Investigation of Generated Near-infrared Crack Defect Dataset for Solar Cells

  • Zuo Chen
  • , Wende Liu*
  • , Taotao Zhang*
  • , Yadong Chen
  • , Haiyong Gan*
  • , Limin Xiong
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The generation of near infrared electroluminescence (EL) crack defect images of polycrystalline silicon (poly-Si) solar cells is investigated. For complex environments (such as background object interference) and weak target signals (such as low pixel ratio), object detection is generally difficult, as is the case for the direct generation of such images. A typical example is the crack defect detection on poly-Si solar cells. The irregular complex environment may cover the defects and cause the generation model to learn irrelevant background and noise information, thus affecting the generation of defect images. Based on StyleGAN model and image processing methods, a two-stage “generation + fusion” simulation method is proposed to generate the EL crack defect images for poly-Si solar cells. The similarity between simulated and real images and the effectiveness of the simulated dataset for small sample learning are studied. The method can also be used for other similar industrial defect detection and data enhancement research.

Original languageEnglish
Title of host publicationConference on Spectral Technology and Applications, CSTA 2024
EditorsZhe Wang, Hongbin Ding
PublisherSPIE
ISBN (Electronic)9781510683082
DOIs
StatePublished - 2024
Event2024 Conference on Spectral Technology and Applications, CSTA 2024 - Dalian, China
Duration: 9 May 202411 May 2024

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume13283
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference2024 Conference on Spectral Technology and Applications, CSTA 2024
Country/TerritoryChina
CityDalian
Period9/05/2411/05/24

Keywords

  • Crack defect
  • Electroluminescence Imaging
  • Fréchet Inception Distance
  • Generative Adversarial Network
  • Object detection
  • Solar Cells

Fingerprint

Dive into the research topics of 'Investigation of Generated Near-infrared Crack Defect Dataset for Solar Cells'. Together they form a unique fingerprint.

Cite this