透過您的圖書館登入
IP:3.144.93.73
  • 期刊

進階雙預測器DPCM影像傳輸系統

Advanced DP-DPCM Image Transmission System

摘要


本論文是針對雙預測器誤差脈衝編碼調變(double predictor differential pulse code modulation,DP-DPCM)影像壓縮編碼法加以改進,以增加影像壓縮率和傳輸速度;我們利用四分樹影像分割(quadtree segmentation)法,將影像切為不同尺寸大小的小區域影像方塊,影像變化大的細節(details)部分被分割成較小的區域處理,而變化小的背景部分則被分割成較大的區域。因此,區域影像方塊輸入至DP-DPCM系統中的預測器時,藉由鄰近像素點間差異性降低的特點,減小預測誤差分佈範圍,達到縮小量化器階數及降低位元使用率的目的。此外,在考慮無誤差、無雜訊(error free)的傳輸通道下,傳統的DPCM系統容易造成過大的量化誤差回授,而導致預測準確率大幅下降;本研究是以採用雙預測器DPCM的架構,來避免量化誤差回授的影響。本論文提出的概念和方法,可以在不增加太多複雜度(complexity)下,就提升整個系統的效能(performance),大約比傳統DPCM的作法多出5~6 dB的SNR值。

並列摘要


This paper is focused on the improvement of Double Predictor Differential Pulse Code Modulation algorithm for image data compression to increase the image transmission speed. We have utilized the Quadtree segmentation algorithm to divide the real-world images into regions having widely different perceptual importance. The detail regions of a given image will be segmented into blocks with smaller block size, and the background regions of the image will be assigned larger block size to the image blocks. After the Quadtree segmentation procedure, the differential values between the nearby pixels within an image block are reduced. Therefore, we can decrease the distribution range of the prediction error as well as reduce the quantization levels and the bit rate. We have also adopted the double predictor DPCM architecture for the proposed image coding system. Conventional DPCM image coding systems can be easily affected by a larger fed-back quantization error while transmitted over an error-free channel. The advantage of double predictor DPCM system is to reduce the fed-back quantization error and not to increase the system complexity. The system performance of the proposed variable block-size Double Predictor DPCM image encoder/decoder system has shown about 5 to 6 dB coding gain in Signal-Noise-Ratio than a conventional DPCM system.

延伸閱讀