Image-based matching of color and appearance of coatings containing effect pigments, using binary patterns derived from color model properties
Disclosed here is a method of matching color and appearance of coatings containing at least one effect pigment. The method involves: obtaining, for a coating sample to be matched, sample color image data having pixel-specific color model properties; creating local binary patterns based on the pixel-specific color model properties; generating a binary pattern histogram for the local binary patterns, wherein the binary pattern histogram arranges the local binary patterns into N distinct bins to obtain N bin counts, and wherein the binary pattern histogram represents an appearance feature descriptor of the coating sample to be matched; and utilizing the binary pattern histogram for comparison against a previously generated binary pattern histogram that serves as an appearance feature descriptor of a candidate coating specimen.
1. A method of matching color and appearance of coatings containing at least one effect pigment, the method comprising:
obtaining, for a coating sample to be matched, sample color image data comprising pixel-specific color model properties, wherein the sample color image data is based on a digital image of the coating sample to be matched, as captured by a digital image capture system;
creating local binary patterns based on the pixel-specific color model properties;
generating a binary pattern histogram for the local binary patterns, wherein the binary pattern histogram arranges the local binary patterns into N distinct bins to obtain N bin counts, and wherein the binary pattern histogram represents an appearance feature descriptor of the coating sample to be matched; and
utilizing the binary pattern histogram for comparison against a previously generated binary pattern histogram that serves as an appearance feature descriptor of a candidate coating specimen,
wherein the local binary patterns are placed into a bin depending on the positions of 0s and 1s in the local binary pattern relative to each other without regard to the numerical value of the local binary pattern.
2. The method of claim 1 , wherein the utilizing step comprises:
creating a sample histogram vector from the binary pattern histogram, wherein the sample histogram vector is an N-dimensional vector comprising the N bin counts.
3. The method of claim 2 , wherein the utilizing step further comprises:
comparing the sample histogram vector against a specimen histogram vector, wherein the specimen histogram vector is an N-dimensional vector derived from pixel-specific color model properties and corresponding binary patterns obtained for the candidate coating specimen in response to a captured digital image of the candidate coating specimen.
4. The method of claim 3 , wherein the comparing step comprises:
calculating a cosine similarity value from the sample histogram vector and the specimen histogram vector as a measure of an angle between the sample histogram vector and the specimen histogram vector, wherein the cosine similarity value indicates how well the candidate coating specimen matches the coating sample to be matched.
5. The method of claim 3 , wherein the comparing step comprises:
calculating a distance metric from the sample histogram vector and the specimen histogram vector, wherein the calculated distance metric indicates how well the candidate coating specimen matches the coating sample to be matched.
6. The method of claim 1 , further comprising:
calculating a match score for the candidate coating specimen relative to the coating sample to be matched.
7. The method of claim 6 , further comprising:
retrieving a production formulation associated with the candidate coating specimen, when the calculated match score satisfies predetermined matching criteria.
8. The method of claim 1 , further comprising:
receiving the digital image captured by the digital image capture system, the digital image represented by captured color image data that is formatted in accordance with a first color model; and
converting the captured color image data into the sample color image data, wherein the sample color image data is formatted in accordance with a second color model that includes the color model properties.
9. The method of claim 8 , wherein:
the first color model is an RGB color model;
the second color model includes pixel-specific hue values as one of the color model properties; and
the pixel-specific hue values are hue angles within the range of 0 degrees to 360 degrees, inclusive.
10. The method of claim 1 , further comprising:
segmenting the sample color image data to identify sparkle pixels that exhibit visually distinguishable sparkle effects, wherein the creating step only creates local binary patterns for the identified sparkle pixels.
11. The method of claim 1 , wherein at least one of the N distinct bins is established and maintained to group local binary patterns that are associated with similar color and appearance characteristics.
12. At least one non-transitory machine-readable storage medium that stores instructions executable by at least one processor, the instructions configurable to cause the at least one processor to perform a method of matching color and appearance of coatings containing at least one effect pigment, the method comprising:
obtaining, for a coating sample to be matched, sample color image data comprising pixel-specific color model properties, wherein the sample color image data is based on a digital image of the coating sample to be matched, as captured by a digital image capture system;
creating local binary patterns based on the pixel-specific color model properties;
generating a binary pattern histogram for the local binary patterns, wherein the binary pattern histogram arranges the local binary patterns into N distinct bins to obtain N bin counts, and wherein the binary pattern histogram represents an appearance feature descriptor of the coating sample to be matched; and
utilizing the binary pattern histogram for comparison against a previously generated binary pattern histogram that serves as an appearance feature descriptor of a candidate coating specimen, wherein the local binary patterns are placed into a bin depending on the positions of 0s and 1s relative to each other in the local binary pattern without regard to the numerical value of the local binary pattern.
