DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers

Solar energy is one of the most important resources that can be a clean and renewable alternative to traditional fuels. The collection process of solar energy mainly rely on the photovoltaic solar cells. The defects, such as microcracks and finger interruption on the photovoltaic solar cells can red...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xiangying Xie, Hu Liu, Zhixiong Na, Xin Luo, Dong Wang, Biao Leng
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Solar energy defects detection transformer multi-head self-attention multi-scale aggregation Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/4ecfd9af6b304e6384aadea431014f7b
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4ecfd9af6b304e6384aadea431014f7b
record_format	dspace
spelling	oai:doaj.org-article:4ecfd9af6b304e6384aadea431014f7b2021-11-24T00:03:05ZDPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers2169-353610.1109/ACCESS.2021.3119631https://doaj.org/article/4ecfd9af6b304e6384aadea431014f7b2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9568905/https://doaj.org/toc/2169-3536Solar energy is one of the most important resources that can be a clean and renewable alternative to traditional fuels. The collection process of solar energy mainly rely on the photovoltaic solar cells. The defects, such as microcracks and finger interruption on the photovoltaic solar cells can reduce its efficiency a lot. To solve this problem, defects detection of solar cells have attracted attention from many researchers. In this paper, we propose a novel transformer based network to detect defects on solar cells efficiently and effectively. First, we introduce convolutions into the transformer to enable positional information and spatial context more accurate and precise. Secondly, cross window based multi-head self-attention (CW-MSA) is proposed to enlarge the window relation modeling capacity. Finally, we propose a multi-scale aggregation block to merge the low-level features into deep semantically strong features by attention. Extensive experiments on the elpv dataset demonstrate DPiT can consistently bring significant improvements over its strong baseline Swin Transformer with subtle extra computational overhead. The visualization results show that the proposed DPiT is able to detect various complex defects correctly. In particular, DPiT can achieve impressive 91.7 top-1 accuracy and greatly outperforms other competitive counterparts.Xiangying XieHu LiuZhixiong NaXin LuoDong WangBiao LengIEEEarticleSolar energydefects detectiontransformermulti-head self-attentionmulti-scale aggregationElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 154292-154303 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Solar energy defects detection transformer multi-head self-attention multi-scale aggregation Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Solar energy defects detection transformer multi-head self-attention multi-scale aggregation Electrical engineering. Electronics. Nuclear engineering TK1-9971 Xiangying Xie Hu Liu Zhixiong Na Xin Luo Dong Wang Biao Leng DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers
description	Solar energy is one of the most important resources that can be a clean and renewable alternative to traditional fuels. The collection process of solar energy mainly rely on the photovoltaic solar cells. The defects, such as microcracks and finger interruption on the photovoltaic solar cells can reduce its efficiency a lot. To solve this problem, defects detection of solar cells have attracted attention from many researchers. In this paper, we propose a novel transformer based network to detect defects on solar cells efficiently and effectively. First, we introduce convolutions into the transformer to enable positional information and spatial context more accurate and precise. Secondly, cross window based multi-head self-attention (CW-MSA) is proposed to enlarge the window relation modeling capacity. Finally, we propose a multi-scale aggregation block to merge the low-level features into deep semantically strong features by attention. Extensive experiments on the elpv dataset demonstrate DPiT can consistently bring significant improvements over its strong baseline Swin Transformer with subtle extra computational overhead. The visualization results show that the proposed DPiT is able to detect various complex defects correctly. In particular, DPiT can achieve impressive 91.7 top-1 accuracy and greatly outperforms other competitive counterparts.
format	article
author	Xiangying Xie Hu Liu Zhixiong Na Xin Luo Dong Wang Biao Leng
author_facet	Xiangying Xie Hu Liu Zhixiong Na Xin Luo Dong Wang Biao Leng
author_sort	Xiangying Xie
title	DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers
title_short	DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers
title_full	DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers
title_fullStr	DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers
title_full_unstemmed	DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers
title_sort	dpit: detecting defects of photovoltaic solar cells with image transformers
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/4ecfd9af6b304e6384aadea431014f7b
work_keys_str_mv	AT xiangyingxie dpitdetectingdefectsofphotovoltaicsolarcellswithimagetransformers AT huliu dpitdetectingdefectsofphotovoltaicsolarcellswithimagetransformers AT zhixiongna dpitdetectingdefectsofphotovoltaicsolarcellswithimagetransformers AT xinluo dpitdetectingdefectsofphotovoltaicsolarcellswithimagetransformers AT dongwang dpitdetectingdefectsofphotovoltaicsolarcellswithimagetransformers AT biaoleng dpitdetectingdefectsofphotovoltaicsolarcellswithimagetransformers
_version_	1718416091533803520

DPiT: Detecting Defects of Photovoltaic Solar Cells With Image Transformers

Ejemplares similares