Technical report | Implementation of the Scale Invariant Feature Transform Algorithm in MATLAB®

Abstract

This report addresses the description and MATLAB implementation of the Scale-Invariant Feature Transform (SIFT) algorithm for the detection of points of interest in a grey-scale image. Some illustrative simulations for code verification are conducted.

Executive Summary

The most important problem in Computer Vision is to detect an object from its images taken from various positions and at variable illumination. The only way to recognize an object from its images, some of which may play the role of training images, is to associate points of interest to which distinctive features can be assigned and matched between different images.

The matching procedure will be successful only if the extracted features are nearly invariant to scale and rotation of the image.

This report addresses the description and MATLAB implementation of the Scale-Invariant Feature Transform (SIFT) algorithm for the detection of points of interest in a grey-scale mage. Some illustrative simulations for code verification are conducted.