chsvopencv
I have written the code for histogram and i want to use it for SVM traning. But the fundamental problem is that i don't understand how many minimum number of bins i should choose so that i can get widely varied distribution among bins for different colors (red,green,yellow,blue,orange).
So, can somebody give me the link/post the image of Hue color wheel for openCV. I need something as shown below but the hue range in it is 0-360 and i need a color wheel used by opencv (i.e. 0-180 )
An alternative approach would be to extract features (keypoints) using the scale-invariant feature transform (SIFT) or Speeded Up Robust Features (SURF).
You can find a nice OpenCV code example in Java, C++, and Python on this page: Features2D + Homography to find a known object
Both algorithms are invariant to scaling and rotation. Since they work with features, you can also handle occlusion (as long as enough keypoints are visible).
Image source: tutorial example
The processing takes a few hundred ms for SIFT, SURF is bit faster, but it not suitable for real-time applications. ORB uses FAST which is weaker regarding rotation invariance.
One of the first mathematically defined color spaces is the CIE 1931 XYZ color space. I think that article on Wikipedia should provide some additional insight in the complexity of what you would like to do. I am sure that someone has done this, but I don't believe it's open source.
The closest chart I found is publish on Wikipedia under Web colors article.
Best Answer
in opencv, the hsv image has to fit into 3 8-bit channels, (no problem for S and V [0..255]) .
since H is in the [0..360] range, the actual values get divided by 2, so it fits into a uchar.
just divide any value in the diagram above by 2. ( or multiply the value you get from opencv )