基于改进Unet的分布式光伏建筑物高精度分割方法

徐孝彬; 张好杰; 白建波; 裴融浩; 胡家宇; 谭治英

doi:10.19912/j.0254-0096.tynxb.2022-1155

PDF(2940 KB)

太阳能学报 ›› 2023, Vol. 44 ›› Issue (11) : 82-90. DOI: 10.19912/j.0254-0096.tynxb.2022-1155

基于改进Unet的分布式光伏建筑物高精度分割方法

徐孝彬¹, 张好杰¹, 白建波¹, 裴融浩², 胡家宇¹, 谭治英¹

作者信息 +

HIGH-PRECISION SEGMENTATION METHOD OF DISTRIBUTED PHOTOVOLTAIC BUILDINGS BASED ON IMPROVED UNET

Xu Xiaobin¹, Zhang Haojie¹, Bai Jianbo¹, Pei Ronghao², Hu Jiayu¹, Tan Zhiying¹

Author information +

文章历史 +

摘要

针对屋顶光伏资源评估中难以准确高效地获取建筑物屋顶区域的问题,该文提出一种基于Unet的FPN_AttentionUnet语义分割网络,用于实现建筑物屋顶的高精度自动提取。该网络融合Soft-Attention注意力机制和双层特征金字塔FPN以提取准确的语义信息,精细化分割结果。Soft-Attention注意力机制用于处理和连接编码部分与解码部分的特征图;双层特征金字塔FPN融合解码部分不同尺度的特征图来获取不同尺度的特征信息。采用无人机获取苏州某区域上空的建筑物数据集和武汉大学WHU公开数据集分别进行训练,训练结果表明：与Unet、AttentionUnet、FPNUnet网络相比,该文提出的FPN_AttentionUnet在建筑物外轮廓提取中具有更高的精度,有效提高边缘提取效果。在自制数据集中类别像素准确率CPA达95.56%,平均交并比MIoU达91.10%,在WHU公开数据集中分割效果同样优于其他对比网络,所提算法能够有效提升建筑物外轮廓边缘的分割精度。最后以河海大学常州校区为例,利用提出的算法从无人机图像中分割建筑物,评估指定区域的光伏发电量与光伏组件安装潜力。

Abstract

Aiming at the problem that it is difficult to obtain the roof area of buildings accurately and efficiently in the evaluation of roof photovoltaic resources, the FPN_AttentionUnet semantic segmentation network is proposed to realize high-precision automatic extraction of building roofs. The network integrates soft attention mechanism and double-layer feature pyramid FPN to extract accurate semantic information and refine segmentation results. The Soft-Attention mechanism is used to process and connect the feature map of the encoding part and the decoding part. Double-layer feature pyramid FPN fuses feature maps of different scales to obtain feature information of different scales. The unmanned aerial vehicle is used to obtain the building data set over a certain area of Suzhou and the WHU public data set of Wuhan University for training, respectively. The training results show that compared with Unet, AttentionUnet and FPNUnet networks, the proposed FPN_AttentionUnet has higher accuracy in building outer contour extraction, which effectively improves the effect of edge extraction. In the self-made dataset, the category pixel accuracy C_PA reaches 95.56%, and the average intersection and union ratio M_IoUreaches 91.10%. In the WHU public dataset, the segmentation effect is also better than other comparison networks. Finally, taking Changzhou Campus of Hohai University as an example, the proposed algorithm was used to segment buildings from UAV images to evaluate the photovoltaic power generation and photovoltaic module installation potential of the area.

导出引用

徐孝彬, 张好杰, 白建波, 裴融浩, 胡家宇, 谭治英. 基于改进Unet的分布式光伏建筑物高精度分割方法[J]. 太阳能学报. 2023, 44(11): 82-90 https://doi.org/10.19912/j.0254-0096.tynxb.2022-1155

Xu Xiaobin, Zhang Haojie, Bai Jianbo, Pei Ronghao, Hu Jiayu, Tan Zhiying. HIGH-PRECISION SEGMENTATION METHOD OF DISTRIBUTED PHOTOVOLTAIC BUILDINGS BASED ON IMPROVED UNET[J]. Acta Energiae Solaris Sinica. 2023, 44(11): 82-90 https://doi.org/10.19912/j.0254-0096.tynxb.2022-1155

