Low-light Video Enhancement with SVD-DWT Algorithm for Multimedia Surveillance Network

Abstract

After the outbreak of the global pandemic COVID-19, video surveillance has become an important aspect of multimedia communication in monitoring-based systems, security systems, object tracing etc. However, due to drawbacks like video frame distortion, the quality of the video in surveillance may be affected in some cases. In this work, a video enhancement-based multimedia surveillance network has been developed by using a Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD). The proposal, in particular, uses SVD-DWT along with median filtering and contrast limited adaptive histogram equalization to enhance low illuminated distorted video frames by using video sensor nodes. The performance of the proposed scheme is evaluated by calculating parameters like mean square error, peak signal to noise ratio, normalized cross-correlation etc. of the video frames in different types of environments. The results delineate that the proposed scheme is better than other available methods in diverse types of environments.