13. The at least one non-transitory machine-readable storage medium of claim 12 , wherein the utilizing step comprises:
creating a sample histogram vector from the binary pattern histogram, wherein the sample histogram vector is an N-dimensional vector comprising the N bin counts.
14. The at least one non-transitory machine-readable storage medium of claim 13 , wherein the utilizing step further comprises:
comparing the sample histogram vector against a specimen histogram vector, wherein the specimen histogram vector is an N-dimensional vector derived from pixel-specific color model properties and corresponding binary patterns obtained for the candidate coating specimen in response to a captured digital image of the candidate coating specimen.
15. The at least one non-transitory machine-readable storage medium of claim 12 , wherein the method performed by the at least one processor further comprises:
calculating a match score for the candidate coating specimen relative to the coating sample to be matched; and
retrieving a production formulation associated with the candidate coating specimen, when the calculated match score satisfies predetermined matching criteria.
16. The at least one non-transitory machine-readable storage medium of claim 12 , wherein the method performed by the at least one processor further comprises the steps of:
receiving the digital image captured by the digital image capture system, the digital image represented by captured image data that is formatted in accordance with an RGB color model; and
converting the captured image data into the sample color image data, wherein the sample color image data is formatted in accordance with a second color model that includes the color model properties.
17. The at least one non-transitory machine-readable storage medium of claim 12 , wherein the method performed by the at least one processor further comprises the step of:
segmenting the sample color image data to identify sparkle pixels that exhibit visually distinguishable sparkle effects, wherein the creating step only creates local binary patterns for the identified sparkle pixels.
18. The at least one non-transitory machine-readable storage medium of claim 12 , wherein at least one of the N distinct bins is established and maintained to group local binary patterns that are associated with similar color and appearance characteristics.
19. A computing system operable to match color and appearance of coatings containing at least one effect pigment, the computing system comprising:
at least one processor; and
at least one non-transitory machine-readable storage medium that stores instructions executable by the at least one processor, the instructions configurable to cause the at least one processor to perform a method comprising the steps of:
obtaining, for a coating that contains at least one effect pigment, color image data comprising pixel-specific color model properties, wherein the color image data is based on a digital image of the coating, as captured by a digital image capture system;
creating local binary patterns based on the pixel-specific color model properties;
generating a binary pattern histogram for the local binary patterns, wherein the binary pattern histogram arranges the local binary patterns into N distinct bins to obtain N bin counts, and wherein the binary pattern histogram represents an appearance feature descriptor of the coating; and
creating a coating histogram vector based on the binary pattern histogram, wherein the coating histogram vector is an N-dimensional vector comprising the N bin counts, wherein the coating histogram vector is usable for comparing color and appearance of the coating against color and appearance of at least one other coating, wherein the local binary patterns are placed into a bin depending on the positions of 0s and 1s relative to each other in the local binary pattern without regard to the numerical value of the local binary pattern.
20. The computing system of claim 19 , wherein the method performed by the at least one processor further comprises the step of:
storing the coating histogram vector in a database associated with the system.
21. The computing system of claim 19 , wherein the method performed by the at least one processor further comprises the steps of:
retrieving a specimen histogram vector for comparison against the coating histogram vector, wherein the specimen histogram vector is an N-dimensional vector derived from pixel-specific color model properties obtained for a candidate coating specimen in response to a captured digital image of the candidate coating specimen; and
comparing the coating histogram vector against the specimen histogram vector.
22. The computing system of claim 19 , wherein the method performed by the at least one processor further comprises the steps of:
calculating a match score for a candidate coating specimen relative to the coating; and
retrieving a production formulation associated with the candidate coating specimen, when the calculated match score satisfies predetermined matching criteria.
23. The computing system of claim 19 , wherein the method performed by the at least one processor further comprises the step of:
segmenting the sample color image data to identify sparkle pixels that exhibit visually distinguishable sparkle effects, wherein the creating step only creates local binary patterns for the identified sparkle pixels.
24. The computing system of claim 19 , wherein at least one of the N distinct bins is established and maintained to group local binary patterns that are associated with similar color and appearance characteristics.