2) La*b* color space
is a color-opponent color space
with dimensions L for lightness and a* and b* for the color-opponent dimensions, based on nonlinearly-compressed CIE XYZ color space
In this method, images are converted into Lab color space
. In order to reduce the influence of luminance, only the average values of data in channels a and b of the VROF region are considered as colored features in this procedure.
. In the vision-based object detection and classification, it is necessary to determine the color space
in which the characteristic of the object appears well.
Table 1: Color Space
Results Method Sensitivity Specificity Accuracy RGB 86.2 100 92 HSV 86.2 100 92 LUV 86.2 100 92 OPP 84.2 95.3 89 LAB 84.7 100 91 Table 2: Texture Features Results Method Sensitivity Specificity Accuracy Texture 49.2 48.6 49 Table 3: Texture Results Method Sensitivity Specificity Accuracy Texture 65.1 79.4 70 + RGB Texture+ 64.2 78.1 69.2 HSV Table 4: Local Features Result Method Sensitivity Specificity Accuracy SURF 89.7 100 94
Let us suppose that we want to classify one image pixel P(x,y,z) from one test image from the FESB MLID dataset in the RGB color space
, so x, y and z are red, green and blue values of that pixel.
B is the transfer matrix from the current color space
coordinates to the base coordinates.
Linear models with 1, 2 and 3 independent variables were tested for each color space
(RGB and HSV).
After this color space
conversion, a pre-determined threshold was applied to the image.
Equation (2) basically is a way of separating the color space
in negative and positive parts and based on that to make isoluminant colors brighter or darker.
Relative compares the extreme highlight of the source color space
to that of the destination color space
and shifts all colors accordingly.
This paper offers a perception-based color space
alternative to the well-known red-green-blue (RGB) color space
and several tools to more effectively convey graphical information to viewers.