SURF And SIFT feature extraction

How to display the number of features extracted and features matched using surf and sift algorithm?

Answer 1

[Omega]

Hi Monika,

You can use inbuillt functions from MATLAB's "Computer Vision Toolbox" to extract and match features.

• "detectSURFFeatures" and "detectSIFTFeatures" functions can be used to detect SURF and SIFT features respectively.
• "extractFeatures" function can be used to extract feature descriptors.
• "matchFeatures" function can be used to find matching features.

Here's a reference code to extract and match features using the SURF algorithm. Feel free to tweak it as per your need.

% Read the input images
image1 = imread('scene_left.png');
image2=imread('scene_right.png');

% Convert images to grayscale
grayImage1 = rgb2gray(image1);
grayImage2 = rgb2gray(image2);

% Detect SURF features
points1 = detectSURFFeatures(grayImage1);
points2 = detectSURFFeatures(grayImage2);

% Extract SURF features
[features1, validPoints1] = extractFeatures(grayImage1, points1);
[features2, validPoints2] = extractFeatures(grayImage2, points2);

% Display the number of features extracted
numFeatures1 = size(features1, 1);
numFeatures2 = size(features2, 1);
disp(['Number of features extracted from image 1: ', num2str(numFeatures1)]);

Number of features extracted from image 1: 241

disp(['Number of features extracted from image 2: ', num2str(numFeatures2)]);

Number of features extracted from image 2: 238

% Match features between the images
indexPairs = matchFeatures(features1, features2);

% Display the number of features matched
numMatches = size(indexPairs, 1);
disp(['Number of features matched: ', num2str(numMatches)]);

Number of features matched: 187

To know more, you can refer to the following documentation links:

• https://www.mathworks.com/help/vision/ref/detectsurffeatures.html
• https://www.mathworks.com/help/vision/ref/detectsiftfeatures.html
• https://www.mathworks.com/help/vision/ref/extractfeatures.html
• https://www.mathworks.com/help/vision/ref/matchfeatures.html