【图像去噪】QAB-去噪（matlab实现）

大家好，欢迎来到IT知识分享网。

 💥💥💞💞欢迎来到本博客❤️❤️💥💥

🏆博主优势：🌞🌞🌞博客内容尽量做到思维缜密，逻辑清晰，为了方便读者。

⛳️座右铭：行百里者，半于九十。

📋📋📋本文目录如下：🎁🎁🎁

目录

💥1 概述

📚2 运行结果

🎉3 参考文献

🌈4 Matlab代码实现

---

💥1 概述

QAB（Quantitative Analysis of Biosystems）是一种用于生物系统定量分析的方法。在QAB中，图像去噪是一个重要的步骤，旨在减少或消除图像中的噪声，以便更准确地分析图像中的生物系统结构和功能。图像去噪是通过应用各种数字信号处理技术来实现的，这些技术可以从图像中识别和过滤掉噪声，同时尽可能地保留有用的信息。常用的图像去噪方法包括均值滤波、中值滤波、高斯滤波、小波变换等。在QAB中，图像去噪的目标是使图像更清晰、更易于分析，并提高后续分析步骤的准确性和可靠性。去噪后的图像可以更好地用于定量分析生物系统的各种特征，例如细胞形态、蛋白质表达水平、细胞器分布等。

📚2 运行结果

主函数部分代码：

close all clc; clear; %load data load('sample_image.mat') %size of signal [M,N] = size(I); NN = N^2; %generate the noise SNR = 15; pI = sum(I(:) .^2) / length(I(:)); B = randn(N,N) .* sqrt(abs(I)); % Poisson noise pB_tmp = sum(sum(B .^2)) / NN; B = B / sqrt(pB_tmp) * sqrt(pI * 10^(- SNR / 10)); J = B + I; %noisy image % Data Ms = 20; % number of iteration in reconstruction pds = 3; % value of palnck's constant sg = 7.5; % Gaussian Variance (smoothing) %image denoising using QAB [I_result] = image_denoising_QAB(I,J,Ms,pds,sg); % Here one may think why we need I as input, since I is the clean image. % Actually in the program, the clean image data I is used just to find out the % best thresholding hyperparameter. In the code, one can see that the clean % image I is only used for the computation of signal-to-noise-ratio (SNR) % and based on this one can tune the best thresholding hyperparameter. % So, one does not need any knowledge about the clean image I for computing % the quantum adaptive basis (QAB) and can tune the thresholding hyperparameter % manually up to their best possible values and modify the code accordingly. % Therefore, one should not get confused after seeing I as an input, because I % does not have any impact on the denoising process. % Calculation of SNR, PSNR and SSIM pnB = sum((I_result(:) - I(:)) .^2) / length(I(:)); SNR_end = 10 * log10(pI / pnB); PSNR_end = calc_PSNR(I,I_result); SSIM_end = ssim_index(I_result, I, [0.01 0.03], fspecial('gaussian', 3, 1.5), max(I(:))) % Print data fprintf('\n\n OUTPUT:\n SNR = %.2f, PSNR = %.2f, and SSIM = %.2f \n\n',SNR_end,PSNR_end,SSIM_end); %plot figures font_size = 12; 

🎉3 参考文献

文章中一些内容引自网络，会注明出处或引用为参考文献，难免有未尽之处，如有不妥，请随时联系删除。

[1]郭江龙,蒋庆,曹松晓等.基于图像处理的微通道流道板表面缺陷检测方法研究[J/OL].电子测量与仪器学报:1-9[2024-03-27].http://kns.cnki.net/kcms/detail/11.2488.TN.20240325.0943.007.html.

[2]张文淇,张浩,牛志杰等.一种基于改进Shallow-UWnet的浑浊水体图像增强方法[J/OL].光电子·激光:1-10[2024-03-27].http://kns.cnki.net/kcms/detail/12.1182.O4.20240321.1604.028.html.

🌈4 Matlab代码实现