應用於無線膠囊內視鏡之低成本與低複雜度無失真影像壓縮積體電路設計

VLSI Implementation of a Low-Cost and Low-Complexity Lossless Image Compression Encoder for Wireless Capsule Endoscopy

陳宥任(Yu-Ren Chen)

指導教授 ： 陳世綸

中原大學/工學院/電子工程研究所/碩士(2014年)

https://doi.org/10.6840/cycu201400582

摘要

本論文提出了一種基於JPEG-LS的低成本與低複雜度無失真影像壓縮演算法，用以實現於無線膠囊內視鏡的積體電路設計。由於本論文中所提出的演算法是基於積體電路設計的實現，而傳統JPEG-LS standard中的Context Model有大量的記憶體需求以及較高的運算複雜度，因此在本演算法中被移除。本論文所提出的演算法中只包含了一個預測模組、常規模組和混合熵編碼模組，而混合熵編碼模組同時使用了霍夫曼編碼與哥倫布編碼技術，少了Context Model，因此演算法的複雜度比JPEG-LS standard的要少得多。除此之外，相較於以前的研究，本論文所提出的演算法在壓縮色彩濾波陣列格式（CFA）的影像時，平均壓縮率提升超過6 ％。 而在硬體方面與其他基於JPEG-LS的電路設計相比，本論文所提出的無失真影像壓縮電路設計採用0.18微米CMOS製程，電路的操作頻率可達到200 MHz，gate counts只有5.54 K，晶片面積為68,291 μm2。與先前的電路設計相比，本論文所提出的電路設計降低了至少52 ％的gate counts和32 ％的記憶體需求。

關鍵字

色彩率波陣列 ； 哥倫布編碼 ； VLSI. ； 霍夫曼編碼 ； 膠囊內視鏡 ； 混合熵編碼 ； 無失真影像壓縮 ； JPEG-LS

並列摘要

In this thesis, a low-cost and low-complexity lossless image compression algorithm based on JPEG-LS is proposed for the implementation of wireless capsule endoscopy. Since the proposed algorithm is developed for VLSI implementation, a context model in traditional JPEG-LS standard is reduced because of the high complexity and memory requirement. The proposed novel low-complexity lossless image compression algorithm contains a prediction model, a regular mode, and a hybrid entropy coding model. The hybrid entropy coding model consists of a Huffman and a Golomb-Rice coding techniques. Since the proposed algorithm is without using context model, the complexity and memory requirement are much less than the JPEG-LS standard. In addition, compared with the previous studies, this work promotes the average compression rate for color-filter-array (CFA) images by over 6 %. To realize the proposed algorithm by VLSI technique, the gate count of this work is only 5.54 k and the core area is 68,291 μm2 synthesized by using TSMC 0.18 μm CMOS process. The operating frequency can be up to 200 MHz. Compared with the previous lossless compression designs, this work reduces at least 52 % gate counts and 32 % memory requirement.

並列關鍵字

JPEG-LS ； Hybrid Entropy Coding ； VLSI ； Capsule endoscopy ； lossless image compression ； color filter array ； Huffman ； Golomb-Rice

參考文獻

[3] X. Xie, G. L. Li, X. W. Li, Z. H. Wang, C. Zhang, D. M. Li, and L. Zhang “A New Approach for Near-lossless and Lossless Image Compression with Bayer Color Filter Arrays,” in Proc. ICIG 2004, pp. 357–360.

[4] X. K. Chen, H. J. Jiang, X. W. Li, and Z. H. Wang, “A Novel Compression Method for Wireless Image Sensor Node,” in Proc. IEEE Asian Solid-State Circuit Conf. (ASSCC’07), Jeju, Nov. 2007, pp. 184-187.

[5] M. J. Weinberger, G. Seroussi, and G. Sapiro, “The LOCO-I Lossless Image Compression Algorithm: Principles and Standardization into JPEG-LS,” IEEE Trans. Image Processing, vol. 9, no. 8, pp. 1309-1324, Aug. 2000.

[6] M. J. Weinberger, J. Rissanen, and R. B. Arps, “Applications of universal context modeling to lossless compression of gray-scale images,” IEEE Trans. Image Processing, vol. 5, pp. 575–586, Apr. 1996.

[7] P. Merlino and A. Abramo, “A Fully Pipelined Architecture for the LOCO-I Compression Algorithm,” IEEE Trans. VLSI Systems, vol. 17, no. 7, pp. 967-971, Jul. 2009.