中图分类号： TK513.5

参考文献

[1] 黄震, 谢晓敏. 碳中和愿景下的能源变革[J]. 中国科学院院刊, 2021, 36(9): 1010-1018.
HUANG Z, XIE X M.Energy revolution under vision of carbon neutrality[J]. Bulletin of Chinese Academy of Sciences, 2021, 36(9): 1010-1018.
[2] 李十中. 推动新能源革命促进实现碳中和目标[J]. 人民论坛·学术前沿, 2021(14): 42-51.
LI S Z.Promoting the new energy revolution and achieving the goal of carbon neutrality[J]. Frontiers, 2021(14): 42-51.
[3] 金秋实, 王晓, 倪依琳, 等. “双碳”背景下光伏行业发展研究与展望[J]. 环境保护, 2022, 50(S1): 44-50.
JIN Q S, WANG X, NI Y L, et al.Development research and outlook on photovoltaic industry under carbon peaking and carbon neutrality goals[J]. Environmental protection, 2022, 50(S1): 44-50.
[4] ZHOU X P, YANG J K, YUAN X D, et al.Solar potential for the solar photovoltaic roof integration system in China explored by the geographic information system[J]. International journal of global energy issues, 2009, 31(1): 50.
[5] 刘光旭, 吴文祥, 张绪教, 等. 屋顶可用太阳能资源评估研究: 以2000年江苏省数据为例[J]. 长江流域资源与环境, 2010, 19(11): 1242-1248.
LIU G X, WU W X, ZHANG X J, et al.Study for evaluating roof-mounted available solar energy resource: case in Jiangsu Province according to its 2000 data[J]. Resources and environment in the Yangtze Basin, 2010, 19(11): 1242-1248.
[6] SUN Y W, HOF A, WANG R, et al.GIS-based approach for potential analysis of solar PV generation at the regional scale: a case study of Fujian Province[J]. Energy policy, 2013, 58: 248-259.
[7] 邱喜兰, 范宏武, 徐强, 等. 上海市分布式光伏发电发展规划研究[J]. 上海节能, 2014(10): 11-15.
QIU X L, FAN H W, XU Q, et al.Study on the development planning of distributed photovoltaic power generation in Shanghai[J]. Shanghai energy conservation, 2014(10): 11-15.
[8] 郭晓琳. 基于屋顶面积的徐州市屋顶太阳能光伏潜力评估[D]. 徐州: 中国矿业大学, 2015.
GUO X L.Rooftop solar PV potential assessment of Xuzhou based on roof area[D]. Xuzhou: China University of Mining and Technology, 2015.
[9] 张华. 城市建筑屋顶光伏利用潜力评估研究[D]. 天津: 天津大学, 2017.
ZHANG H.Research on PV energy potential of rooftop in urban area[D]. Tianjin: Tianjin University, 2017.
[10] 徐辉, 祝玉华, 甄彤, 等. 深度神经网络图像语义分割方法综述[J]. 计算机科学与探索, 2021, 15(1): 47-59.
XU H, ZHU Y H, ZHEN T, et al.Survey of image semantic segmentation methods based on deep neural network[J]. Journal of frontiers of computer science and technology, 2021, 15(1): 47-59.
[11] 张鑫, 姚庆安, 赵健, 等. 全卷积神经网络图像语义分割方法综述[J]. 计算机工程与应用, 2022, 58(8): 45-57.
ZHANG X, YAO Q A, ZHAO J, et al.Image semantic segmentation based on fully convolutional neural network[J]. Computer engineering and applications, 2022, 58(8): 45-57.
[12] SHELHAMER E, LONG J, DARRELL T.Fully convolutional networks for semantic segmentation[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence. 2016: 640-651.
[13] RONNEBERGER O, FISCHER P, BROX T.U-net: convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2015: 234-241.
[14] 何直蒙, 丁海勇, 安炳琪. 高分辨率遥感影像建筑物提取的空洞卷积E-Unet算法[J]. 测绘学报, 2022, 51(3): 457-467.
HE Z M, DING H Y, AN B Q.E-Unet: a atrous convolution-based neural network for building extraction from high-resolution remote sensing images[J]. Acta geodaetica et cartographica sinica, 2022, 51(3): 457-467.
[15] CHEN Z Y, LI D L, FAN W T, et al.Self-attention in reconstruction bias U-net for semantic segmentation of building rooftops in optical remote sensing images[J]. Remote sensing, 2021, 13(13): 2524.
[16] 秦梦宇, 刘勇, 张寅丹, 等. 基于改进U-Net模型的高分辨率遥感影像中城市建筑物的提取[J]. 兰州大学学报(自然科学版), 2022, 58(2): 254-261, 269.
QIN M Y, LIU Y, ZHANG Y D, et al.Extraction of urban buildings from high-resolution remote sensing images based on improved U-Net model[J]. Journal of Lanzhou University (natural sciences), 2022, 58(2): 254-261, 269.
[17] DENG W J, SHI Q, LI J.Attention-gate-based encoder-decoder network for automatical building extraction[J]. IEEE journal of selected topics in applied earth observations and remote sensing, 2021, 14: 2611-2620.
