查詢結果分析
來源資料
頁籤選單縮合
題 名 | 應用於RISC架構之H.264視訊編碼複雜度控制=Complexity Control for H.264 Video Coding on RISC Processors |
---|---|
作 者 | 黃振原; 許順閔; 鄭光宏; 鄧博元; 陳鴻元; | 書刊名 | 電腦與通訊 |
卷 期 | 127 2009.03[民98.03] |
頁 次 | 頁42-49 |
專 輯 | 無線與感測網路系統技術專題 |
分類號 | 448.94 |
關鍵詞 | 視訊編碼; 電力知曉; 複雜度控制; 動態調整電源電壓; Video coding; Power aware; Complexity control; Dynamic voltage scale; DVS; |
語 文 | 中文(Chinese) |
中文摘要 | 隨著無線感測網路的蓬勃發展,相關研究及應用引起越來越多的關注與討論。其裝置輕薄短小、攜帶方便,同時卻被賦予能夠支援多種功能的期望,如:影像擷取、即時視訊通訊……等等。為了使裝置在有限電源供應下同時能夠支援如此高運算量、高頻寬之要求,電力知曉(Power Aware)的設計觀念被應用在電力最佳化。電力知曉視訊編碼系統不僅是低功耗的設計,亦能針對無線視訊編碼系統,依據電池可供電剩餘時間調整編碼之複雜度動態調整功耗,以達到延長電池可供電時間之目的。本研究提出一參數化的視訊編碼架構。除了可以完全調整功耗,同時引入以人類主觀視覺之視訊品質評估方式,建立一個功率-位元率-失真(Power-Rate-Distortion)之視訊編碼系統模型。該系統可以調整其複雜度控制參數以符合裝置所能提供之能量,同時將視訊品質提升至最高。本研究在Intel Xscale處理器開發平台上實際探討,實驗結果顯示,此視訊編碼模型可減少系統高達75%之功耗。 |
英文摘要 | Wireless sensor networks is a rising trend. Related research and application topics attract more and more attention. A user's expectation for a device is light, thin, small, but powerful enough to support the increasing number of functions, such as photo capture, real-time video communications. To support these operations requiring high computing and high bandwidth on a device with limited energy, a power aware design concept is developed and applied in order to achieve power optimization. A power aware video coding system is not only a conventional low power design, but also a design that can adaptively adjust its encoding complexity based on the remaining battery supply to prolong the battery life. We developed a parametric video encoding architecture, which is fully scalable in power consumption, and imports a quality metric for subjective video quality measure. We established a power-rate-distortion model of the video encoding system. The system is able to adjust its complexity control parameters based on the available energy supply of the device while maximizing the picture quality. A development platform based on Intel's Xscale microprocessor is introduced. Experiments show that the system is able to save up to 75% power consumption in video coding based on this model. |
本系統中英文摘要資訊取自各篇刊載內容。