Detect the number of Outlier matching point

I am using OpenCV code (in the link below) to do the matching between two images to compare few detection and description algorithms. The matching is going well but I would to detect the number of inlier and outlier matching points. Would you assist me how to do it please?

OpenCV: Feature Matching + Homography to find Objects


you’ve found the tutorial. run the code. inspect the variables. what variables look like they contain your answer?

Thank you.

I have run the code and it works find, I got the matching for both images as you may see in the image below. However, I would to know how to identify the number of good matching points( TP) and number of bad matching points (outlier/TF). This can help me to see what algorithms such as SIFT, SURF, ORB, KAZE etc has higher inlier and less outlier.

secondly, I can calculate Precision and Recall

  • Precision = TruePositives / (TruePositives + FalsePositives)