液晶与显示, 2017, 32 (4): 294, 网络出版: 2017-05-03  

改进离散格子玻尔兹曼方法的图像去噪

Image denoising based on modified discrete lattice Boltzmann method
作者单位
浙江越秀外国语学院 网络传播学院, 浙江 绍兴 312000
摘要
为解决生物医学图像处理过程中存在的图像反差低下且噪声污染严重等问题, 本文对离散格子玻尔兹曼方法进行了改进, 提出了一种改进离散格子玻尔兹曼方法的图像去噪算法。算法对分段线性拉伸函数进行构造, 并作为LBM公式的外力因子, 构建演化求解方程, 通过调节变化因子σ, 同步控制图像降噪与反差增强的相关程度, 消除降噪与反差增强之间的对立; 通过优化平衡分布加权系数, 折中扩散方程的求解精度和速度; 将边缘特征和迭代步长嵌入松弛因子中, 实现医学图像非线性扩散的去噪, 解决医学图像处理中的同步降噪与反差增强间的矛盾。仿真实验表明, 该算法与其他降噪算法相比, 在有效抑制噪声的同时, 细节处理更为有效。
Abstract
In order to solve the problem of low image contrast and serious noise pollution in the process of biomedical image processing, a modified lattice Boltzmann method is proposed. The algorithm structures the piecewise linear stretching function, and constructed it as the external force factor of the LBM formula to build the evolution equation. The algorithm adjusts the change factor σto synchronously control the correlation between image noise reduction and contrast enhancement to eliminate the contrast between noise reduction and contrast enhancement. By optimizing the weighting coefficients of the equilibrium distribution, the solution accuracy and speed of the tradeoff diffusion equation are obtained. The edge feature and the iterative step are embedded in the relaxation factor to realize the non-linear diffusion denoising of the medical image and to solve the contradiction between the synchronous noise reduction and the contrast enhancement in the medical image processing. The simulation results show that the proposed algorithm is more effective than other noise reduction algorithms in suppressing noise effectively.

于宁宁. 改进离散格子玻尔兹曼方法的图像去噪[J]. 液晶与显示, 2017, 32(4): 294. YU Ning-ning. Image denoising based on modified discrete lattice Boltzmann method[J]. Chinese Journal of Liquid Crystals and Displays, 2017, 32(4): 294.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!