Can anybody explain me how OPENCV fisherfaceclassifier works?
How it recognize the difference faces ?
It's important to understand that all OpenCV algorithms (usually) are based on a research papers or topics that can be researched and understood.
For FisherFace you can read about the background of it here to understand exactly how it works, this article discussed the background and implementation.
As for implementation of the Fisherface using (as your question is a little unclear) OpenCV provide a tutorial on how to get up and running, they also explain some background and importantly link out too all the underlying papers and research
I didn't really think this question is appropriate for SO as you really could have just googled your question and come to the same answer. I just thought it might be useful to understand how OpenCV is built up.
Related
Can anyone suggest me a good source to learn?
I am a newbie in ML
As I am a newbie, I have not done anything in this.
This might be an excellent place to start. You can create a new kernel straight from the dataset page, and the data will be ready for you when you enter the kernel. You can also look at other people's kernels who have used that dataset, and I bet you'll find plenty of helpful examples.
You'll get lots of hate for asking this kind of question, since it doesn't fit in S.O. question parameters, but I prefer to be a useful human.
I am trying to Code a genetic algorithm in Matlab but really dont know how it works in images and how to proceed? Is there any basic tutorial that can help me understand how to apply GA on images (starting from 2d to multidimentional images ).
That will be a great help for me.
Thanking everyone in anticipations.
Kind Regards.
For GA you need two things: a fitness function that can evaluate any solution and tell how good it is, and a representation of your solution so that you can do crossover and mutation. Once you have these, you are good to go. I'm not an expert on image processing so I can't help you with that exactly.
Look at the book Essentials of metaheuristics which is a very good resource for start with evolutionary computation (and not only that) in general. It's free.
There is a paper on this subject which you can find at the IEEE library. I believe it solves the problem you vaguely describe.
I am trying to use ORB descriptors with LshMatcher for a faster matching.
I have found somewhere LSH implementations (example: https://code.ros.org/trac/wg-ros-pkg/browser/branches/trunk_diamondback/stacks/object_recognition_experimental/rbrief/src/lsh.cpp)
But it seems it is not implemented yet in opencv 2.4.2.
Do you have any hint how to include LshMatcher within opencv?
I have asked the same question on the OpenCV dev forum, without a good answer.
http://answers.opencv.org/question/503/how-to-use-the-lshindexparams/
Yet, I hope for some more docs. You can just check it again in a few days to see whether there is a new answer.
BTW, if you try to use it with SIFT/SURF/ORB, which are float descriptors, as I know, it will not work LSH are for binary descriptors only.
Edit
It seems to be a bug in OpenCV (2.4.2), as stated in the accepted answer here
http://answers.opencv.org/question/503/how-to-use-the-lshindexparams/
all the pattern-recognition related posts here are dealing with face-, hand-, characters-recognition.
I wonder, is someone has successivly used OpenCV for recognizing cells on a picture from a microscope.
What I am currently able to do, is counting cells using threshold and shape-detection (change threshold, count shapes, discard shapes with invalide size).
My next task is recognizing among about 20 types of cells. It would be very interesting to exchange the experience. May be, the OpenCV is not the proper/overweighted tool for this?
Regards,
Valentin Heinitz
EDIT
An up-vote has drown my attention to this old question again. I completed the task finally with OpenCV. It works good and the tool was approved by FDA last year as a software part of a diagnostic device :-)
Now I think OpenCV was the perfect tool for this, despite I had to implement Haralick-Features myself.
your problem is a machine learning problem. OpenCV offer a few tools for that, SVM would be a good fit for what you are trying to do. I have experience with Kohonen neural networks, which would be a good idea too, if you have a good database of your cells. However, I do not know how good you are in C++, but it is always a good idea to start with matlab, get your algorithm to work, and then rewrite it in C++. To test your ideas quickly, matlab is definitely a better tool, with a lot of ML functions availables.
I am working with OpenCV for a project used for recognition and I had a general question regarding the API and it's terms. I've looked online and couldn't find anything specific to this but I was wondering what the differences were regarding the Discrete Adaboost, Real AdaBoost, LogitBoost, and Gentle AdaBoost. If anyone could direct me to a pros v cons or a general description about these so that I may research which would be useful.
Update
I have added a link to a powerpoint file that goes over the different variations of the Boosting techniques. Hope this hopes someone else out there.
Adaboost powerpoint
Thanks in advance
There isn't really a simple "always use technique X" otherwise there wouldn't be a need for all the others . You really have to understand the details and experiment.
see The opencv discussion and A list of papers and technical summaries