Image quality defect detection via sensor characterization and halftone dot classification

A system for image quality defect detection on customer prints, the system comprising:

• a memory and a storage medium;
• and
• a processor in communication with said storage medium and said memory, said processor executing machine readable instructions for performing:
• receiving an image of a document printed using a color marking device intended to be analyzed for image quality defects;
• characterizing a resolving power of a sensor used to scan said image such that a deficiency of said resolving power can be determined;
• compensating for a resolving power deficiency of a sensor that generated said image, said compensating comprising:
• determining a response of a de-blurring filter appropriate for compensating for said sensor's resolving power deficiency, said response being determined with respect to a Modulation Transfer Function (MTF) of said sensor;
• and
• applying an image enhancement to said image using said filter to obtain an enhanced image;
• classifying halftone dots on said enhanced image according to colorants present at a corresponding location of said image, said classification comprising:
• comparing a value for each pixel of said compensated image and a plurality of colorant combinations of said color marking device and, in response to said value being within a range of one of said colorant combinations, classifying said halftone dots as one of said colorant combinations, otherwise:
• deriving a plurality of mean colors, where each mean color represents an output signal value of said sensor corresponding to a known colorant combination of said color marking device;
• determining color differences between each pixel of said compensated image and said plurality of mean colors;
• and
• using said color differences to classify each halftone dot based on an index of a mean color that yields a minimum difference;
• and
• analyzing said classified halftone dots of at least one color separation to determine whether said color separation contains an image quality defect.

A method for image quality defect detection on customer prints, the method comprising:

• receiving an image of a document printed using a color marking device intended to be analyzed for image quality defects;
• characterizing a resolving power of a sensor used to scan said image such that a deficiency of said resolving power can be determined;
• compensating for a resolving power deficiency of a sensor that generated said image, said compensating comprising:
• determining a response of a de-blurring filter appropriate for compensating for said sensor's resolving power deficiency, said response being determined with respect to a Modulation Transfer Function (MTF) of said sensor;
• and
• applying an image enhancement to said image using said filter to obtain an enhanced image;
• classifying halftone dots on said enhanced image according to colorants present at a corresponding location of said image, said classification comprising a two stage approach:
• in stage 1 of said two stage approach, comparing a value for each pixel of said compensated image and a plurality of colorant combinations of said color marking device and, in response to said value being within a range of one of said colorant combinations, classifying said halftone dots as one of said colorant combinations using a first label value, otherwise classifying said halftone dots with a second label value:
• in stage 2 of said two stage approach, for each of said halftone dots classified with the second label value,
• deriving a spatially varying RGB→CMYK LUT which relates an output signal of said sensor to a colorant signal of said color marking device;
• and
• using said spatially varying RGB→CMYK LUT to classify said halftone dots;
• and
• analyzing said classified halftone dots of at least one color separation to determine whether said color separation contains an image quality defect.

A system for image quality defect detection on customer prints, the system comprising:

• a memory and a storage medium;
• and
• a processor in communication with said storage medium and said memory, said processor executing machine readable instructions for performing the method of:
• receiving an image of a document printed using a color marking device intended to be analyzed for image quality defects;
• characterizing a resolving power of a sensor used to scan said image such that a deficiency of said resolving power can be determined;
• compensating for a resolving power deficiency of a sensor that generated said image, said compensating comprising:
• determining a response of a de-blurring filter appropriate for compensating for said sensor's resolving power deficiency, said response being determined with respect to a Modulation Transfer Function (MTF) of said sensor;
• and
• applying an image enhancement to said image using said filter to obtain an enhanced image;
• classifying halftone dots on said enhanced image according to colorants present at a corresponding location of said image, said classification comprising a two stage approach:
• in stage 1 of said two stage approach, comparing a value for each pixel of said compensated image and a plurality of colorant combinations of said color marking device and, in response to said value being within a range of one of said colorant combinations, classifying said halftone dots as one colorant combination using a first label value, otherwise classifying said halftone dots with a second label value;
• in stage 2 of said two stage approach, for each of said halftone dots classified with the second label value,
• deriving a spatially varying RGB→CMYK LUT which relates an output signal of said sensor to a colorant signal of said color marking device;
• and
• using said spatially varying RGB→CMYK LUT to classify said halftone dots;
• and
• analyzing said classified halftone dots of at least one color separation to determine whether said color separation contains an image quality defect.