morphology and internal gray variance[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 9(5):1767-1779.
[16] AWRANGJEB M, ZHANG Chunsun, FRASER C S. Improved building detection using texture information[J].International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2011, 38:143-148.
[17] 赵传,张保明,陈小卫,等.一种基于LiDAR点云的建筑物提取方法[J].测绘通报,2017(2):35-39. ZHAO Chuan, ZHANG Baoming, CHEN Xiaowei, et al.A method of extracting building based on LiDAR point clouds[J].Bulletin of Surveying and Mapping, 2017(2):35-39.
[18] GUO Haonan, SHI Qian, DU Bo, et al. Scene-driven multitask parallel attention network for building extraction in high-resolution remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(5):4287-4306.
[19] LI Zhenshi, ZHANG Xueliang, XIAO Pengfeng, et al. On the effectiveness of weakly supervised semantic segmentation for building extraction from high-resolution remote sensing imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14:3266-3281.
[20] 胡舒,王树根,王越,等.基于Mask R-CNN的高分遥感影像建筑物目标检测研究[J/OL].测绘地理信息.[2021-04-01].https://doi.org/10.14188/j.2095-6045.2020416. HU Shu, WANG Shugen, WANG Yue, et al.Building object detection in high-resolution remote sensing image based on mask R-CNN[J/OL].Journal of Geomatics.[2021-04-01]. https://doi.org/10.14188/j.2095-6045.2020416.
[21] GIRSHICK R. Fast R-CNN[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. New York, USA:IEEE, 2015:1440-1448.
[22] REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2016, 39(6):1137-1149.
[23] HE Kaiming, GKIOXARI G, DOLLAR P, et al. Mask r-cnn[C]//Proceedings of 2017 IEEE International Conference on Computer Vision. New York, USA:IEEE, 2017:2961-2969.
[24] 史文旭,鲍佳慧,姚宇.基于深度学习的遥感图像目标检测与识别[J].计算机应用,2020,40(12):3558-3562. SHI Wenxu, BAO Jiahui, YAO Yu. Remote sensing image target detection and identification based on deep learning[J]. Journal of Computer Applications, 2020,40(12):3558-3562.
[25] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified, real-time object detection[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. New York, USA:IEEE, 2016:779-788.
[26] LIU Wei, ANGUELOY D, ERHAN D, et al. Ssd:single shot multibox detector[C]//Proceedings of 2016 European Conference on Computer Vision. Cham, Switzerland:Springer, 2016:21-37.
[27] 董彪,熊风光,韩燮,等.基于改进Yolo v3算法的遥感建筑物检测研究[J].计算机工程与应用,2020,56(18):209-213. DONG Biao, XIONG Fengguang, HAN Xie, et al. Research on remote sensing building detection based on improved Yolo v3 algorithm[J]. Computer Engineering and Applications,2020,56(18):209-213.
[28] 李响,苏娟,杨龙.基于改进YOLOv3的合成孔径雷达图像中建筑物检测算法[J].兵工学报,2020,41(7):1347-1359. LI Xiang, SU Juan, YANG Long. A SAR image building detection algorithm based on improved YOLOv3[J]. Acta Armamentarii, 2020,41(7):1347-1359.
[29] MA Haojie, LIU Yalan, REN Yuhuan, et al. Detection of collapsed buildings in post-earthquake remote sensing images based on the improved YOLOv3[J]. Remote Sensing, 2020, 12(1):44.
[30] LI Qingpeng, WANG Yunhong, LIU Qingjie, et al. Hough transform guided deep feature extraction for dense building detection in remote sensing images[C]//Proceedings of 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). New York, USA:IEEE, 2018:1872-1876.
[31] JIANG Kaiyu, LI Qingpeng. TQR-Net:Tighter quadrangle-based convolutional neural network for dense building instance localization in remote sensing imagery[C]//Proceedings of 2019 International Conference on Image and Graphics. Cham, Switzerland:Springer, 2019:281-291.
[32] WANG Kun, LIU Maozhen, YE Zhaojun. An advanced YOLOv3 method for small-scale road object detection[J]. Applied Soft Computing, 2021:107846.
[33] IOFFE S, SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//Proceedings of 2015 International conference on machine learning. Calgary Canada:PMLR, 2015:448-456.
[34] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE conference on computer vision and pattern recognition. Las Vegas, NV, USA:IEEE, 2016:770-778.