Binary image feature extraction

I want to perform feature based image registration on two binary images.

What feature extraction algorithm should i use, to extract sufficient features from the binary images.

Answer 1

[Gautam]

Hello Nisha,

The common feature extraction methods used for binary images are:

1. SIFT: Although traditionally used for grayscale images, it can be applied to binary images after some preprocessing. Use the MATLAB function “detectSIFTFeatures” to implement this.
2. ORB: It can be particularly effective for binary images. USE the function “detectORBFeatures” to implement this.
3. BRISK: Another good algorithm that’s good for binary images. Use “detectBRISKFeatures” for its implementation.

Here are the links to the functions for more information on them

1. detectSIFTFeatures: https://www.mathworks.com/help/vision/ref/detectsiftfeatures.html
2. detectORBFeatures: https://www.mathworks.com/help/vision/ref/detectorbfeatures.html
3. detectBRISKFeatures: https://www.mathworks.com/help/vision/ref/detectbriskfeatures.html