H.264即時影像解碼器之設計與實現

Abstract

隨著多媒體應用的普及與影像處理技術的進步，多媒體影像的效果與檔案的大小都有相當程度的提升。因此，多媒體影像的新格式與相對應的加解碼技術就廣泛的被討論與開發研究。由ISO/IEC MPEG與ITU-T VCEG聯合組織(JVT)所發展出來的H.264被視為下一代的影像壓縮技術。H.264提供了兩種壓縮方式：依減少空間的相依性壓縮的I frame；與減少時間的相依性壓縮的P frame。在這篇論文中，我們於H.264的軟體解壓縮上提出一個快速而有效的方法。透過組合語言與C語言的整合，對H.264的I frame提出快速的解壓縮技術。我們達成H.264即時且快速的解碼速度。實驗結果顯示我們的結果在平均情況下可以對I frame解碼得到多達70倍的改進，並且不會影響影像的品質。

As multimedia application popularization and image processing technique progression, both performance and size of multimedia file have a big amount of increment. Therefore, the new video compressed standard and its encoding/decoding techniques are discussed and developed broadly. The Joint Video Team (JVT) of ISO/IEC MPEG and ITU-T VCEG developed a new standard “H.264”, is regarded as the next generation video compression standard. In H.264, it provides two video compression schemes: Compress data by reducing the spatial redundancy, called I frame; compress data by reducing the temporal redundancy, called P frame. In this thesis, we proposed a fast and efficient method in H.264 software decoding in I frame. By using assembly language coding and special C coding techniques, we can reach real-time decoding in H.264 I-frame. Experimental Results shows we can achieve 70 times improvements on average in QCIF format, compared with JM decoder, without any video quality loss.