Please wait ...
0% Complete
فارسی
Home
/
پانزدهمین کنفرانس بین المللی فناوری اطلاعات و دانش
A Data-Efficient Approach to Solar Panel Micro-Crack Detection via Self-Supervised Learning
Authors :
Alireza Akhavan safaei
1
Pegah Saboori
2
Reza Ramezani
3
Morteza Tavana
4
1- دانشگاه اصفهان
2- دانشگاه اصفهان
3- دانشگاه اصفهان
4- شرکت آسمان رصد هادی
Keywords :
Data Augmentation،Micro-Crack Detection،Convolutional Neural Network،Self-Supervised Learning،Transfer Learning
Abstract :
This study presents a method for the automatic identification of micro-cracks in photovoltaic solar modules using deep learning techniques. The main challenge in this research is the lack of labeled data and class imbalance for the detection of micro-cracks. The proposed method employs a multi-stage approach. Initially, 10% of the dataset is manually labeled to train a simple convolutional neural network model. This model is then used to generate pseudo-labels for the unlabeled data using a self-supervised approach. The pseudo-labels are manually reviewed to increase the number of micro-crack samples in the training set. Data augmentation techniques are also applied to increase the size and diversity of the training dataset. Finally, the pre-trained ResNet-50 model is fine-tuned on the expanded labeled dataset for accurate detection of micro-cracks. Advanced preprocessing steps, including solar cell segmentation, cropping, and data augmentation, have been performed. The class imbalance problem is addressed through undersampling and weighted loss functions. The experimental results demonstrate the effectiveness of the proposed method, achieving an accuracy of 0.9782 and an F1-score of 0.7776 in the detection of micro-cracks in electroluminescence images of solar panels. This study provides insights into the use of limited labeled data for training robust deep learning models for the identification of defects in solar modules.
Papers List
List of archived papers
From Faces to Words: An Efficient Persian Visual Lip Reading
Mana Amini - Sajjad Aemmi - Azadeh Ashouri - Reza Akhoundzadeh - Kourosh Hassanzadeh - Mohammad Reza Mohammadi
Targeted Vaccination for COVID-19 Using Mobile Communication Networks
Mohammadmohsen Jadidi - Pegah Moslemi - Saeed Jamshidiha - Iman Masroori - Abbas Mohammadi - Vahid Pourahmadi
A Blockchain-Based Smart Contract Framework for Peer-to-Peer Energy Trading in Smart Grids
Hossein Shahinzadeh - Farshad Ebrahimi - S. Mohammadali Zanjani - Amirafshin Zamani - Saiedeh Mehrabani-Najafabadi - Gevork B. Gharehpetian
Effective Classifier for Predicting Churn in Payment Terminals Using RFM model and Deep Neural Network
Dr Mahila Dadfarnia - Ali Alemi Matinpour - Dr Monireh Abdoos
Automatic identification and reconstruction of Tuberculosis in microscopic images using convolutional auto-encoder network
Ahmad Reza Nadafi - Farahnaz Mohanna
ParsEL 1.0: Unsupervised Entity Linking in Persian Social Media Texts
Majid Asgari-bidhendi - Farzane Fakhrian - Dr Behrouz Minaei-bidgoli
Distributed Learning Automata-based Algorithm for Finding K-Clique in Complex Social Networks
Mohammad Mehdi Daliri Khomami - Alireza Rezvanian - Ali Mohammad Saghiri - Mohammad Reza Meybodi
NFV-Based Distributed Service Function Chaining with Imperfect Information
Mahsa Alikhani - Marzieh Sheikhi - Dr Vesal Hakami
طرحی برای تبدیل نمودارهای رفتاری BPMN به نمودار UML و تولید کد از آن
مهدیس صفری - احمد عبدالله زاده بارفروش
Sparse Beamforming Design for Non-Coherent UD-CRAN with mm-Wave Fronthaul Links
Alireza M. Hosseini - Dr Abbas Mohammadi
more
Samin Hamayesh - Version 45.0.1