Abstract

Image filtering and enhancement play a pivotal role in ensuring the quality and clarity of visual content, particularly in multimedia video streaming applications. Existing filtering techniques often struggle with balancing noise reduction, detail preservation, and real-time performance, resulting in suboptimal outcomes in dynamic video environments. Furthermore, video streaming systems demand adaptive solutions that cater to diverse lighting and noise conditions. To address these challenges, a novel Enhanced Adaptive Image Filtering and Enhancement framework combining Deep Artificial Neural Networks (Deep ANN) with Adaptive Histogram Equalization (AHE) is proposed. This method leverages the powerful learning capabilities of Deep ANN to identify noise patterns and preserve critical details, while AHE dynamically adjusts contrast to improve visual quality in varying lighting conditions. The proposed framework is tested on real-time video streaming datasets, simulating environments with low light, noise, and high-motion scenarios. The results show significant improvements over traditional filtering methods. Experimental evaluations show an increase in Peak Signal-to-Noise Ratio (PSNR) to 42.3 dB, compared to 37.1 dB achieved by conventional methods. Structural Similarity Index Measure (SSIM) reached 0.96, reflecting enhanced detail preservation and perceptual quality. Moreover, the framework achieved a 35% reduction in Mean Squared Error (MSE) and maintained an average processing speed of 28 frames per second, making it suitable for real-time applications. These findings highlight the potential of combining advanced neural network capabilities with adaptive histogram techniques to enhance multimedia video streaming quality. This method ensures superior performance in diverse environments, paving the way for immersive and reliable video streaming experiences.