Hybrid SVR-ACNN model: Proposed Video Super-Resolution method for video enhancement

Abstract

Video super-resolution techniques are the need of the hour for the high-resolution display devices as the current high-resolution videos are the basic question. Even though there are a large number of techniques employed for the video super-resolution, all these existing techniques face a hectic challenge at various conditions. Thus, this paper proposes an effective video resolution strategy using the hybrid Support vector regression-Actor Critic Neural Network (SVR-ACNN) model for video enhancement. The super-resolution images formed using the individual SVR model and Actor Critic Neural Network are integrated using the weighted average concept. The Actor Critic Neural Network is tuned optimally using the proposed Fractional-based Sine Cosine algorithm (F-SCA) that is responsible for the global optimal convergence. The experimentation of the proposed method utilizes three videos taken from the Cambridgedriving Labeled Video Database (CamVid), and the results are analyzed for three scaling factors. The outcome of the analysis proves that the proposed method offers a better super-resolution image with a better PSNR, SSIM, and SDME of 33.6447dB, 0.9398, and 45.2779, respectively.