EFFECTIVE PIGMENT IDENTIFICATION METHOD, IDENTIFICATION SYSTEM, IDENTIFICATION PROGRAM, AND RECORDING MEDIUM THEREFOR

• US 20070172113A1
• Filed: 01/09/2007
• Published: 07/26/2007
• Est. Priority Date: 01/20/2006
• Status: Active Grant

• Current Assignee: Kansai Paint Company Limited
• Original Assignee: Kansai Paint Company Limited
• Inventors: Arai, Yoshinori, Masuda, Yutaka, Sai, Keisuke
• Granted Patent: US 8,290,275 B2
• Time in Patent Office: Days
• US Class Current: 382/159
• CPC Class Codes: G01J 3/46 Measurement of colour; Colour measuring devices, e.g. colorimeters measuring colour temperature G01J5/60
• CPC Class Codes: G01J 3/463 Colour matching
• CPC Class Codes: G01J 3/504 Goniometric colour measurements, for example measurements of metallic or flake based paints
• CPC Class Codes: G01N 21/25 Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands

Abstract

The method of identifying an effective pigment comprises:

a first step of imaging a target effective pigment to obtain image data thereof;

a second step of subjecting the obtained image data to background processing and extracting image data concerning a region containing one particle of the effective pigment as image data for processing;

a third step of extracting image characteristic parameters from the image data for processing; and

a fourth step of identifying the target effective pigment, based on the image characteristic parameters of the target effective pigment extracted in the third step, using a pre-prepared database that stores information on various kinds of effective pigments in such a manner as to correlate with the image characteristic parameters of various kinds of effective pigments extracted by conducting the second and third steps.

Claims

1. A method of identifying an effective pigment comprising:

a first step of imaging a target effective pigment to obtain image data thereof;

a second step of subjecting the obtained image data to background processing and extracting image data concerning a region containing one particle of the effective pigment as image data for processing;

a third step of extracting image characteristic parameters from the image data for processing; and

a fourth step of identifying the target effective pigment, based on the image characteristic parameters of the target effective pigment extracted in the third step, using a pre-prepared database that stores information on various kinds of effective pigments in such a manner as to correlate with the image characteristic parameters of the various kinds of effective pigments extracted by conducting the second and third steps.

2. The method according to claim 1 wherein the image characteristic parameters include characteristic parameters representing the color and surface condition of the effective pigment.

3. The method according to claim 2 wherein the characteristic parameters representing the color of the effective pigment include sine and cosine values of the dominant color hue angle, and intensity and saturation of the dominant color,

the dominant color hue angle being the mode of the hue angle histogram of the pixels in an image of one particle of the effective pigment in the HSI color space,

the intensity of the dominant color being the average intensity of pixels having the dominant color hue angle, and

the saturation of the dominant color being the average saturation of pixels having the dominant color hue angle.

4. The method according to claim 2 wherein the characteristic parameters representing the surface condition of the effective pigment include the number of constituent colors, the average intensity gray level and the average saturation gray level,

the number of constituent colors being the number of colors that occupy at least a certain proportion of a color-reduced image of one particle of the effective pigment,

the average intensity gray level being determined by converting an image of one particle of the effective pigment into an intensity grayscale image, subjecting the grayscale image to edge extraction, and calculating the average intensity gray level of the pixels in the thus-obtained image;

and

the saturation gray level average being determined by converting an image of one particle of the effective pigment into a saturation grayscale image, subjecting the grayscale image to edge extraction, and calculating the average saturation gray level of the pixels in the thus-obtained image.

5. The method according to claim 2 wherein the image characteristic parameters further include a characteristic parameter representing the particle shape of the effective pigment.

6. The method according to claim 5 wherein the characteristic parameters representing the particle shape of the effective pigment include particle size, circularity, outline condition, and number of notches,

the outline condition being the number of peaks and valleys in a two-dimensional contour profile obtained by plotting the distance from the centroid of one particle of the effective pigment in an image to the outline pixels, in the order of tracing the outline, and

the number of notches being the number of deep valleys in the two-dimensional contour profile.

7. The method according to claim 1 which comprises identifying the target effective pigment using a neural network that uses the image characteristic parameters as input units and information on the effective pigment as output units.

8. The method according to claim 7 wherein the output values of the output units are real numbers, and a predetermined number of effective pigments are selected from the database according to the output values of the output units.

12. The method according to claim 3 wherein the image characteristic parameters further include a characteristic parameter representing the particle shape of the effective pigment.

13. The method according to claim 4 wherein the image characteristic parameters further include a characteristic parameter representing the particle shape of the effective pigment.

9. An effective pigment identification system comprising:

an image-capturing device for imaging the target effective pigment and obtaining image data thereof;

a characteristic parameter detector for subjecting to background processing image data input from the image-capturing device, extracting an image of one particle of the target effective pigment, then subjecting the extracted image of one particle of the target effective pigment to image processing and computing various kinds of image characteristic parameters of the target effective pigment; and

a recording device for storing a database that stores information on various types of effective pigments in such a manner as to correlate with image characteristic parameters calculated by the characteristic parameter detector;

the characteristic parameter detector identifying the target effective pigment, based on the image characteristic parameters calculated from the image of the target effective pigment input from the image-capturing device, using the database.

10. An effective pigment identification program providing:

a first function of subjecting to background processing image data of a target effective pigment input to a computer and extracting image data of a region containing one particle of the effective pigment as image data for processing;

a second function of extracting image characteristic parameters from the image data for processing; and

a third function of identifying the target effective pigment, based on the image characteristic parameters thereof extracted by the second function, using a pre-prepared database that stores information on various types of effective pigments in such a manner as to correlate with image characteristic parameters thereof extracted through the process by the first and second functions.

11. A computer readable recording medium which records the effective pigment identification program of claim 10.