# how to use the SVM algorithm to find the inner points in the job of ellipse fitting

i need to fit an ellipse with many points of noise points. i konw the RANSAC algorithm can do this job. and this algorithm can find the inner points fitting the ellipse and reject the noise points. recently, i have learnt SVM algorithm, consider that if find the appropriate Kernel function, we can do this nonlinear classification. according this kernal function, we can mapped all the point to high-dimensional space, to solve the linear inseparable problem in the original space.

so i want to konw whether the SVM can be used to find the inner points and how to do this job?

or have other method to solve this problem?

this is the image of points including noise used to fit an ellipse

thanks!

I have experience with SVMS and have not a single clue on how to do this with SVMS. That techniques is simply not created for that purpose...

ok, thanks, i find the SVM algorithm cannot do this job. but it can move the noise points and leave the efficient points. for example, leave the points looking like making up an circle in the image above