Computer Vision Project

Using non-maximum suppression, I got 113 good matches with 2 bad matches with 98% accuracy.

Project 2: Local Feature Matching

Interest Points Detection

Algorithm

I used the Harris corner detector algorithm to find interest points of an image. For the Harris corner detector, I would first have patches that when shifting them, it would cause a drastic change. At the implementation level, I used derivatives of gaussian filters to get the horizontal and vertical gradients. Then I filtered the images, and calculated second moment matrix and cornerness scores. Below is the algorithm which I followed.

1) Compute x and y derivatives of image
2) Compute products of derivatives of every pixel
3) Compute the sums of products of derivatives at each pixel
4) Define at each pixel (x,y) the matrix
5) Compute the response of the detector at each pixel

These parameters were tweaked to allow for more accuracy:

gauss = fspecial('gaussian', 4*sigma+1, sigma);


	

	corner = gix2 .* giy2 - gixy .* gixy - alpha * (gix2 + giy2) .* (gix2 + giy2);
	corner = corner .* (corner > threshold * maxvalue);

Non-maximum suppression was also implemented. This method would ignore some of the less significant points. for suppressing some of the unimportant points.

Using non-maximum suppression, I got 215 good matches and 0 bad matches with 100% accuracy.

Without using non-maximum suppression, I got 276 good matches and 4 bad matches with 99% accuracy.

Local Features Description

For finding descriptors, I used the SIFT algorithm by taking the gradients of the image and calculating the magnitude and the orientation at every pixel. SIFT performs really well since it is invariant to scale and orientation.

Feature Matching

For matching features, I simply used Matlab's k nearest neighbor search algorithm with k=2. My code for matching is as follows:


	[index, d] = knnsearch(features1, features2, 'K', 2);
	matches = []; confidences = [];
	for i = 1:size(index, 1)
	    indexes = index(i, :); distances = d(i, :);
	    if distances(1) / distances(2) < 0.76
	        matches = [matches; indexes(1) i]; confidences = [confidences; 1 - distances(1)];
	    end
	end

Results

Analysis