[18] ALSABHAN W, ALOTAIBY T.Automatic building extraction on satellite images using unet and ResNet50[J]. Computational intelligence and neuroscience, 2022, 2022: 5008854.
[19] DELIBAŞOĞLU İ. INCSA-UNET: spatial attention inception UNET for aerial images segmentation[J]. Computing and informatics, 2021, 40(6): 1244-1262.
[20] YE H R, LIU S, JIN K, et al.CT-UNet: an improved neural network based on U-net for building segmentation in remote sensing images[C]//2020 25th International Conference on Pattern Recognition (ICPR). Milan, Italy, 2021: 166-172.
[21] SCHLEMPER J, OKTAY O, SCHAAP M, et al.Attention gated networks: learning to leverage salient regions in medical images[J]. Medical image analysis, 2019, 53: 197-207.
[22] YU M Y, CHEN X X, ZHANG W Z, et al.AGs-unet: building extraction model for high resolution remote sensing images based on attention gates U network[J]. Sensors, 2022, 22(8): 2932.
[23] SUN X Y, XIAO Y, JI Y F, et al.Multi scale UNet encoder-decoder network for building extraction[C]//2021 3rd International Conference on Information Technology and Computer Communications. Guangzhou, China, 2021.
[24] FENG D J, XIE Y K, XIONG S F, et al.Regularized building boundary extraction from remote sensing imagery based on augment feature pyramid network and morphological constraint[J]. IEEE journal of selected topics in applied earth observations and remote sensing, 2021, 14: 12212-12223.
[25] 季顺平, 魏世清. 遥感影像建筑物提取的卷积神经元网络与开源数据集方法[J]. 测绘学报, 2019, 48(4): 448-459.
JI S P, WEI S Q.Building extraction via convolutional neural networks from an open remote sensing building dataset[J]. Acta geodaetica et cartographica sinica, 2019, 48(4): 448-459.
[26] 于文玲, 刘波, 刘华, 等. 基于Attention Gates和R2U-Net的遥感影像建筑物提取方法[J]. 地理与地理信息科学, 2022, 38(3): 31-36, 42.
YU W L, LIU B, LIU H, et al.Building extraction from remote sensing images based on the R2U-Net model and attention gates[J]. Geography and geo-information science, 2022, 38(3): 31-36, 42.
[27] 李传林, 黄风华, 胡威, 等. 基于Res_AttentionUnet的高分辨率遥感影像建筑物提取方法[J]. 地球信息科学学报, 2021, 23(12): 2232-2243.
LI C L, HUANG F H, HU W, et al.Building extraction from high-resolution remote sensing image based on Res_AttentionUnet[J]. Journal of geo-information science, 2021, 23(12): 2232-2243.
[28] NIU Z Y, ZHONG G Q, YU H.A review on the attention mechanism of deep learning[J]. Neurocomputing, 2021, 452: 48-62.
[29] HU J, SHEN L, SUN G.Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA, 2018: 7132-7141.
[30] LI X, WANG W H, HU X L, et al.Selective kernel networks[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach, CA, USA, 2020: 510-519.
[31] LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA, 2017: 936-944.
[32] ZHAO B J, ZHAO B Y, TANG L B, et al.Multi-scale object detection by top-down and bottom-up feature pyramid network[J]. Journal of systems engineering and electronics, 2019, 30(1): 1-12.
[33] 赵斐, 张文凯, 闫志远, 等. 基于多特征图金字塔融合深度网络的遥感图像语义分割[J]. 电子与信息学报, 2019, 41(10): 2525-2531.
ZHAO F, ZHANG W K, YAN Z Y, et al.Multi-feature map pyramid fusion deep network for semantic segmentation on remote sensing data[J]. Journal of electronics & information technology, 2019, 41(10): 2525-2531.
[34] 崔卫红, 熊宝玉, 张丽瑶. 多尺度全卷积神经网络建筑物提取[J]. 测绘学报, 2019, 48(5): 597-608.
CUI W H, XIONG B Y, ZHANG L Y.Multi-scale fully convolutional neural network for building extraction[J]. Acta geodaetica et cartographica sinica, 2019, 48(5): 597-608.
[35] TIAN Q L, ZHAO Y J, LI Y, et al.Multiscale building extraction with refined attention pyramid networks[J]. IEEE geoscience and remote sensing letters, 2022, 19: 1-5.
[36] DONG X, LI F, BAI H, et al.Dual attention based image pyramid network for object detection[J]. KSII transactions on internet and information systems, 2021, 15(12): 4439-4455.