形狀自適應影像壓縮的係數掃描和分割技術

在近幾年的視訊編碼技術中，對不規則區域的影像作壓縮編碼已經變成一個非常重要的功能。形狀自適應編碼(Shape adaptive coding)的優點在於：這種方法可以運用同一個不規則影像區域中色彩強度的高相關性來對不規則形狀內的影像資訊做更好的壓縮，以達到更高的壓縮率。相較於傳統基於方型區塊的影像壓縮，形狀自適應的影像編碼可產生相當少的區塊效應以及變形失真的情形。這是由於基於方型區塊的影像壓縮忽略了影像的內容與特徵。傳統上，形狀自適應影像編碼技術要依賴格拉姆-施密特正交化(Gram-Schmidt Orthogonalization)來得到不規則影像區域的單位正交基底。然而，這種方法的運算複雜度過於龐大。相反地，雖然有其它運算複雜度較低的形狀自適應影像編碼技術，但這些方法的編碼效率並不高。在這篇論文中，我們提出了一個以基於三角形和梯形的二維正交離散餘弦轉換(Two dimensional orthogonal DCT expansion in triangular and trapezoid regions)的新式形狀自適應影像編碼技術。一般來說，任何的影像區塊都可以看成是很多多邊形的組合，而這些多邊形又可以分割為許多的三角形或梯形區域。因此，在本篇論文中我們提出了一些不規則影像區塊的近似梯形和三角形切割演算法除此之外，如同其他的形狀自適應轉換，基於三角形和梯形的二維正交離散餘弦轉換也會造成不規則形狀的係數區塊。然而，現有的鋸齒形掃描(Zigzag scan)並沒有考慮係數區塊的形狀。因此，我們提出了對於不規則形狀係數區塊的掃描方法，這種係數掃描方式可以簡單的置換於現有的影像壓縮技術中。從實驗結果證明，我們提出的不規則影像區塊的近似梯形和三角形切割演算法可以有效的找到幾乎和原來不規則影像區塊匹配的近似梯形和三角形。此外，我們提出的不規則形狀係數區塊的掃描方法相對於傳統的鋸齒形掃描也有較佳的編碼效率。最後，整合起來的基於三角形和梯形的二維正交離散餘弦轉換的新式形狀自適應影像編碼技術能夠比傳統的JPEG影像壓縮技術能有更好的壓縮效果。關鍵字：JPEG、形狀自適應編碼、係數掃描、影像切割。

關鍵字

影像壓縮；影像編碼；形狀自適應編碼；係數掃描；影像切割

並列摘要

Nowadays, coding of arbitrarily shaped image region is becoming an important feature in many visual coding applications. The advantage of shape-adaptive coding is that it can employ the information of arbitrarily-shaped region to exploit the high correlation of the color values within the same image segment in order to achieve a higher compression ratio. Compared to the conventional block-based image coding, shape-adaptive image coding produces significantly less blocking artifacts and distortions in other forms which typically emerges in block-based image coding since its negligence of the image content and characteristics. Traditionally, shape adaptive image coding relies on the Gram-Schmidt process the obtain orthonormal bases for an arbitrarily-shaped image segment. However, this method causes severe complexity issue. On the contrary, other methods may take low computation complexity but are not efficient in shape adaptive image coding. In this thesis, a new approach of object coding based on the two dimensional orthogonal DCT expansion in triangular and trapezoid regions is proposed. In general, any image segment can be viewed as a polygon and a polygon can be decomposed into several triangular or trapezoid regions. In this thesis, we propose a triangular and trapezoid segmentation algorithm to achieve the trapezoid approximation of an image segment. Besides, like the other shape adaptive transformations, applying the 2-D DCT expansion on these trapezoid regions leads up to irregular-shaped coefficient block. However, the existed zigzag scanning order does not concern the shape of the coefficient block. Solutions of scanning these irregular-shaped coefficients according to its shape are proposed on this thesis. The proposed coefficient scanning scheme can easily be implemented in existing image compression scheme. Results show that the trapezoid approximation of any image segments can be found and it nearly matches the shape of the image segments. Moreover, the proposed coefficient scanning scheme for generalized shaped can improve the coding efficiency. Finally, the proposed image compression scheme achieves better performance than the conventional JPEG image compression standard. Index terms— JPEG, Shape adaptive transform, coefficient scanning, image segmentation.

並列關鍵字

MPEG ； JPEG ； Image compression ； Shape adaptive ； Coefficient scanning ； Image segmentation ； object-based video coding

參考文獻

A. My Conference Paper

B. Digital Image Compression

[2] T. Acharya amd A. K. Ray, Image Processing Principles and Applications, John Wiley & Sons, New Jersey

[3] B. Zeng and J. Fu, “Directional discrete cosine transforms—A new framework for image coding,” IEEE Trans. Circuits Syst. Video Technol., vol. 18, no. 3, pp. 305-313, Mar 2008.

[4] N. Yamane, Y. Morikawa, T. Nairai, and A. Tsuruhara, “An image coding method using DCT in skew coordinates,” Electron. Commun. Japan, vol. 83, no.1, pp. 53-62, 2000.

國際替代計量

形狀自適應影像壓縮的係數掃描和分割技術

全文下載

主題瀏覽