基于深度残差注意力网络的光伏组件分割

李鹏; 宁昊; 宿雲龙; 孟庆伟; 陈继明

doi:10.19912/j.0254-0096.tynxb.2024-1510

PDF(3055 KB)

太阳能学报 ›› 2026, Vol. 47 ›› Issue (1) : 72-81. DOI: 10.19912/j.0254-0096.tynxb.2024-1510

基于深度残差注意力网络的光伏组件分割

李鹏, 宁昊, 宿雲龙, 孟庆伟, 陈继明

作者信息 +

PHOTOVOLTAIC MODULE SEGMENTATION BASED ON DEEP RESIDUAL ATTENTION NETWORK

Li Peng, Ning Hao, Su Yunlong, Meng Qingwei, Chen Jiming

Author information +

文章历史 +

摘要

针对遥感图像中光伏组件的分割和提取问题,提出一种基于深度残差注意力网络的遥感图像光伏组件语义分割方法。首先基于U-Net结构搭建光伏组件遥感图像分割框架;然后,使用深度残差神经网络替换原始U-Net的特征提取部分,提升网络的图像特征提取和表达能力;最后,在网络的残差模块中引入一种高效局部注意力机制,用于进一步增强局部特征的表达能力,提高算法对光伏组件的分割和提取精度。利用该算法在遥感图像光伏组件公开数据集上进行分割提取实验,结果表明改进算法在3类不同空间分辨率的数据集上表现优于DeepLabv3+、UCTransNet、UDTransNet、HRNetV2、SegFormer等方法,相较于原始U-Net网络的mIoU、Dice、mPA和Precision等评价指标平均提升5.80%、2.91%、3.06%和3.92%。

Abstract

To address the challenge of segmenting photovoltaic （PV） modules in remote sensing images, this paper proposes a semantic segmentation method based on a deep residual attention network. The method builds on the U-Net architecture by integrating a deep residual network for enhanced feature extraction and representation. Additionally, a local attention mechanism is incorporated within the residual modules to further refine local feature expression, improving segmentation accuracy. Experimental results on a public remote sensing PV dataset demonstrate that the proposed method consistently outperforms DeepLabv3+, UCTransNet, and UDTransNet across multiple spatial resolutions, achieving average improvements of 5.80%, 2.91%, 3.06% and 3.92% in mIoU, Dice, mPA, and Precision, respectively, over the original U-Net.

导出引用

李鹏, 宁昊, 宿雲龙, 孟庆伟, 陈继明. 基于深度残差注意力网络的光伏组件分割[J]. 太阳能学报. 2026, 47(1): 72-81 https://doi.org/10.19912/j.0254-0096.tynxb.2024-1510

Li Peng, Ning Hao, Su Yunlong, Meng Qingwei, Chen Jiming. PHOTOVOLTAIC MODULE SEGMENTATION BASED ON DEEP RESIDUAL ATTENTION NETWORK[J]. Acta Energiae Solaris Sinica. 2026, 47(1): 72-81 https://doi.org/10.19912/j.0254-0096.tynxb.2024-1510

中图分类号： TP751

参考文献

[1] 杨静丽. 发展“阳光”产业推动能源转型[N]. 中国石化报, 2024-07-22(008). YANG J L. Developing the “Sunshine” industry to promote energy transition[N]. China Petrochemical News, 2024-07-22(008).
[2] HANSEN K, BREYER C, LUND H.Status and perspectives on 100% renewable energy systems[J]. Energy, 2019, 175: 471-480.
[3] 刘桂生, 丁鑫, 祝锐, 等.结合HDC和Attention的高分遥感影像光伏组件提取研究[J]. 计算机工程与应用, 2024, 60(14): 357-366.LIU G S, DING X, ZHU R, et al. Research on photovoltaic panel extraction from high-resolution remote sensing images combining HDC and Attention[J]. Computer engineering and applications, 2024, 60(14): 357-366.
[4] 洪向共, 周世芬. 基于灰度共生矩阵和区域生长算法的红外光伏面板图像分割[J]. 科学技术与工程, 2018, 18(34): 92 97.HONG X G, ZHOU S F. Infrared photovoltaic panel image segmentation based on gray level co-occurrence matrix and region growing algorithm[J]. Science technology and engineering, 2018, 18(34): 92-97.
[5] KHAN Z, YANG J.Nonparametric K-means clustering-based adaptive unsupervised colour image segmentation[J]. Pattern analysis and applications, 2024, 27(1): 17.
[6] EALI S N J, BHATTACHARYYA D, NAKKA T R, et al. A novel approach in bio-medical image segmentation for analyzing brain cancer images with U-NET semantic segmentation and TPLD models using SVM[J]. Traitement du signal, 2022, 39(2): 419-430.
[7] 翁飘, 陆彦辉, 齐宪标, 等. 基于改进的全卷积神经网络的路面裂缝分割技术[J]. 计算机工程与应用, 2019, 55(16): 235-239, 245.WENG P, LU Y H, QI X B, et al. Pavement crack segmentation technology based on improved fully convolutional neural network[J]. Computer engineering and applications, 2019, 55(16): 235-239, 245.
[8] RONNEBERGER O, FISCHER P, BROX T.U-Net: convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015. Cham: Springer International Publishing, 2015: 234-241.
[9] CHEN L C, PAPANDREOU G, KOKKINOS I, et al.Semantic image segmentation with deep convolutional nets and fully connected CRFs[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2017, 40(8): 834-848.
[10] SIMONYAN K, ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2014-09.
[11] 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.
[12] CHEN L C, PAOANDREOU G, SCHROFF F, et al.Rethinking atrous convolution for semantic image segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2017-06.
[13] CHEN L C, ZHU Y K, PAPANDREOU G, et al.Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Computer Vision-ECCV 2018. Cham: Springer, 2018: 833-851.
[14] WANG H N, CAO P, WANG J Q, et al.UCTransNet: rethinking the skip connections in U-Net from a channel-wise perspective with transformer[J]. Proceedings of the AAAI conference on artificial intelligence, 2022, 36(3): 2441-2449.
[15] WANG H N, CAO P, YANG J Z, et al.Narrowing the semantic gaps in U-Net with learnable skip connections: the case of medical image segmentation[J]. Neural networks, 2024, 178: 106546.
[16] MANAKITSA N, MARASLIDIS G S, MOYSIS L, et al.A review of machine learning and deep learning for object detection, semantic segmentation, and human action recognition in machine and robotic vision[J]. Technologies, 2024, 12(2): 15.
[17] WINSTON D P, MURUGAN M S, ELAVARASAN R M, et al.Solar PV’s micro crack and hotspots detection technique using NN and SVM[J]. IEEE access, 2021, 9: 127259-127269.
[18] 谢七月, 孙武彪, 申忠利, 等. 基于灰度聚类的光伏红外热斑图像分割方法[J]. 太阳能学报, 2023, 44(9): 117-124. XIE Q Y, SUN W B, SHEN Z L, et al. Infrared Hotspot image segmentation method for photovoltaic systems based on grayscale clustering[J]. Acta energiae solaris sinica, 2023, 44(9): 117-124.
[19] 孙海蓉, 李莉, 周映杰, 等. 基于特征金字塔融合高分辨率网络的光伏热斑识别[J]. 太阳能学报, 2023, 44(9): 109-116.SUN H R, LI L, ZHOU Y J, et al. Photovoltaic Hotspot detection based on feature pyramid and high-resolution network[J]. Acta energiae solaris sinica, 2023, 44(9): 109-116.
[20] 孙海蓉, 李帆. 基于注意力机制的光伏热斑识别[J]. 太阳能学报, 2023, 44(2): 453-459. SUN H R, LI F. Photovoltaic hot spot recognition based on attention mechanism[J]. Acta energiae solaris sinica, 2023, 44(2): 453-459.
[21] 王银, 沈灵鑫, 李茂环, 等.融合视觉特征的光伏组件语义分割模型研究[J]. 太阳能学报, 2024, 45(4): 500-511.WANG Y, SHEN L X, LI M H, et al. Research on semantic segmentation model of photovoltaic modules integrating visual features[J]. Acta energiae solaris sinica, 2024, 45(4): 500-511.
[22] 沈灵鑫, 王银, 李杰, 等. 融合边缘特征的DeepLabv3+光伏面板语义分割模型研究[J]. 控制工程, 2025, 32(4): 707-719.SHEN L X, WANG Y, LI J, et al. Research on deepLabv3+semantic segmentation model for photovoltaic panels integrating edge features[J]. control engineering, 2025, 32(4): 707-719.
[23] CASTELLO R, ROQUETTE S, ESGUERRA M, et al.Deep learning in the built environment: automatic detection of rooftop solar panels using convolutional neural networks[J]. Journal of physics: conference series, 2019, 1343(1): 012034.
[24] ZHUANG L, ZHANG Z J, WANG L.The automatic segmentation of residential solar panels based on satellite images: a cross learning driven U-Net method[J]. Applied soft computing, 2020, 92: 106283.
[25] CAMILO J, WANG R, COLLINS L M, et al. Application of a semantic segmentation convolutional neural network for accurate automatic detection and mapping of solar photovoltaic arrays in aerial imagery[J]. arXiv preprint arXiv:1801.04018, 2018.
[26] HE K M, ZHANG X Y, REN S Q, et al.Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA, 2016.
[27] 张文悦, 贾子彦, 李青, 等. 基于ResNet和UNet的胃癌病理图像诊断系统[J]. 河北大学学报(自然科学版),2024, 44(4): 433-440.ZHANG W Y, JIA Z Y, LI Q, et al. Gastric cancer pathological image diagnosis system based on ResNet and UNet[J]. Journal of Hebei University(natural science edition), 2024, 44(4): 433-440.
[28] 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. Cham, Switzerland: Springer, 2015: 234-241.
[29] 张明星, 徐健, 刘秀平, 等. 改进U-Net的路面裂缝检测方法[J]. 计算机工程与应用, 2024, 60(24): 306-313.ZHANG M X, XU J, LIU X P, et al. Improved U-Net method for pavement crack detection[J]. Computer engineering and applications, 2024,60(24):306-313.
[30] YANG Z Y, XIONG F, PEI Y Y, et al.A stay cable icing identification method based on the fusion of U-Net and ResNet50[J]. Applied sciences, 2024, 14(7): 3062.
[31] JIANG H, YAO L, LU N, et al.Multi-resolution dataset for photovoltaic panel segmentation from satellite and aerial imagery[J]. Earth system science data, 2021, 13(11): 5389-5401.