提出一个小波域多尺度马尔柯夫随机场模型用于模拟视觉系统在图像分割中的若干功能。针对人类视觉系统具有特征检测器、等级层次性、双向连续性、学习机制等功能，对输入场景，该模型用小波变换提供该场景图像的稀疏表示，模拟特征检测器功能 用金字塔结构模拟等级层次性 用两类信息流模拟双向连接性，分别刻画自底向上的输入图像特征提取过程以及自顶向下的反馈过程 用迭代过程模拟学习机制 采用多尺度马尔柯夫随机场模型实现图像分割。-Put forward a wavelet domain multi-scale markov random field model used to simulate the visual system in image segmentation of some function. According to human visual

Sleep system has feature detector, grade level, two-way continuity, learning mechanism, and other functions, to input scene, this model using wavelet transform provides the scene image sparse said, simulation feature detector function　Use the pyramid structure simulation grade level　Use two kinds of information flow simulation two-way connectivity, respectively depict bottom-up input image feature extraction process and the top-down feedback process　Using iterative process simulation study mechanism　Using multi-scale markov random field model to realize image segmentation.