基于图像特征检测——使用相位拉伸变换（Matlab代码实现）

% test script to test PST function
clc  % clear screen
clear all  % clear all variables
close all   % close all figures
% import original image
Image_orig=imread('lena_gray_512.tif');
% if image is a color image, convert it to grayscale
try
    Image_orig=rgb2gray(Image_orig);
catch
end
%show the original image
subplot(1,2,1)
imshow(Image_orig)
title('Original Image')
% convert the grayscale image do a 2D double array
Image_orig=double(Image_orig);
% low-pass filtering (also called localization) parameter
handles.LPF=0.21; % Gaussian low-pass filter Full Width at Half Maximum (FWHM) (min:0 , max : 1)
% PST parameters
handles.Phase_strength=0.48;  % PST  kernel Phase Strength
handles.Warp_strength=12.14;  % PST Kernel Warp Strength
% Thresholding parameters (for post processing)
handles.Thresh_min=-1;      % minimum Threshold  (a number between 0 and -1)
handles.Thresh_max=0.0019;  % maximum Threshold  (a number between 0 and 1)
% choose to compute the analog or digital edge
Morph_flag = 1 ; %  Morph_flag=0 to compute analog edge and Morph_flag=1 to compute digital edge.
% Apply PST and find features (sharp transitions)
[Edge PST_Kernel]= PST(Image_orig,handles,Morph_flag);
if Morph_flag ==0
    % show the detected features    
    subplot(1,2,2)
    imshow(Edge/max(max(Edge))*3)
    title('Detected features using PST')
else
    subplot(1,2,2)
    imshow(Edge)
    title('Detected features using PST')
    % overlay original image with detected features
    overlay = double(imoverlay(Image_orig, Edge/1000000, [1 0 0]));
    figure
    imshow(overlay/max(max(max(overlay))));
    title('Detected features using PST overlaid with original image')
end
% show the PST phase kernel gradient
figure
[D_PST_Kernel_x D_PST_Kernel_y]=gradient(PST_Kernel);
mesh(sqrt(D_PST_Kernel_x.^2+D_PST_Kernel_y.^2))
title('PST Kernel phase Gradient')