Saturday, October 13, 2007

HSV Fuzzy Linking

In 2005, Konstantinidis et al proposed the extraction of a fuzzy-linking histogram based on the color space CIE-La*b*. The necessity though, of the transportation of the image from RGB field to CIEXYZ and finally to CIELab field made this method noticeably timeconsuming. HSV color space demands smaller computation power in comparison with CIELAB, because it emerges after a direct transportation of the RGB color space. In [1], a fuzzy system is proposed to produce a fuzzy-linking histogram, which regards the three channels of HSV as
inputs, and forms a 10 bin histogram as an output. Each bin represents a present color as follows: (0) Black, (1) White, (2) Grey, (3) Red, (4) Orange, (5) Yellow, (6) Green, (7) Cyan, (8) Blue and (9) Magenta. The inputs of the system are analyzed as follows: Hue is divided into 8 fuzzy areas. (0) Red to Orange, (1) Orange, (2) Yellow, (3) Green, (4) Cyan, (5) Blue, (6) Magenta, (7) Blue to Red.

S is divided in only 2 fuzzy areas. This channel defines the shade of a color based on white. The first area in combination with the position of the pixel in channel V is used to define if the color is clear enough to be ranked in one of the categories which are described in H histogram, or if it is a shade of white or gray color.

The third input, channel V, is divided in 3 areas. The first one is actually defining substantially when the pixel will be black, independently from the values that gives to the other inputs. The second fuzzy area, in combination with the value of channel S gives the gray color.

For the evaluation of the consequent variables two methods have been used. Initially LOM algorithm (Largest of Maximum) was used. This method assigns the input to the output bin which is defined from the rule that gives the greater value of activation. Next a Multi Participate algorithm was used, which assigns the input to the output bins which are defined from all the rules that are being activated. The experimental results show that the second algorithm performs better.


No comments: