Application of MRF to Depth Information Estimation of Micro Image Defocus
ZENG Xiang-jin1, HUANG Xin-han1, WU Qian2, WANG Min1
1. Department of Control Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China: 2. Laser Fusion Research Center, China Academy of Engineering Physics, Mianyang 621000, China
Abstract：For the problem of depth information estimation of micro vision image,a defocus characteristic parameter model based on Markov random field(MRF)is presented,which converts the depth information estimation problem of defocus characteristics into the optimization problem of energy function.With iterated conditional mode(ICM)algorithm applied to fulfill optimization and least squares estimate(LSE)algorithm employed to estimate the parameters of initial point,performance of the ICM algorithm is improved and the result of the algorithm is prevented from getting into local optimum.The experiments and simulations prove that the model and algorithm are effective and feasible.
[1] Namboodiri V P,Chaudhuri S.On defocus,diffusion and depth estimation[J].Pattern Recognition Letters,2007,28(3):311～319.
[2] Ziou D,Deschenes F.Depth from defocus estimation in spatial domain[J].Computer Vision and Image Understanding,2001,81(2):143～165.
[3] Deschenes F,Ziou D,Fuchs P.An unified approach for a simultaneous and cooperative estimation of defocus blur and spatial shifts[J].Image and Vision Computing,2004,22(1):35～57.
[4] 李锋,周源华.基于频域的视差估算[J].上海交通大学学报,1999,33(5):516～519.Li Feng,Zhou Yuan-hua.Disparity estimation based on frequency domain[J].Journal of Shanghai Jiaotong University,1999,33(5):516～519.
[5] Pentland A P.A new sense for depth of field[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.1987,9(4):523～531.
[6] 陆明俊,王润生.计算机视觉中的Markov随机场方法[J].电子科学学刊,2000,22(6):1028～1037.Lu Ming-jun,Wang Run-sheng.Markov random field methodology in computer vision[J].Journal of Electronics,2000,22(6):1028～1037.
[7] Besag J.On the statistical analysis of dirty pictures[J].Journal of the Royal Statistical Society B,1986,48(3):259～302.
[8] Rajagopalan A N,Chandhuri S.An MRF model-based approach to simultaneous recovery of depth and restoration from defocused images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1999,2l(7):577～589.
[9] Rajagopalan A N,Chaudhuri S.A variational approach to recovering depth from defocusod images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(10):1158～1164.
[10] Subbarao M.Parallel depth recovery by changing camera parameters[A].Proceedings of the IEEE International Conference on Computer Vision[C].Piscataway,NJ,USA:IEEE.1988.149～155.
