基于改进Segformer的光伏组件分割算法

曹冠华; 党建武; 杨景玉

doi:10.19912/j.0254-0096.tynxb.2023-1563

PDF(5202 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (2) : 425-432. DOI: 10.19912/j.0254-0096.tynxb.2023-1563

基于改进Segformer的光伏组件分割算法

曹冠华¹, 党建武^1,2, 杨景玉¹

作者信息 +

PHOTOVOLTAIC MODULE SEGMENTATION ALGORITHM BASED ON IMPROVED SEGFORMER

Cao Guanhua¹, Dang Jianwu^1,2, Yang Jingyu¹

Author information +

文章历史 +

摘要

针对光伏图像中光伏组件与复杂背景的边界特征模糊、难以区分的问题,提出一种基于Segformer的改进算法。在编码阶段,引入全新设计的block5模块用来解析更高级的语义信息;在解码阶段,引入LPCB（边界强化）模型用来识别更精确的边界特征,最终得到精确的分割结果。实验结果表明：改进后的光伏组件分割算法在光伏组件图像数据集上的分割效果优于经典的主流分割算法,可以对光伏组件区域进行有效分割。

Abstract

Addressing the issue of blurred and indistinguishable boundary features between Photovoltaic modules and complex backgrounds in photovoltaic images, an improved algorithm based on Segformer is introduced. During the encoding phase, a novel "block5" module is incorporated to extract higher-level semantic information. In the decoding phase, an LPCB （Learning to Predict Crisp Boundaries） module is introduced to identify more precise boundary features, resulting in highly accurate segmentation. Experimental results demonstrate that the improved photovoltaic module segmentation algorithm outperforms classical mainstream segmentation methods on a dataset of photovoltaic module images, effectively segmenting the photovoltaic module regions .

导出引用

曹冠华, 党建武, 杨景玉. 基于改进Segformer的光伏组件分割算法[J]. 太阳能学报. 2025, 46(2): 425-432 https://doi.org/10.19912/j.0254-0096.tynxb.2023-1563

Cao Guanhua, Dang Jianwu, Yang Jingyu. PHOTOVOLTAIC MODULE SEGMENTATION ALGORITHM BASED ON IMPROVED SEGFORMER[J]. Acta Energiae Solaris Sinica. 2025, 46(2): 425-432 https://doi.org/10.19912/j.0254-0096.tynxb.2023-1563

中图分类号： TP391.41

参考文献

[1] 童光毅. 关于当代能源转型方向的探讨[J]. 智慧电力, 2018, 46(10): 1-3, 25.
TONG G Y.Probe into modern energy structure transition[J]. Smart power, 2018, 46(10): 1-3, 25.
[2] 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.Photovoltaic infrared hot spot image segmentation method based on gray clustering[J]. Acta energiae solaris sinica, 2023, 44(9): 117-124.
[3] 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 photoelectric panels image segmentation based on gray level co-occurrence matrix and region growing algorithm[J]. Science technology and engineering, 2018, 18(34): 92-97.
[4] 陈文勤, 郝慧杰, 肖建, 等. 一种高精度的光伏阵列红外图像分割算法[J]. 计算机技术与发展, 2020, 30(11): 153-157.
CHEN W Q, HAO H J, XIAO J, et al.A high-precision photovoltaic array infrared image segmentation algorithm[J]. Computer technology and development, 2020, 30(11): 153-157.
[5] 周树芳, 张小咏, 陈正超, 等. 面向光伏电站识别的深度实例分割方法[J]. 福州大学学报(自然科学版), 2022, 50(4): 497-504.
ZHOU S F, ZHANG X Y, CHEN Z C, et al.Deep instance segmentation method for photovoltaic power station recognition[J]. Journal of Fuzhou University (natural science edition), 2022, 50(4): 497-504.
[6] 李青, 李海涛, 李辉, 等. 注意力机制和全局卷积在光伏板分割的应用[J]. 计算机工程与应用, 2024, 60(04): 237-248.
LI Q, LI H T, LI H, et al.Photovoltaic panel segmentation using an attention mechanism and global convolute-on[J]. Computer engineering and applications, 2024, 60(4): 237-248.
[7] WANG J X, CHEN X, JIANG W C, et al.PVNet: a novel semantic segmentation model for extracting high-quality photovoltaic panels in large-scale systems from high-resolution remote sensing imagery[J]. International journal of applied earth observation and geoinformation, 2023, 119: 103309.
[8] ZHU R, GUO D X, WONG M S, et al.Deep solar PV refiner: a detail-oriented deep learning network for refined segmentation of photovoltaic areas from satellite imagery[J]. International journal of applied earth observation and geoinformation, 2023, 116: 103134.
[9] 任喜伟, 韩欣, 钟弋, 等. 基于改进U-Net网络的光伏板图像分割方法[J]. 陕西科技大学学报, 2023, 41(2): 155-161.
REN X W, HAN X, ZHONG Y, et al.Photovoltaic panel image segmentation method based on improved U-Net network[J]. Journal of Shaanxi University of Science & Technology, 2023, 41(2): 155-161.
[10] 刘桂生, 丁鑫, 祝锐, 等. 结合HDC和Attention的高分遥感影像光伏板提取研究[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 extracting photovoltaic panels from high resolution remote sensing images by combining HDC and Attention[J]. Computer engineering and applications, 2024, 60(14): 357-366.
[11] 肖运启, 张美玲, 郭治昌, 等. 山地光伏电站固定式组件安装角度优化方法[J]. 太阳能学报, 2020, 41(5): 329-335.
XIAO Y Q, ZHANG M L, GUO Z C, et al.Optimum design for tilt and orientation angle of fixed pv module of PV power station on mountainous area[J]. Acta energiae solaris sinica, 2020, 41(5): 329-335.
[12] DENG R X, SHEN C H, LIU S J, et al.Learning to predict crisp boundaries[M]. Cham: Springer International Publishing, 2018: 570-586.
[13] XIE E Z, WANG W H, YU Z D, et al.SegFormer: Simple and efficient design for semantic segmentation with transformers[J]. Advances in neural information processing systems, 2021, 34: 12077-12090.
[14] 刘文婷, 卢新明. 基于计算机视觉的Transformer研究进展[J]. 计算机工程与应用, 2022, 58(6): 1-16.
LIU W T, LU X M.Research progress of Transformer based on computer vision[J]. Computer engineering and applications, 2022, 58(6): 1-16.
[15] WANG X T, XIE L B, DONG C, et al.Real-ESRGAN: training real-world blind super-resolution with pure synthetic data[C]//2021 IEEE/CVF International Conference on Computer Vision Workshops(ICCVW). Montreal, BC, Canada, 2021: 1905-1914.
[16] 李洪安, 郑峭雪, 陶若霖等. 基于深度学习的图像超分辨率研究综述[J]. 图学学报, 2023, 44(01): 1-15.
LI H A, ZHENG Q X, TAO R L, et al.Review of image super-resolution based on deep learning[J]. Journal of graphics, 2023, 44(1): 1-15.
[17] TUBA E, BAČANIN N, STRUMBERGER I, et al. Convolutional neural networks hyperparameters tuning[M]. Cham: Springer International Publishing, 2021: 65-84.
[18] GU W W, TANDON A, AHN Y Y, et al.Principled approach to the selection of the embedding dimension of networks[J]. Nature communications, 2021, 12(1): 3772.
[19] VASWANI A, SHAZEER N, PARMAR N, et al.Attention is all you need[J]. Advances in neural information processing systems, 2017, 30.
[20] SANDLER M, HOWARD A, ZHU M L, et al.MobileNetV2: inverted residuals and linear bottlenecks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA, 2018: 4510-4520.
[21] AZAD R, ASADI-AGHBOLAGHI M, FATHY M, et al.Attention Deeplabv3+: multi-level context attention mechanism for skin lesion segmentation[M]. Cham: Springer International Publishing, 2020: 251-266.