A5051458012- Digital Media Forensic Detection
- Music and Audio Processing
- Solar Radiation and Photovoltaics
- Energy and Environment Impacts
- Photovoltaic System Optimization Techniques
- Anomaly Detection Techniques and Applications
- Currency Recognition and Detection
- Industrial Vision Systems and Defect Detection
- Advanced Neural Network Applications
SK Group (South Korea)
2022-2023
Methods that enable the visual inspection of solar panels are currently in demand, as a huge number now being deployed sustainable energy source. One solutions for automation is an end-to-end deep learning framework, but this not recommended problem because such framework requires only powerful computational resources, also large-scale class-balanced dataset. In study, we present cost-effective panel defect detection method. We emphasize spatial feature defects by utilizing attention map...
Modern society interests on the renewable energy and maintaining efficiency of them. In case using solar energy, recognition response to defectives as soon possible is recommended because defects in panels reduces efficiency. same context interest it would be better use a proper lightweight defective detection model than high-performance heavy model. order reduce computational load training procedure, we define statistical features from panel those for detection. Also, assuming that...
Recent studies show edge computing-based road anomaly detection systems which may also conduct data collection simultaneously. However, the computers will have small storage but we need to store collected audio samples for a long time in order update existing models or develop novel method. Therefore, should consider an approach efficient management methods while preserving high-fidelity audio. A hardware-perspective approach, such as using low-resolution microphone, is intuitive way reduce...