Compare ground truth scatter plot with tracker's scatter plot

I am using a particle tracker that outputs coordinates of tracked particles in an image into a matrix. I want to assess the tracker's performance looking at the tracker as a binary classifier. In order to obtain the "ground truth" I marked particles by eye in imageJ and I obtained a matlab matrix with the coordinates of the particles for the "ground truth".

I now want to use Matlab to count the number of true positives, false positives, and false negatives. Is there a function I can use to count these for me?

Example:

%Consider the following scatter plot:

X=[4,7;2,4;7,1;3,2]

Y=[7.1,1.1;3.1,2.1;9,2;4.1,7.1]

scatter(X(:,1),X(:,2),'x')

hold on

scatter(Y(:,1),Y(:,2),'o')

%Imagine that the exes represent the coordinates for the ground truth and the circles represent the coordinates output by the tracker. I would want a function that would %tell me there are 3 true positives (given a maximum tolerable distance I input), 1 false positive, and 1 false negative.

Thank you so much!

Answer 1

[Hrishikesh Borate]

Hi,

From my understanding, you are trying to identify tracker output points in the vicinity of ground truth points given a maximum tolerable distance. This can be done by using rangesearch function.

Upon identifying the points, you can classify and label your data. Once you have labelled points, you can use confusionchart to identify the TP, FP, FN, TN.