Near-infrared defect detection method for photovoltaic panels based on the FBCT-YOLOv8 algorithm
Zhongxing Zhang, Shaotong Pei, Chenlong Hu, Yunpeng Liu, Jie Liu, Guiting Na, Pengda Wang, Jie Mei, Chang Liu, Peizhe Zeng, Lianghui Li · 2026
Read Near-infrared defect detection method for photovoltaic panels based on the FBCT-YOLOv8 algorithm by Zhongxing Zhang, Shaotong Pei, Chenlong Hu, Yunpeng Liu, Jie Liu, Guiting Na, Pengda Wang, Jie Mei, Chang Liu, Peizhe Zeng, Lianghui Li (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13500561.txt.
Description
Zhongxing Zhang and colleagues upgrade in 2026 an object detection algorithm for finding defects in solar panels from near-infrared images. Panel faults vary wildly in scale and shape, and tiny defects are easy to miss, yet they sap power output and shorten panel life. Their method builds on the YOLOv8 detector, adding dynamic convolution that adapts to each input and a lightweight module that fuses features across scales. An attention-based detection head and a reshaped loss function that adjusts its focus during training complete the design.
The combined system reached a mean average precision near the top of the reported scale while cutting computational cost to a level the authors judge fit for real-time inspection. Ablation tests, where parts are removed one at a time, and comparisons with rival detectors each supported the design choices. Field relevance spans factory quality control and maintenance scanning of installed arrays.
Catalog record
| CREATOR | Zhongxing Zhang, Shaotong Pei, Chenlong Hu, Yunpeng Liu, Jie Liu, Guiting Na, Pengda Wang, Jie Mei, Chang Liu, Peizhe Zeng, Lianghui Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37781 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |