Lab_12

Image matching using SIFT & SURF features

Date: 20.05.2021

Additional Requirements

• Computer Vision Toolbox - SURF features
• VLFeat Toolbox - SIFT features VLFeat Toolbox installation guide for MATLAB

Installation of VLFeat Toolbox

1. Download the binary package of vlfeat to the current directory from VLFeat - Dowload page
2. Unpack the file. Terminal(bash) command to do so on Linux systems is

   tar -xzvf vlfeat-0.9.21-bin.tar.gz

3. Execute the following command in the MATLAB Command Window with the current directory as the workspace directory

   run('./vlfeat-0.9.21/toolbox/vl_setup')
   vl_version versbose

Note: vlfeat-0.9.21 - the version number might be different on basis of future updates to the toolbox. Personal recommendation to follow VLFeat Toolbox installation guide for MATLAB

Files

• ex-12--image-mathcing-using-sift-features-and-surf-features.pdf: Problem Statement for this Lab experiment
• SIFT.m: MatLab code to extract SIFT features for a given image
• SURF.m: MatLab code to extract & visually annotate the SURF feeature & keypoints

Output

Output of SURF.m
Output of SIFT.m

References

• Open file selection dialog box - MATLAB uigetfile
• How to change the window title of a MATLAB plotting figure? - StackOverflow
• How can I customize the text of the figure window title in MATLAB? - MathWorks
• Error: Function IMRESIZE expected input number 2, MAP, to be a valid colormap. Valid colormaps cannot have values outside the range [0,1]. Solution - MathWorks