Abstract
In order to solve the problems of high cost and low efficiency of manual fault detection of photovoltaic modules in real life, a typical fault classification and recognition system for photovoltaic modules based on thermal imaging photography and machine processing was proposed. Feature extraction network is used to extract features from thermal imaging images. Classify different types of faults and make training sets; The YOLOV5 target detection algorithm was used to train the model through subset pre-training, weight allocation, multi-group multi-number training and other methods, and the test set was used for multiple tests. The test results show that the system has a strong ability to identify faults, and the average accuracy of network detection in the task of fault classification reaches 84.2%. In the complex area images including the target photovoltaic modules, the accuracy of the result reached 79.1% after the test set tested the training model. It is judged that the system can replace manual work to complete the typical fault classification and identification of photovoltaic modules.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Lecun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436 (2015)
Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1137–1149 (2017)
Wang, W., Wu, B., Yang, S., et al.: Road Damage Detection and Classification with Faster R-CNN. In: 2018 IEEE International Conference on Big Data (Big Data). IEEE (2018)
Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6) (2015)
Chen, Z., Wang, H., Zhang, L., et al.: Visual saliency detection based on homology similarity and an experimental evaluation. J. Vis. Commun. Image Repres. 40(pt.A), 251–264 (2016)
Li, X., Zhao, L., Wei, L., et al.: DeepSaliency: multi-task deep neural network model for salient object detection. IEEE Trans. Image Process. Publ. IEEE Signal Process. Soc. 25(8), 3919 (2016)
Nian, B., Fu, Z., Wang, L., et al.: Automatic detection of defects in solar modules: image processing in detecting. In: International Conference on Wireless Communications Networking & Mobile Computing. IEEE (2010)
Laganiere, R.: OpenCV 2 computer vision application programming cookbook : over 50 recipes to master this library of programming functions for real-time computer vision (2011)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Xu, S. (2022). Typical Fault Classification and Recognition of Photovoltaic Modules Based on Deep Learning and Thermal Imaging Picture Processing. In: Su, R., Zhang, YD., Liu, H. (eds) Proceedings of 2021 International Conference on Medical Imaging and Computer-Aided Diagnosis (MICAD 2021). MICAD 2021. Lecture Notes in Electrical Engineering, vol 784. Springer, Singapore. https://doi.org/10.1007/978-981-16-3880-0_44
Download citation
DOI: https://doi.org/10.1007/978-981-16-3880-0_44
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-3879-4
Online ISBN: 978-981-16-3880-0
eBook Packages: EngineeringEngineering (R0)