Fusion of color space data to extract dominant color
AssigneeHoneywell International Inc.
InventorsLokesh R. Boregowda, Lalitha M. Eswara
Life of this patent
Estimating the term…
Abstract
A system and a method to extract dominant color information using fusion of color space data. The color space data may include Red, Green, Blue (RGB) color space data and may include Hue (H), Saturation (S), and Intensity Value (V) color space data. Red, Green and Blue color space histograms may be computed and maximum count values of each color space may be obtained, which may help finding ‘shine’ information in the object. Normalized HSV color space may be fused with Normalized RGB color space for dominant color determination. The dominant color decision conditions involved in the fusion process may change based on a ‘shine’ flag.
Claims15
A tangible machine-readable medium storing a sequence of instructions that, when executed by a computer, cause the computer to determine dominant color of object by performing:
- analyzing, via the computer, color space of the object to generate color histograms;
- analyzing, via the computer, color difference of the object to generate normalized color difference data;
- analyzing, via the computer, hue, saturation and intensity value color space of the object to generate hue, saturation, and intensity value sum data;
- and
- fusingtogether, via the computer, the color histograms, the normalized color difference data and the hue, saturation, and intensity value sum data,
- wherein analyzing color space of the object to generate color histograms includes:
- determining components of specific colors of the object;
- and
- transforming the color components to normalized color space for each specific color.
A tangible machine-readable medium storing a sequence of instructions that, when executed by a computer, cause the computer to perform:
- generating normalized red, green, blue color space components of an object;
- generating red, green and blue color histograms of the object;
- generating normalized hue, saturation, and intensity value components of the object;
- and
- determining a dominant color of the object using the generated normalized components and histograms by
- determining a maximum count value and its corresponding position for each histogram;
- determining a mean value for each histogram;
- determining a number of counts greater than mean value for each histogram;
- and
- computing an average value of their positions.
A computer implemented method to identify predominant color of an object in an image, the method comprising:
- using one or more processors to execute instructions retained in machine-readable medium to perform at least some portion of the following:
- extracting red, green and blue color pixels;
- performing a color space analysis based on normalization of the red, green and blue color pixels;
- identifying a first chromaticity factor across pixels based on sum and difference of the normalized red, green and blue color pixels;
- transforming the color components to a color space representing perceptual aspects of the image;
- identifying a second chromaticity factor based on sum and difference of the perceptual aspects;
- determining predominant color of the object based on at least one predetermined condition using the first and second chromaticity factors;
- and
- presenting determined predominant color of the object to a user.
A tangible machine-readable medium storing a sequence of instructions that, when executed by a computer, cause the computer to:
- extract red, green and blue color pixels;
- perform a color space analysis based on normalization of the red, green and blue color pixels;
- identify a first chromaticity factor across pixels based on sum and difference of the normalized red, green and blue color pixels;
- transform the color components to a color space representing perceptual aspects of the image;
- identify a second chromaticity factor based on sum and difference of the perceptual aspects;
- and
- determine predominant color of the object based on at least one predetermined condition using the first and second chromaticity factors.
Dependencies are inferred from the claim text.