Capture by chroma differentiation: We can analyze that in an image each color spectrum can be represented by a color code. Color numbers in computing can be found in different formats, depending on their precision. We can have from 1 bit of color, like black and white. We have the normal standards in 8-bit RGB, and those for graphic works from 16 bits and above.
Thus, using 8-bit coding, we have the code (#000000) for black, that is, without the presence of a light signal, all other colors are between (#000000) and white (#ffffff). For example, red would be (#ff0000). Thus, by digitizing the image we would have a matrix with the color spectrum of the object under study, so using this matrix with a programmed system we can develop analyzes of the object in different ways.
The Color Matrix: A standard digitized image is just a matrix with the respective colors of the image. Thus, with mathematical matrix tools, we can modify the image in various ways to suit the objectives of capturing specific data for analysis. Each row and column point of this matrix is defined as a pixel. Thus, a matrix of 10 columns by 10 rows would be a 10x10 pixel image. A standard image from a simple camera is 640x480 pixels. We will use the square pixel as it will facilitate our study.
Turning an RGB image gray: An RGB (red, green, blue) colored image can be changed by modifying the values of the image matrix. We saw that colors can be written as variations of (#ffffff), as the values are in hexadecimal, each value (f) corresponds to (16) in decimal, so another way to determine colors is (16x16,16x16,16x16 ) = (256,256,256) = (red, green, blue).
We can simplify an image by converting different values to an average of equal values, so we would only have to analyze 256 tonal values.
Making a gray image black and white: To make a gray image black and white, you just need to determine the shift value. In other words, a shade of gray goes from 256 (white) to 0 (black), if you determine that the turning point is at point 128, then what is above will be white and what is below will be black. Therefore, a gray image can be easily converted to black and white.
Simplifying the visible spectrum of an object: Once the background of the image where the object is located is simplified to just one color, we can now completely isolate the object from the background in which it is located.
Turning the image to gray, and knowing that the background has a single color, see the image, we have the background color (#00), or (0), so to isolate the object it is only necessary to change any value other than ( #00), to (#ff), and so everything that is white will correspond to the analyzed object.
