A novel Motion Estimation architecture, used for homeland security video surveillance systems, proposed in this paper. The proposed architecture is suitable to be adopted in H.265 High Efficient Video Coding (H.265/HEVC) encoder of video surveillance systems. The proposed architecture uses level A and level B data reuse to minimize memory I/O while fully utilizing the hardware resources. An efficient local memory is used for performing data reuse while loading both the current block and the search area inside the Processing Element (PE) array. The proposed architecture has a smooth and a regular data flow. The video resolution accuracy of the proposed architecture is very high. Subjective and quantitative measures are used to measure the performance of the proposed architecture. Our architecture can be adopted in the encoder of video surveillance systems and any High Quality (HQ) video related applications such as High Definition (HD) broadcasting. The time verification of the proposed architecture is tested using Modelism-version10.4a. © 2018 IEEE.

View