Distinguishing Colors of Illuminated Objects Using Machine Vision

• US 20100172573A1
• Filed: 01/07/2009
• Published: 07/08/2010
• Est. Priority Date: 01/07/2009
• Status: Active Grant

• Current Assignee: National Instruments Corporation (Emerson Electric Company)
• Original Assignee: National Instruments Corporation (Emerson Electric Company)
• Inventors: Bailey, Michael
• Granted Patent: US 8,320,662 B2
• Time in Patent Office: Days
• US Class Current: 382/159
• CPC Class Codes: G06T 2207/30141 Printed circuit board [PCB]
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]
• CPC Class Codes: G06V 10/462 Salient features, e.g. scale invariant feature transforms [SIFT]

Abstract

System and method for distinguishing colors of illuminated objects using machine vision. A color-balanced image that includes at least one lit area is received, as well as an indication of a region of interest that includes one of the one or more lit areas. A mask image is generated based on the region of interest. A color-balanced image of the region of interest is generated by masking the color-balanced image with the mask image, and a plurality of image attributes for the region of interest is determined by analyzing the color-balanced image of the region of interest. A color is determined based on the plurality of image attributes using a trained classifier, and the determined color stored, e.g., in a memory medium.

Claims

1. A computer-implemented method for distinguishing colors of illuminated objects using machine vision, the method comprising:

receiving a color-balanced image that includes at least one lit area;

receiving an indication of a region of interest, wherein the region of interest includes one of the one or more lit areas;

generating a mask image based on the region of interest;

generating a color-balanced image of the region of interest by masking the color-balanced image with the mask image;

determining a plurality of image attributes for the region of interest by analyzing the color-balanced image of the region of interest;

determining a color based on the plurality of image attributes using a trained classifier; and

storing the determined color.

2. The method of claim 1, wherein said generating a mask image based on the region of interest comprises:

generating a binary image from the color-balanced image;

generating an initial mask image based on the region of interest; and

generating the mask image based on the initial mask and the binary image.

3. The method of claim 2, wherein said generating a binary image from the color-balanced image comprises:

generating a grayscale image from the color-balanced image; and

generating the binary image from the grayscale image based on a specified threshold.

4. The method of claim 1, wherein said analyzing the color-balanced image of the region of interest comprises:

computing a histogram on the color-balanced image of the region of interest; and

determining the plurality of attributes based on the histogram.

5. The method of claim 1,

wherein said determining a color further comprises:

determining a confidence value for the color using the trained classifier,

wherein the confidence value indicates a degree to which the determined color matches the color of the lit area in the region of interest; and

wherein said storing further comprises storing the confidence value.

6. The method of claim 1, wherein the plurality of image attributes comprises attributes with respect to multiple color planes.

7. The method of claim 1, wherein the plurality of image attributes comprises two or more of:

average value, minimum value, maximum, and standard variation for each of red, green, blue, hue, saturation, and intensity color planes.

8. The method of claim 1, further comprising training the classifier, comprising:

receiving a control image, wherein the control image comprises at least one lit area of each color to be classified;

for each color to be classified,

analyzing the lit area of the color; and

determining a respective plurality of attributes for the region of interest with respect to the color based on said analyzing; and

training the classifier with respect to each color using the plurality of attributes and respective identifiers for the colors to the classifier as training input.

9. The method of claim 8, wherein said analyzing the lit area of the color comprises computing a histogram with respect to the lit area, and wherein said determining the respective plurality of attributes for the region of interest based on said analyzing comprises determining the respective plurality of attributes based on the histogram.

10. The method of claim 8, wherein the respective plurality of attributes comprises two or more of:

average value, minimum value, maximum, and standard variation with respect to each of red, green, blue, hue, saturation, and intensity color planes.

11. The method of claim 1, wherein the image comprises an image of an object, wherein the image is received from an image acquisition device in an industrial process, the method further comprising:

using the determined color to analyze the object.

12. The method of claim 11, the method further comprising:

performing one or more operations with respect to the object based on the analysis of the object.

13. A computer readable memory medium that stores program instructions executable by a processor to perform:

receiving a color-balanced image that includes at least one lit area;

receiving an indication of a region of interest, wherein the region of interest includes one of the one or more lit areas;

generating a mask image based on the region of interest;

generating a color-balanced image of the region of interest by masking the color-balanced image with the mask image;

determining a plurality of image attributes for the region of interest by analyzing the color-balanced image of the region of interest;

determining a color based on the plurality of image attributes using a trained classifier; and

storing the determined color.

14. The memory medium of claim 13, wherein said generating a mask image based on the region of interest comprises:

generating a binary image from the color-balanced image;

generating an initial mask image based on the region of interest; and

generating the mask image based on the initial mask and the binary image.

15. The memory medium of claim 14, wherein said generating a binary image from the color-balanced image comprises:

generating a grayscale image from the color-balanced image; and

generating the binary image from the grayscale image based on a specified threshold.

16. The memory medium of claim 13, wherein said analyzing the color-balanced image of the region of interest comprises:

computing a histogram on the color-balanced image of the region of interest; and

determining the plurality of attributes based on the histogram.

17. The memory medium of claim 13,

wherein said determining a color further comprises:

determining a confidence value for the color using the trained classifier, wherein the confidence value indicates a degree to which the determined color matches the color of the lit area in the region of interest; and

wherein said storing further comprises storing the confidence value.

18. The memory medium of claim 13, wherein the plurality of image attributes comprises attributes with respect to multiple color planes.

19. The memory medium of claim 13, wherein the plurality of image attributes comprises two or more of:

average value, minimum value, maximum, and standard variation for each of red, green, blue, hue, saturation, and intensity color planes.

20. The memory medium of claim 13, wherein the program instructions are further executable to train the classifier, wherein training the classifier comprises:

receiving a control image, wherein the control image comprises at least one lit area of each color to be classified;

for each color to be classified,

analyzing the lit area of the color; and

determining a respective plurality of attributes for the region of interest with respect to the color based on said analyzing; and

training the classifier with respect to each color using the plurality of attributes and respective identifiers for the colors to the classifier as training input.

21. The memory medium of claim 20, wherein said analyzing the lit area of the color comprises computing a histogram with respect to the lit area, and wherein said determining the respective plurality of attributes for the region of interest based on said analyzing comprises determining the respective plurality of attributes based on the histogram.

22. The memory medium of claim 20, wherein the respective plurality of attributes comprises two or more of:

average value, minimum value, maximum, and standard variation with respect to each of red, green, blue, hue, saturation, and intensity color planes.

23. The memory medium of claim 13, wherein the image comprises an image of an object, wherein the image is received from an image acquisition device in an industrial process, wherein the program instructions are further executable to perform:

using the determined color to analyze the object.

24. The memory medium of claim 23, wherein the program instructions are further executable to perform one or more operations with respect to the object based on the analysis of the object.

25. A system, comprising:

a processor; and

a memory medium coupled to the processor, wherein the program instructions are executable by the processor to;

receive a color-balanced image that includes at least one lit area;

receive an indication of a region of interest, wherein the region of interest includes one of the one or more lit areas;

generate a mask image based on the region of interest;

generate a color-balanced image of the region of interest by masking the color-balanced image with the mask image;

determine a plurality of image attributes for the region of interest by analyzing the color-balanced image of the region of interest;

determine a color based on the plurality of image attributes using a trained classifier; and

store the determined color.