An efficient field-programmable gate array-based hardware oriented block motion estimation algorithm based on diamond adaptive rood pattern search algorithm for multi-standard video codec

Abstract

Motion estimation and motion compensation techniques are effectively used for removing temporal redundancy between adjacent frames during video compression, but video quality will be degraded with the traditional pixel-based matching algorithms. To overcome this issue, diamond adaptive rood pattern search algorithm-based block-matching motion estimation algorithms are proposed in this manuscript for multi-stranded codec hardware design, which provide a high compression rate with less computational complexity. In this work, a diamond adaptive rood pattern search algorithm is used, which is a combination of a diamond-search algorithm and adaptive rood pattern search. In addition, the proposed algorithm applies multiple search strategies and terminates unnecessary searching while reserving high data regularity. The proposed diamond adaptive rood pattern search algorithm-based block-matching motion estimation algorithm for multi-stranded codec hardware design provides good performance in terms of complex calculations reduction and reaches a high speed in comparison to traditional search algorithms. The proposed architecture is implemented in Xilinx ISE design 14.5 suite and the implemented results are compared to the existing motion estimation techniques on field-programmable gate array devices. Simulation results shows that the diamond adaptive rood pattern search algorithm for multi-stranded codec hardware design provides a very good improvement in terms of computations overhead reduction, with 96.4%, 97.3% latency and 98.7% throughput and achieves a high compression rate without video quality degradation.

An efficient VLSI architecture for fast motion estimation by hybridizing parallel spiral and adaptive threshold star diamond search algorithm‐based search algorithm in HEVC

“…4 A direct BMA that comprehensive search as whole blocks of feasible candidate as mentioned area is termed as a full search block-matching (FSBM). 5,6 To prove the procedure of feasible reference current block matching frame via well-modified method is preferred as an approximation of block-based motion. The encryption of the video utilizes ME to remove the temporary sequences of the video.…”

“…Every standard is blocked, because the approximate motion module is activated in BMAs 4 . A direct BMA that comprehensive search as whole blocks of feasible candidate as mentioned area is termed as a full search block‐matching (FSBM) 5,6 . To prove the procedure of feasible reference current block matching frame via well‐modified method is preferred as an approximation of block‐based motion.…”