Important Parameters for Image Color Analysis: An Overview - Visual Computing: Scientific Visualization and Imaging Systems - page 83

Fig. 4.4 Computer vision system 9 human vision system

property of the object, lighting, and camera in response to the stimulus received, as
compared by Fig. 4.4 .
Just like the human eye, a digital camera has sensors that convert radiation
(stimulus) into electrical charges, which are then digitalized. The image sensors
used in most digital cameras are charge coupled device (CCD) or complementary
metal oxide semiconductor (CMOS). Each sensor is composed of two filters, one
black and white with sensitivity to different intensities of radiation and another
sensitive to red, green, and blue, called the Bayer filter, deployed in order to make
the acquisition of radiation by the camera as similar as possible to the human eye.
Comparing the spectral distributions of filters used in some models of cameras,
one can observe a difference between models and manufacturers. Comparing the
spectral distribution of the cameras with the distribution of color equalization
functions of Commission Internationale de l'Eclairage—the International Com-
mission on Illumination (CIE) there is a big difference between them, technically
demonstrating the need for calibration and characterization of these when used in
quantitative automatic inspection systems.
The process of image digitalization is generated from the mapping of an analog
image, in which the latter is divided into several points, forming a two-dimen-
sional array f(x, y). Each point of this matrix is called the pixel of the image, and,
due to the color filter used, these two elements are connected with the spatial
resolution of the image's color, which is typically represented by color systems to
enable their representation. Similarly to the human eye, the color system that is
most similar to the image acquisition system is the RGB due to the formation of
the three additive colors red, green, and blue.
Various features can be extracted from the RGB values obtained during the
processing step, including some appearance characteristics such as the color. Other
color systems are also used in the analysis of color of an image from the