區塊分割免搜尋碎形影像編碼

REGIONAL DIVISION NO SEARCH FRACTAL IMAGE CODING

徐蔚仁(Wei-Jen Hsu)

指導教授 ： 汪順祥

大同大學/電機資訊學院/通訊工程研究所/碩士(2007年)

摘要

傳統的碎形影像編碼(fractal image coding)已經能得到不錯的影像
品質，但是花費在編碼的時間相當長，所以後來有學者提出了免搜尋
的碎形影像編碼。這個方法的確大幅提升了編碼的速度，但是會相當
程度地降低影像品質。我們參考了數篇提升碎形編碼影像重建品質的
方法之後，提出了將免搜尋碎形影像編碼與區塊分割編碼(block
truncation coding)結合的編碼方式。針對影像細部所使用的區塊分割編
碼可以有效彌補免搜尋碎形編碼普遍在劇烈變化部份表現不佳的缺
點；並且同時保留了免搜尋編碼在壓縮上能得到優勢的四元樹(quadtree)
特性。此外我們也將這個編碼方式導入彩色影像的編碼中，藉由部分
編碼共用的方式進一步地提升影像的壓縮比率。

關鍵字

免搜尋 ； 碎形影像編碼 ； 區塊分割

並列摘要

Conventional fractal image coding can achieve an extremely high
visual quality, but it requires lots of time in encoding. To solve the problem,
the no search method has been proposed in the literature. It reduces the
encoding time indeed, but the reconstruction quality is also reduced. By
investigating several papers that focus on raising the quality of image, we
propose a new method that combines the advantages of the no search
fractal coding method and block truncation coding method. The block
truncation coding method that focus on the detail of the image can alleviate
the drawback of no search method which cannot work well in the detail of
image. We also extend it to color image coding. By sharing part of the
codes, the compression ratio can be raised further.

並列關鍵字

no search ； fractal image coding ； regional division

參考文獻

[1] Arnaud E. Jacquin, “Image coding based on a fractal theory of iterated contractive

image transformations,” IEEE Transactions on Image Processing, Vol. 1, No. 1, pp.18 –

[2] Mario Polvere and Michele Nappi, “Speed-up in fractal image coding: comparison of

methods,” IEEE Transactions on Image Processing, Vol. 9, No. 6, pp.1002 – 1009, June

[3] Shen Furao, and Osamu Hasegawa, “An effective fractal image coding method