qmriso
Member level 5
I am trying to extract some features from two images in Matlab. The first image is a black and white picture of an apple and the second is of a bat. I'm trying to see if all the combined features can linearly separate the two classes but not sure if my steps are correct.
I used the command regionprops to extract each feature (solidity, circularity, eccentricity).
I put each feature in a matrix Y=[solidity; circularity; eccentricity] for the apple and put the bat features in another variable Z.
That is where I think I went wrong. I'm not sure if I am suppose to put those features into a matrix like that.
Anyways, next I already have a pre-programmed perceptron algorithm that I ran to see if they are linearly separable but it seems like it doesn't reach a conclusion.
Any suggestions?
Thanks!
I used the command regionprops to extract each feature (solidity, circularity, eccentricity).
I put each feature in a matrix Y=[solidity; circularity; eccentricity] for the apple and put the bat features in another variable Z.
That is where I think I went wrong. I'm not sure if I am suppose to put those features into a matrix like that.
Anyways, next I already have a pre-programmed perceptron algorithm that I ran to see if they are linearly separable but it seems like it doesn't reach a conclusion.
Any suggestions?
Thanks!