A neural network framework for selecting real-time video enhancement algorithms on mobile devices

Real-time video enhancement on mobile devices is crucial for modern services like video calling, augmented reality, medical imaging, and surveillance. However, limited processing power, battery life, and memory limit the selection of video enhancement algorithms. Techniques aim to reduce noise, improve resolution, and enhance contrast, but their effectiveness depends on processing speed, visual quality, power consumption, and implementation complexity. Balancing these performance parameters is a challenge in real-time applications. Optimally selecting real-time video enhancement algorithms is very difficult and risky when using the classical model. Therefore, we develop a new decision-making model based on fuzzy neural networks with Sugeno-Weber norms. The proposed model selects a real-time video enhancement algorithm. The proposed model provides Deep Learning Super Resolution as an optimal real-time video enhancement algorithm. The decision results are verified based on well-known decision approaches to assess the accuracy of the comparison.