【解包裹】基于加权最小二乘算法实现解包裹附matlab代码

✅作者简介：热爱科研的Matlab仿真开发者，修心和技术同步精进，matlab项目合作可私信。

🍎个人主页：Matlab科研工作室

🍊个人信条：格物致知。

更多Matlab仿真内容点击👇

智能优化算法  神经网络预测 雷达通信  无线传感器

信号处理 图像处理 路径规划 元胞自动机 无人机

⛄ 内容介绍

在移相法测量光学波面或物体形貌过程中,相位解包裹是条纹自动分析中的关键技术,而质量图对解包裹相位算法起着至关重要的作用.用计算机模拟干涉图获得相位导数偏差质量图,指出其在标识相位数据质量方面的不足,并根据调制度结合相位梯度构造出新的质量图(称之为调制度-相位梯度偏差质量图)来弥补此缺陷.再以新质量图数据为权值,采用加权最小二乘解包裹算法验证新质量图的可靠性

⛄ 部分代码

%%%%%%%%%%%%%%%%%%%%% test phase unwrap

% parameters

close all

N = 512;

ampPhase = 20;

noise = 0.5;

[x, y] = meshgrid(linspace(-1,1,N));

%%%%% (1) unweighted case

% original unwrapped phase

phi = exp(-(x.*x+y.*y)/2/0.2^2) * ampPhase + (x + y) * ampPhase/2;

% wrapped phase

psi = wrapToPi(phi + randn(size(phi))*noise);

% unweighted case

abc = tic;

phi2 = phase_unwrap(psi);

disp(sprintf('Unweighted phase unwrap of a %dx%d image takes %f seconds', N, N, toc(abc)));

% show the images

close all;

subplot(2,2,1);

imagesc(phi); title('Original phase');

subplot(2,2,2);

imagesc(psi); title('Wrapped phase with noise');

subplot(2,2,3);

imagesc(ones(N)); title('Weight');

subplot(2,2,4);

imagesc(phi2); title('Unwrapped phase');

%%%%% (2) now test the weighted case

weight = ones(N);

xregion = floor(N/4):floor(N/2);

yregion = floor(N/4):floor(N/2);

weight(yregion, xregion) = 0;

% change the zero-weighted region to noise only

psi3 = psi;

psi3(yregion, xregion) = randn([length(yregion), length(xregion)]);

% now unwrap

bac = tic;

phi3 = phase_unwrap(psi3, weight);

disp(sprintf('Weighted phase unwrap of a %dx%d image takes %f seconds', N, N, toc(bac)));

% show the images

figure;

subplot(2,2,1);

imagesc(phi); title('Original phase');

subplot(2,2,2);

imagesc(psi3); title('Wrapped phase with noise');

subplot(2,2,3);

imagesc(weight); title('Weight');

subplot(2,2,4);

imagesc(phi3); title('Unwrapped phase');

%%%%% (3) test the weighted case (with noise in the border)

weight4 = zeros(N)+0.01;

xregion = floor(N/5):floor(4*N/5);

yregion = floor(N/5):floor(4*N/5);

weight4(yregion, xregion) = 1;

% change the zero-weighted region to noise only

psi4 = randn(size(psi));

psi4(yregion, xregion) = psi(yregion, xregion);

% now unwrap

acb = tic;

phi4 = phase_unwrap(psi4, weight4);

disp(sprintf('Weighted phase unwrap of a %dx%d image takes %f seconds', N, N, toc(acb)));

% show the images

figure;

subplot(2,2,1);

imagesc(phi); title('Original phase');

subplot(2,2,2);

imagesc(psi4); title('Wrapped phase with noise');

subplot(2,2,3);

imagesc(weight4); title('Weight');

subplot(2,2,4);

imagesc(phi4); title('Unwrapped phase');

⛄ 运行结果

⛄ 参考文献

[1]朱勇建, 刘立人, 栾竹,等. 相位解包裹算法中基于调制度的新质量图[J]. 中国激光, 2006.

❤️ 关注我领取海量matlab电子书和数学建模资料

❤️部分理论引用网络文献，若有侵权联系博主删除