Take the 2-minute tour ×
Programmers Stack Exchange is a question and answer site for professional programmers interested in conceptual questions about software development. It's 100% free, no registration required.

I need it for my thesis and for some reason I am having a hard time finding decent books or websites for it. My thesis topic is "Classification of Modern Art Paintings using Machine Learning Approach". My goal is to classify examples of modern art paintings to its respective modern art movement(expressionism, realism,etc..) using machine learning approach.

Also, suggestions and comments about my thesis are greatly appreciated.

share|improve this question
    
This is not an answer, but... cirl.lcsr.jhu.edu/wiki/images/0/09/Project6.pdf You should give a copy of this prior work to your professor for pre-approval, and during your project you should first verify their findings, as well as document every new idea and improvements you tried (whether successful or not). Summarize your final report by stating your new contributions. Good luck. –  rwong Jul 31 '11 at 6:16

4 Answers 4

It looks like a tough subject. Finding a suitable feature space to make your problem solvable is probably going to take some hard work and experimenting. Anyway, challenges are always good! :)

If you provide your background it would help a lot answering your question. Assuming you're new to the subject:

Take a look at libsvm, read the papers, run the example and learn the python bindings. You can also approach the problem using neural networks or other classifiers but at least between my peers that work in research (not many, I must admit) SVM seems more common today.

Since you're not going to do pixel to pixel classification - only the whole image matters, not individual pixels - you can have a rough start by classifying histograms from different color spaces (histograms for R, G, B, H, S, V, mean values, standard deviations). I'd try a Hough transform and other feature extraction techniques to expand the feature space into something broader.

Good luck :)

share|improve this answer

If I understand right, you are trying to classify an image based on properties that will be identified by a machine. This is not much different than an optical character recognition program. Most OCR applications use Neural Networks that are trained to convert images to text. You may want to look into how Neural Netwoks operate and see if it can be practically applied to your task. There are plenty of resources available online for Neural Networks.

share|improve this answer

Pattern Recognition & Machine Learning by Bishop is an excellent text for post graduate level.

There is a sample chapter here.

I'd really recommend you take a look at the sample and try and get hold of the book from a university library, or obviously buy it if you have the funds.

Bishop is a leading expert in the probabilistic/Bayesian approach, so I think it would be very well suited to your classification problem. Some of his published papers may also be relevant to you.

Your project sounds like fun:-)

share|improve this answer

There are a few free sources on machine learning in the web, for example a book about a java based program: Ian H. Witten, Eibe Frank: Data Mining: Practical Machine Learning Tools and Techniques, Third Edition (The Morgan Kaufmann Series in Data Management Systems). And their website.

There is also a Video course on machine learning by the Stanford University, it's pretty tough though (on the course website itself are a few sources).

As I asked my professor about some good sources on machine learning he gave me: Pat Langley - Elements of machine learning.

Last but not least there is a quite useful online course on neural networks by the indian institute of technology, Kharagpur (I haven't found any sources there, but it's very interesting).

share|improve this answer
    
Here's also the website of the first "data mining" book, I couldn't put it in my initial post, because as new user you can only post 2 uris within a post: cs.waikato.ac.nz/ml/weka –  beyeran Apr 29 '11 at 8:30

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.