一種改進(jìn)的馬爾可夫隨機(jī)場(chǎng)圖象恢復(fù)與分割模型
A MODIFIED VERSION OF MARKOV RANDOM FIELD MODEL FOR IMAGE RESTORATION AND SEGMENTATION
-
摘要: 在現(xiàn)有的馬爾可夫隨機(jī)場(chǎng)圖象恢復(fù)與分割模型中,圖象場(chǎng)能量最低組態(tài)被看成是原始景物的一種最優(yōu)估計(jì)。但在圖象灰度值發(fā)生變化的邊界上,能量最低組態(tài)不對(duì)應(yīng)于原始景物,從而造成恢復(fù)(或分割)誤差。本文對(duì)這類模型作了改進(jìn),利用改進(jìn)的模型給出了一種引入邊界信息的松弛算法,并給出了應(yīng)用該算法對(duì)低信噪比圖象進(jìn)行恢復(fù)處理的計(jì)算機(jī)模擬結(jié)果。Abstract: The current Markov random field models for image restoration and segmentation are discussed. A configuration of the image field is regarded as an optimal estimate of the original scene when its energy is the lowest. However, the lowest energy configuration does not correspond to the scene on the edges, which results in errors of restoration or segmentation. Improvements of the model are made and a relaxation algorithm based on the improved model is presented using edge information obtained by a coarse-to-fine procedure. Some examples are also presented on the application of the algorithm to restoration of noisy images.
-
Gem an S, Geman D. IEEE Trans. on PAMI, 1984, PAM1-6(6):721-741.[2]Besag J. J. Roy. Statist. Soc.B, 1986, 48(3):259-302.[3]Derin H. Elliott H. IEEE Trans. on PAMI., 1987, PAMI-9(1):39-55.[4]匡錦瑜,姚小燕.計(jì)算機(jī)學(xué)報(bào),1991,14(7); 514-522.[5]Hu R, Fahmy M M. Signal Processing, 1992,16(3):285-305.[6]Won C S, Derin H. CVGIY Graphical Models anti Imnge Processing, 1992, 54(4):308-328.[7]Geman D, et al. IEEE Trans. on PAM1, 1990, PAMI-12(7):609-528.[8]Mallat S, Hwang W L. IEEE Trans. on IT. 1992, 1T-38 (2):617-643.[9]郭宇春,匡錦瑜.北京師范大學(xué)學(xué)報(bào)(自然科學(xué)版),1994,30(1): 60-66.[10]Marr D. Hildreth E. Proc. Royal Soc. London B. 1980, 207(1167):187-218.[11]Kundu A. Pattern Recognition, 1990, 13(5):423- 440. -
計(jì)量
- 文章訪問數(shù): 1964
- HTML全文瀏覽量: 104
- PDF下載量: 895
- 被引次數(shù): 0
下載:
