software for data presentation, statistical analysis, marketing and prediction.

Free download:
LeoStatistic.zip or
(selfextracting winzip file)


  • Introduction
  • Data
  • Statistics
  • Results presentation
  • Samples
  • Popular statistics and data analysis
  • Image palette analysis.

    One of the most recognizable characteristics of materials is their color. The procedure of comparison of color of probe with standard sample looks like trivial. One can place one above the other and observe with naked eye is colors are differ. Quality answer of our brain will be very accurate in vast number of cases. But let presume that we need apply a deviation of color from standard as a criteria for quality of the product. We will be forced to establish formal criteria of acceptable deviation of colors. That will demand from us to assign  a numerical expression to color.

    Numerical characteristic of color.

    There are several models to present a color of one pixel. Most common in software development is RGB value. One can define an RGB value as normalized reading from three light detectors with red, blue, green filters. One of the most important advantages of the RGB scheme is the fact that it completely consistent with anatomy of our eye. Three different proteins of cone cells permits us to detect preliminary red, blue and green light waves each. In the spite that similarity with RGB concept is not 100% precise but it good enough for any practical consideration. Other popular and useful model is HSV (Hue - direction from white color - the color itself, Saturation - intensity of color, Value - darkness).

    If for one point (pixel) of image one three numbers to characterize a color is enough for area of the sample there are number of points in the area. These points could have or have not the same numerical characteristics of the color. In general case one can use histogram distribution of the RGB or HSV colors along the image to characterize the palette it painted.

    Use LeoStatistic to have color distribution.

    Let analyze the test image painted with three color palette with our application for statistical analysis - LeoStatistic.

    The image  is made by MS Paint with usage of three colors with RGB colors. Backgrounds - (240,240,130) and (220,160,130); text (0,0,128) and saved in three main formats: BMP - , JPG -   and GIF - . They look similar if not identical but in real distributions of colors in them are quite differ.

    Using LeoStatistic one can instantly create histograms of distributions of colors for the picture file in practically any image format. Let's compare these for images above. We will use HSV model for characterization of the color (Hue (color itself) - is violet on histograms, Saturation (intensity of color) - is blue, Value - (brightness) is black.

    Only three main colors on palette with traces of mixed color representing borders between  main colors.  Exact as expected given that size of file is 6454 bytes - practically exact 3 bytes per pixel as it should be to save exact information about every pixel.

    Picture in JPG format shows distribution of colors. Size of file is 1730 bytes represents significant compression of information. But for cost. Is the deviation from initial image a good trade for size of file? Your guess is as valid as anybodies.


    Size of file is - 1476 that is best from three. But represented in image colors shows less then expected result. In sense hue and saturation - characteristics of the color itself we can see almost the same palette as for BMP format with noticeable exception of addition of pure grey color. But values of the color points on existents of shows at least 7 values brushes instead 3 initials in BMP file. The detail explanation of the fact is far from goal of the articles. In general the idea behind gif format is limit palette with given number of colors and to represent color from BMP image with two or more nearby points these together best approximate initial pixel. 

    Screenshots of the LeoStatistic software:
    click on picture to enlarge

    Building histograms
    Building histogram

    Distribution of two variables.
    Distribution of two variables.

    Approximation (constructor style interface).
    (constructor style interface).

    3D view.
    3D view.

    DOW trend.
    DOW trend.

    Signals revealing
    Signals revealing.

    Near neighbors method
    Near neighbors method.

    Harmonic analysis.
    Harmonic analysis.

    Fit with free format formula.
    Fit with free format formula.

    Curve fit of crystal growth rate.
    Curve fit of crystal growth rate.

    Get data from image file.
    Get data from image file.

    Data analysis  Crystal growth simulation  Internet robot  Photoshop and image analyzer  NetCDF editor  Calculator
    Software archive  Expert database  Photo album  Maverick thoughts  Open forum  Search for cheap sale 
    Home  Products  Partners  Service  Contact
    Copyright by LeoKrut