簡易檢索 / 詳目顯示

回結果列表
研究生： 宋題均
Sung, Ti-Chun.
論文名稱： 利用GPU架構實作碎形影像壓縮演算法
Implementation of Fractal Image Compression on GPU
指導教授： 陳朝欽
Chen, Chaur-Chin
口試委員: 朱學亭
陳建彰
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 資訊系統與應用研究所
Institute of Information Systems and Applications
論文出版年： 2019
畢業學年度： 107
語文別： 英文
論文頁數： 32
中文關鍵詞： 影像壓縮 、碎形影像 、平行計算
外文關鍵詞： Fractal Image, Image Compression, Gpu Architecture
相關次數： 點閱：1  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 碎形影像壓縮演算法是一種利用影像自身相似性的壓縮方法，儘管他提供許多很好的特性，但此種方法需要消耗非常多的壓縮時間。雖然許多演算法的改良都努力想要縮短壓縮時間，不過壓縮過程依舊十分耗時。隨著硬體設備的不斷的進步，圖形處理器(GPU)廣泛的被使用在科學計算上，數百或數千個內核，可運行大量的計算工作。這讓我們能夠藉由高度的平行化運算，來加速碎形影像壓縮。

    此篇論文中將介紹碎形影像壓縮的基本原理，以及 GPU 架構的運行機制。我們研讀了利用分類方法來改良的壓縮演算法，並在 GPU 架構上實作此方法，藉由高度的平行化運算及優化，來大幅縮短影像壓縮的時間。

    Fractal image compression (FIC) is a lossy compression method based on the local similarity of image structure. Despite that it provides many advantages, it suffered from its large amount of computation needed for the encoding stage. Many researches and algorithms have been developed in order to reduce compression time, but its encoding procedure still remains a huge consumption. However, with the rapid improvement of the Graphics Processing Unit (GPU) technology, it becomes much easier that we can utilize GPU to solve more and more computational problems. And due to the process of fractal image encoding can be parallel independent computation, we can reduce the encoding time to the extreme extent by means of running on GPU.

    In this study, we introduced the basic theorem behind the FIC, then implement the baseline FIC and APCC-based (the absolute value of Pearson correlation coefficient) FIC algorithms on GPU parallel architectures to get significantly less compression time.

    1 Introduction . . . . . . . . . . . . 1 2 Related Work . . . . . . . . . . . . 3 2.1 Collage theorem . . . . . . . . . . . 3 2.2 Fundamental FIC . . . . . . . . . . . 4 2.3 APCC-based FIC . . . . . . . . . . . 7 2.4 Architecture of General-purpose GPU . . . . . 8 2.5 CUDA Programming . . . . . . . . . . 10 3 Methodology. . . . . . . . . . . . 11 3.1 The Number of Parallel Threads . . . . . 11 3.2 Usage of Shared Memory . . . . . . . . 12 3.3 Algorithm of Fundamental FIC . . . . . . 12 3.4 Algorithm of APCC-based FIC . . . . . . . 13 3.5 Implementation of two algorithms with CUDA . . 15 4 Experiment and Results. . . . . . . . . 19 4.1 Implementation environment . . . . . . . 19 4.2 Experiment . . . . . . . . . . . . 20 4.2.1 Encoding time of different size images . . 20 4.2.2 PSNR and SSIM . . . . . . . . . . . 20 4.2.3 Encoding with different range block sizes. . 21 4.2.4 Color image encoding . . . . . . . . . 21 5 Conclusion and Future Work. . . . . . . . 29 References. . . . . . . . . . . . . . . 31

    References
    [AlSa2017] N.-M.-G. AlSaidi and A.-H. Ali. Towards enhancing of fractal image com-pression performance via block complexity.Annual Conference on NewTrends in Information Communications Technology Applications-(NTICT’2017)7 - 9 March:246–251, 2017.
    [Alva2015] O. Alvarado-Nava. Gpgpu implementation of fic using texture memory.In-ternational Work Conference on Bio-inspired Intelligence:185–190, 2015.ISBN:978-1-4673-7846-8/15.
    [Barn2006] M.-F. Barnsley and J. Hutchinson. New methods in fractal imaging.Pro-ceedings of the International Conference on Computer Graphics, Imagingand Visualisation:1–6, 2006. ISBN:0-7695-2606-3/06,IEEE.
    [Chu2001] H.-T. Chu and C.-C. Chen. Accelerating fractal compression with a real-time decoder.Joural of Information Science and Engineering, 17:417–427,2001.
    [Geor2009] E. George and A. AL-HILO. Fractal color image compression by adaptivezero-mean method.2009 International Conference on Computer Technol-ogy and Development:525–529, 2009. ISBN:978-0-7695-3892-1/09,IEEE.
    [Josh2017] M. Joshi, R. Belwal, and B. Gupta. A review on different techniques offractal image compression.International Journal of Advanced Research inComputer EngineeringTechnology, 6:1164–1177, 8, 2017.ISSN: 2278-1323.[Komi2001] J. Kominek. Algorithm for fast fractal image compression.Proc. SPIE 2419,Digital Video Compression: Algorithms and Technologies, 1995.
    [Li2018] W. Li. Research on image fractal compression coding algorithm based ongene expression programming.2018 17th International Symposium on Dis-tributed Computing and Applications for Business Engineering and Sci-ence:88–91, 2018.
    [Li2000] Z. Li and L. Zhao. Fractal color image compression:185–192, 2000. ISBN:0-7695-0878-2/00,IEEE.[Seel2012] D.-S. Seeli and M.-K. Jeyakumar. A study on fractal image compression us-ing soft computing techniques.International Journal of Computer ScienceIssues, 9, 6,No2, 2012.ISSN: 1694-0814.
    [Shar2013] M.-P. Sharabayko and N.-G. Markov. Fractal compression of grayscale andcolor images, 2013. ISBN:978-1-4673-1773-3/12,IEEE.
    [Wang2017] J. Wang, P. Chen, and J. Liu. Fast sparse fractal image compression.PLOSONE, http://doi.org/10.1371/journal.pone.0184408, 2017.
    [Wang2013] J. Wang and N. Zheng. A novel fractal image compression scheme withblock classification and sorting based on pearson’s correlation coefficient.IEEE Transactions on image processing, 22,No.9:3690–3702, 2013. ISBN:1057-7149.[Github]https://github.com/jimmy43333/GPU_Fractal_Image_Compression.[Web01]https://http.download.nvidia.com/developer/cuda/seminar/TDCI_Arch.pdf.
    [Web02]http://haifux.org/lectures/267/Introduction-to-GPUs.pdf.[Web03]http://www.netprint101.com/text/43373744-413.html.[Web04]https : / / blog . csdn . net / sunmc1204953974 / article /details/51088899.[Web05]https : / / core . ac . uk / download / pdf / 62876379 . pdf ?fbclid=IwAR0lOrHjP9R5fhFzb_OMRJ5Yk4Vbctln__NLOzOTS7uhKT4-w90E4B9DlXw.

    QR CODE