基于免疫克隆選擇算法的圖像分割
Application of Immune Clone Selection Algorithm to Image Segmentation
-
摘要: 圖像分割是圖像處理領(lǐng)域中不可缺少的一個(gè)分支。該文基于閾值分割方法，將免疫克隆選擇優(yōu)化算法應(yīng)用到圖像分割中，提出了一種新的圖像分割算法。詳細(xì)敘述了算法機(jī)理，并對(duì)算法復(fù)雜度進(jìn)行了理論分析以及實(shí)驗(yàn)數(shù)據(jù)比較。在仿真實(shí)驗(yàn)中，將遺傳算法和免疫克隆選擇算法分別獨(dú)立運(yùn)行10次，對(duì)10次得到的閾值以及均值、方差進(jìn)行了比較，并將函數(shù)評(píng)價(jià)次數(shù)作為算法復(fù)雜度的評(píng)價(jià)指標(biāo)。該文算法不僅能夠?qū)D像進(jìn)行準(zhǔn)確的分割，而且在同樣的種群規(guī)模下能夠以較少的迭代代數(shù)和較低的函數(shù)評(píng)價(jià)次數(shù)得到最優(yōu)閾值。仿真結(jié)果表明，該方法應(yīng)用在圖像分割中是可行的、有效的。
-
關(guān)鍵詞:
- 圖像分割;人工免疫系統(tǒng);克隆選擇; 遺傳算法
Abstract: Image segmentation is a significant part in image processing field. Inspired by the threshold-based segmentation methods, a novel algorithm based on immune clone selection and optimal entropy theory is presented in this paper. Immune clone selection algorithm performs not only local but also global search, and has better performance than Genetic Algorithm(GA) in searching for the optimal entropy threshold of images. The algorithm is depicted in detail and the computational complexity is given. In experiments, natural image and SAR image are selected, and the algorithm runs ten times independently and the mean numbers of function values are presented as the evaluation of the algorithm complexity. It shows that the algorithm presented in this paper can find better solutions with small generation and mean numbers of function values. So this method has better performance in stabilization and convergence than GA. Experimental results show that this method is feasible and effective. -
Kapur J N, Sahoo P K ,Wong A K C. A new method of gray level picture thresholding using the entropy of the histogram [J].Computer Vision, Graphics, and Image Processing.1985, 29(2):273-[2]Pal N R, Pal S K. A review on image segmentation techniques. Pattern Recognition, 1993, 26(9): 12771294. .[3]Pun T. A new method for gray-level picture thresholding using the entropy of the histogray[J].Signal Processing.1980, 2(3):223-[4]Yen J C, Chang F J, Chang S. A new criterion for automatic multilevel thresholding[J].IEEE Trans. on Image Processing.1995, 4(3):370-[5]Sahoo P K, Wong A K C. A survey of thresholding techniques[J].Computer Vision, Graphics, and Image Processing.1988, 41:233-[6]焦李成，杜海峰. 人工免疫系統(tǒng)進(jìn)展與展望. 電子學(xué)報(bào). 2003, 31(10): 1540.1548.[7]陳國(guó)良，王煦法等. 遺傳算法及其應(yīng)用. 北京：人民郵電出版社，1999.[8]杜海峰. 免疫克隆計(jì)算與人工免疫網(wǎng)絡(luò)研究與應(yīng)用，博士后研究工作報(bào)告，西安電子科技大學(xué)，2003. -
計(jì)量
- 文章訪問數(shù): 2413
- HTML全文瀏覽量: 117
- PDF下載量: 1232
- 被引次數(shù): 0
下載:
