Detection of PV Solar Panel Surface Defects using Transfer Learning of the Deep Convolutional Neural Networks
Imad Zyout, Abdulrohman Oatawneh
20202020 Advances in Science and Engineering Technology International Conferences (ASET)31 citationsDOI
Abstract
The need for automatic defect inspection of solar panels becomes more vital with higher demands of producing and installing new solar energy systems worldwide. Deep convolutional neural networks (CNN) remarkably perform very well for solving the image classification task from different domains. In this paper, the convolutional neural network is applied to characterize the surface of the PV panel and to detect the presence of the defect. The application of transfer learning with AlexNet CNN provided a very promising performance and reveal the potential of the approach for the detection of various defects in the surface of the solar panel.
Topics & Concepts
Convolutional neural networkTransfer of learningDeep learningComputer scienceArtificial intelligenceInstallationArtificial neural networkPattern recognition (psychology)Surface (topology)Machine learningMathematicsOperating systemGeometryIndustrial Vision Systems and Defect DetectionPhotovoltaic System Optimization TechniquesAdvanced Neural Network Applications