基于噪音受益的快速图像分割算法

Fast image segmentation algorithm based on noise benefit

图像分割是指将一幅图像分解为若干互不交迭的区域的集合。当用已有的改进高斯混合模型于图像分割时,如何加快其分割过程是一个有研究意义的课题。基于最新的噪音受益EM算法,通过人工加噪来加快已有的改进高斯混合模型的收敛速度,从而达到加快图像分割的目的。当添加的噪声满足噪音受益EM定理时,加性噪声加快了EM算法收敛到局部最大值的平均收敛速度。改进的高斯混合模型是EM算法的特例,因此,噪音受益EM定理同样适用于改进的高斯混合模型。实验表明,提出的算法进行图像分割时,其收敛速度明显加快,时间复杂度明显变小。

Image segmentation denotes a process by which a raw image is partitioned into nonoverlapping regions. When using the existing improved Gaussian mixture model in image segmentation, how to speed up its segmentation process is a significant research topic. Based on the latest noise-benefit EM algorithm, this paper speeds up the convergence speed of the existing improved Gaussian mixture model by adding artificial noise, which achieves the goal of speeding up image segmentation. Additive noise speeds up the average convergence of the EM algorithm to a local maximum of the likelihood surface when adding noise to meet the noise-benefit EM theorem. Improved Gaussian mixture model is a special case of the expectation-maximization algorithm, therefore, noise-benefit EM theorem applies to improved gaussian mixture model. Experimental results indicate that the algorithm speeds up the convergence speed when it is used for image segmentation, and the time complexity is decreased significantly.