北京生物醫(yī)學(xué)工程

改進(jìn)的基于遺傳算法稀疏分解的腦CT圖像壓縮

Brain CT image compression with sparse decomposition based on improved genetic algorithm

作者：黃平安胡晏婷王俊

單位：南京郵電大學(xué)通信與信息工程學(xué)院(南京210003)

分類號(hào)：

出版年·卷·期（頁碼）：2012·31·4（356-360）

摘要：

目的提出一種新型的稀疏分解算法，對(duì)腦部CT圖像進(jìn)行壓縮。方法本文采用改進(jìn)的遺傳算法(genetic algrithm，GA)與匹配追蹤（matching pursuit，MP）算法相結(jié)合以實(shí)現(xiàn)稀疏分解，對(duì)腦部CT圖像進(jìn)行壓縮以節(jié)約存儲(chǔ)空間。針對(duì)原有遺傳算法計(jì)算時(shí)間長、匹配率低的不足，本方法優(yōu)化了迭代次數(shù)的選擇、競爭、變異等操作。結(jié)果利用該算法對(duì)腦部CT圖像分塊壓縮，使運(yùn)算速度、壓縮比和信噪比均得到提高。結(jié)論通過分析與實(shí)驗(yàn)驗(yàn)證，改進(jìn)的方法壓縮比例更大，失真更小，運(yùn)行時(shí)間更短，為腦部CT圖像的壓縮提供了一種新方法。

Objective This paper proposes a novel sparse decomposition algorithm to compress the brain CT image. Methods Sparse decomposition is achieved with improved genetic algorithm (GA) and matching pursuit (MP) algorithm to compress the brain CT image to save storage space. For overcoming some shortcomings in original algorithm, such as long time assuming and low matching rate, the improved genetic algorithm optimizes certain operations such as selection, competition and mutation of iterations. Results We compress the brain CT images by the improved algorithm, making progress in computing speed and computational accuracy. Conclusions This improved method shows greater compression ratio, smaller distortion and shorter running time, and proposes a new method for brain CT image compression.

參考文獻(xiàn)：

［1］ MallatS, Zhang Z, Matching pursuit with time-frequency dictionaries ［J］. IEEE Trans on Signal Processing,1993,41(12):3397-3415.
［2］Bergeau F, Mallet S. Match pursuit of images［C］. Piladelphia, PA, USA：Proceedings of IEEE-SP, 1994： 330-333.
［3］王建英，尹忠科，張春梅. 信號(hào)與圖像的稀疏分解及初步應(yīng)用［M］.成都：西南交通大學(xué)出版社，2006:152-168.
［4］張曉，戴冠中，徐乃平. 一種新的優(yōu)化搜索算法——遺傳算法［J］. 控制理論與應(yīng)用，1995,12(3):265-273.
［5］Tu ZW, Chen XR, Yuille A L,et al. Image parsing unifying segmentation, detection and recognition［J］. Int J Comput Vis, 2005, 63: 113-140.
［6］Yin ZK, Xie M, Wang JY. Fast MP of images based on atom energy property［J］. Journal of University of Electronic Science and Technology of China(UESTC), 2005, 34(2):156-159.
［7］尹忠科, 王建英. 一種新的圖像稀疏分解快速算法［J］. 計(jì)算機(jī)應(yīng)用, 2004,24(10): 92-96.
Yin Zhongke, Wang Jianying. New fast algorithm to sparsely decompose images［J］. Journal of Computer Applications, 2004,24(10): 92-96.
［8］Wang CG, Liu JJ, Sun JX. Compression algorithm for electrocardiograms based on sparse decomposition［J］. Frontiers of Electrical and Electronic Engineering in China, 2009(3): 10-14.
［9］尹忠科,邵君, Vandergheynst P. 利用FFT實(shí)現(xiàn)基于MP的信號(hào)稀疏分解［J］. 電子與信息學(xué)報(bào), 2006,28 (4): 614-618.
Yin Zhongke, Shao Jun, Vandergheynst P. MP based signal sparse decomposition with FFT［J］. Journal of Electronics & Information Technology , 2006,28 (4): 614-618.
［10］Neff R, Zakhor A. Very low bit-rate video coding based on matching pursuit. IEEE Transactions Circuits and Systems for Video Technology, 1997,1(1):158-171.
［11］Chen S, Donoho D, Saunders M. Atomic decomposition by basis on matching pursuit［J］.SIAM Journal Inform Theory, 1992,38:713-718.

服務(wù)與反饋：

【文章下載】【加入收藏】

提示：您還未登錄，請(qǐng)登錄！點(diǎn)此登錄

51黑料吃瓜在线观看,51黑料官网|51黑料捷克街头搭讪_51黑料入口最新视频