Fuzzy Holoentropy-Based Adaptive Inter-Prediction Mode Selection for H.264 Video Coding

Fuzzy Holoentropy-Based Adaptive Inter-Prediction Mode Selection for H.264 Video Coding

Srinivas Bachu (Department of ECE, Marri Laxman Reddy Institute of Technology and Management, Dubdigal, Hyderabad, Telangana, India) and N. Ramya Teja (Department of ECE, Guru Nanak Institutions Technical Campus (Autonomous), Ibrahimpatnam, Hyderabad, Telangana, India)
DOI: 10.4018/IJMCMC.2019040103

Abstract

Due to the advancement of multimedia and its requirement of communication over the network, video compression has received much attention among the researchers. One of the popular video codings is scalable video coding, referred to as H.264/AVC standard. The major drawback in the H.264 is that it performs the exhaustive search over the interlayer prediction to gain the best rate-distortion performance. To reduce the computation overhead due to exhaustive search on mode prediction process, this paper presents a new technique for inter prediction mode selection based on the fuzzy holoentropy. This proposed scheme utilizes the pixel values and probabilistic distribution of pixel symbols to decide the mode. The adaptive mode selection is introduced here by analyzing the pixel values of the current block to be coded with those of a motion compensated reference block using fuzzy holoentropy. The adaptively selected mode decision can reduce the computation time without affecting the visual quality of frames. Experimentation of the proposed scheme is evaluated by utilizing five videos, and from the analysis, it is evident that proposed scheme has overall high performance with values of 41.367 dB and 0.992 for PSNR and SSIM respectively.
Article Preview
Top

1. Introduction

Advancement in the multimedia applications has paved the way for giving importance to video compression techniques. Video compression is a way of representing the video with lower data bits. Several applications using digital media have made use of video compression standards. Also, ever increasing demand for internet sources, have paved the way for live streaming of events all over the world (Irannejad & Mahdavi-Nasab, 2017). In the late ’90s, several video streaming resources came into the picture, and they allowed the compression process through video encoding schemes. Owing to their complexity and advancement in the digital era, various advanced video coding standards, such as MPEG-4 Part 2, H.263, H.262/MPEG-2, and H.261, are introduced (Vigneash & Marimuthu, 2017). These technologies have revolutionized the video market, and also made digital sources at an affordable rate (Liu. et al., 2014). Video processing schemes developed in recent era concentrates on area and power of the equipment, and they indented to reduce the computational intensity. As video quality acts as prime criteria, several techniques have less focused on reducing power consumption (Basha & Kannan, 2017). The video is the rapid movement of image frames, and thus, frame sequence can be categorized as ‘I’ frame, ‘P’ frame, and ‘B’ frame. The ‘I’ frame constituted in the video is indicated as a key/reference frame, since it has information contained in the upcoming frame. The ‘P’ frames are said to be prediction frames as they are the continuation or small movement from ‘I’ frame. The ‘B’ frame has pixel movement in both forward and reverses direction (Dolly., et al., 2017). Various multimedia applications, such as TV, video conference, mobile, and video streaming make use of video compression schemes.

Motion estimation schemes discussed in literature can be subdivided into two major categories, namely pixel-based motion estimation and block-based motion estimation. In a pixel-based motion estimation scheme, varying motion between frames is identified by constructing a motion vector for each pixel in the frame. Rather, in block-based motion estimation scheme, frames are subdivided into macroblocks and motion vector is constructed for blocks (Hemanth & Anitha, 2017). Rapid development in video processing schemes allowed researchers to provide improved video compression schemes. Compressing video size without compromising the quality of video frames acts as criteria for video compression algorithms. The algorithms try to find temporal correlation among the successive frames for achieving high compression ratio (Liu. et al., 2014). Video communication standards require highly efficient video compression techniques, as they reduce memory requirement and also improves the communication speed. Video compression tries to represent the video using less number of bit and thus, reduces the size of the video (Hemanth & Anitha, 2017). Major criterion involved in video compression schemes is that the compression process should not alter the visual quality of the video (Hemanth & Anitha, 2017). Video compression is done by removing temporal redundancy prevailing in video sequences.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2020): Forthcoming, Available for Pre-Order
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing