This paper proposes a frame level rate control scheme for HEVC based on support vector machine. Firstly, textual features of frames are extracted from video sequence. According to different complexities of textual features, support vector machine is chosen as a classifier for video classification. Secondly, a frame level rate control scheme is proposed to achieve higher coding efficiency based on classification results. According to the experiment results, the proposed algorithm achieves better coding performance than the original rate control scheme JCTVC -K0103 in HEVC.
Abstract-Considering the weak signal of partial radiant point, the traditional TDOA(Time Difference of Arrival) detector cannot meet the requirement. A weak signal detection method based on double time difference collaborative mechanism is proposed. In this thesis, the double time difference collaborative detector is designed, and theoretical analysis and mathematical derivation of the performance of the detector is implemented. Then the processing gain is given out. At last, the simulation has confirmed the performance of the detector.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.