Ask Your Question

Ideas for separating LDA clusters to improve SVM accuracy?

asked 2015-08-10 00:30:33 -0500

jackbrucesimspon gravatar image

I've been trying to classify some tags (each tag is a 24*24 pixel image) that I have on the backs of insects. There are 3 tag types: circle (O), rectangle (I) and Queen (Q). There's also a fourth tag type: unknown (U) because sometimes the angle that the tags are extracted at or the lighting mean that it isn't possible to tell what type of tag it is. There's quite a bit of variation in lighting and tag quality too, which you can see in the image below (image 7.png is actually a rectangle that's a little bit blurry although in the image I said it was unknown):

image description

I've tried to use PCA and LDA on this data, LDA initially looked promising when I only trained it with easily lit tags, however once I include all tag types, the LDA looks a little less clear:

image description

If I apply histogram equalisation to the tags, the queen tag seems to separate out reasonably well, and if I train an SVM with two classes (queen vs all others) it's 95% accurate.

image description

I could train two SVMs where the first extracts the queen and then I use the other one to separate out the circles and rectangles. The problem is that there seems to be a fair bit of overlap between the LDA distributions of these shapes. Does anyone have any advice on other image processing, dimensionality reduction or machine learning techniques that I could try to separate these two groups?

edit retag flag offensive close merge delete


How about using non-linear SVM kernels? They should be able to seperate those two clusters by projecting the data to another higher dimension space?

StevenPuttemans gravatar imageStevenPuttemans ( 2015-08-10 02:09:34 -0500 )edit

I've tried using an SVC with RBF kernel but that didn't seem to work any better unfortunately.

jackbrucesimspon gravatar imagejackbrucesimspon ( 2015-08-10 09:52:46 -0500 )edit

1 answer

Sort by ยป oldest newest most voted

answered 2015-08-13 03:02:35 -0500

Tony_Ge gravatar image

I suggest you use feature extractor first instead of raw pixels. feature with LDA are much better than just raw pixels. Have a look of this paper: file:///home/tony/Downloads/nicta_publication_slides_8868.pdf

edit flag offensive delete link more


You link is not working...

StevenPuttemans gravatar imageStevenPuttemans ( 2015-08-13 04:18:51 -0500 )edit

Question Tools

1 follower


Asked: 2015-08-10 00:30:33 -0500

Seen: 1,619 times

Last updated: Aug 10 '15