Abstract: What is disclosed is a system and method for determining a pixel classification threshold for vehicle occupancy determination. An IR image of a moving vehicle is captured using a multi-band IR imaging system. A driver's face is detected using a face recognition algorithm. Multi-spectral information extracted from pixels identified as human tissue of the driver's face is used to determine a pixel classification threshold. This threshold is then used to facilitate a classification of pixels of a remainder of the IR image. Once pixels in the remainder of the image have been classified, a determination can be made whether the vehicle contains additional human occupants other than the driver. An authority is alerted in the instance where the vehicle is found to be traveling in a HOV/HOT lane requiring two or more human occupants and a determination has been made that the vehicle contains an insufficient number of human occupants.

Abstract: What is disclosed is a system and method for post-processing a multi-spectral image which has already been processed for pixel classification. A binary image is received which contains pixels that have been classified using a pixel classification method. Each pixel in the image has an associated intensity value and has a pixel value of 1 or 0 depending on whether the pixel has been classified as a material of interest or not. A block of size m×n is defined. Pixel values in a block are changed according to a threshold-based filtering criteria such that pixels in the same block all have the same binary value. The block is then shifted by k pixels and pixel processing repeats until all pixels have been processed. Once all blocks have been processed, contiguous pixels having the same binary value are grouped to form objects. In such a manner, pixel classification errors are reduced.

Abstract: What is disclosed is a system and method for post-processing a multi-spectral image which has already been processed for pixel classification. A binary image is received which contains pixels that have been classified using a pixel classification method. Each pixel in the image has an associated intensity value and has a pixel value of 1 or 0 depending on whether the pixel has been classified as a material of interest or not. The image is divided into a plurality of blocks of pixels. On a block by block basis, pixel values in a block are changed according to a threshold-based filtering criteria such that pixels in the same block all have the same binary value. Once all the blocks have been processed, contiguous pixels having the same binary value are grouped to form separate objects. In such a manner, pixel classification errors in the post-processed binary image can be reduced.

Abstract: A methodology is disclosed to achieve adaptive illumination independent matching of out-of-gamut spot colors. In one embodiment, the methodology includes an iterative process to determine a variety device specific recipes for out-of-gamut spot colors across different illumination spectra and gamut mapping techniques, and then automatically choosing and/or recommending the optimal recipe and gamut mapping technique that provides the lowest color dispersion across a variety of illuminants under consideration.

Abstract: A methodology is disclosed to achieve adaptive illumination independent matching of spot colors. In one embodiment, the methodology includes an iterative process to determine a variety device specific recipes for spot colors across different illumination spectra, and then automatically choosing and/or recommending the optimal recipe that provides the lowest color dispersion across a variety of illuminants under consideration. This approach may be used with a variety of gamut mapping techniques. According to a further embodiment, the methodology may be used with a ray-based gamut mapping method.

Abstract: What is disclosed is a novel system and method for generating a reference color difference quantity guide for spot color print applications. RCDQ values are determined for spot colors obtained from a library of spot colors. A reference color difference quantity guide is created from the spot colors and their respective RCDQ values in a manner more fully disclosed herein. Thereafter, when a user desires to render a job in a particular spot color, the associated RCDQ value for that spot color can be obtained from the reference color difference quantity guide. In various embodiments, recommendations in the form of a suggested printer to use, a media type, a halftone screen, and other meaningful assistance can be provided for selection of spot colors for a given print/copy job that are less sensitive to varying illuminations. The present RCDQ guide provides meaningful extensions in color reproduction in diverse print/copy job environments.

Abstract: What is disclosed is a novel system and method for generating a spectral matching guide for spot color print applications. Spectral matching values are determined for spot colors obtained from a library of spot colors. A spectral matching guide is created from the spot colors and their respective spectral matching values in a manner more fully disclosed herein. Thereafter, when a user desires to render a job in a particular spot color, the associated spectral matching value for that spot color can be obtained from the spectral matching guide. In other embodiments, recommendations in the form of a suggested printer to use, a media type, a halftone screen, and other meaningful assistance can be provided for selection of spot colors for a given print/copy job that are less sensitive to varying illuminations. The present spectral matching guide provides meaningful extensions in spectral color reproduction in print/copy job environments.

Abstract: What is disclosed is a system and method for post-processing a multi-spectral image which has already been processed for pixel classification. A binary image is received which contains pixels that have been classified using a pixel classification method. Each pixel in the image has an associated intensity value and has a pixel value of 1 or 0 depending on whether the pixel has been classified as a material of interest or not. A block of size m×n is defined. Pixel values in a block are changed according to a threshold-based filtering criteria such that pixels in the same block all have the same binary value. The block is then shifted by k pixels and pixel processing repeats until all pixels have been processed. Once all blocks have been processed, contiguous pixels having the same binary value are grouped to form objects. In such a manner, pixel classification errors are reduced.

Abstract: What is disclosed is a novel system and method for generating a color inconstancy guide for use in spot color print applications. In a manner more fully disclosed herein, color inconstancy values are calculated for selected spot colors of interest using a color inconstancy metric. A color inconstancy guide is generated from the calculated color inconstancy values and their respective spot colors. Thereafter, when a user desires to render a job in a particular spot color, the associated color inconstancy value for that color can be obtained from the guide. In various embodiments, recommendations in the form of a suggested printer to use, a media type, a halftone screen, and other meaningful assistance can be provided for spot color selection that are less sensitive to varying illuminations for a given print/copy job. The present color inconstancy guide provides meaningful extensions in color quality and color reproduction in print/copy job environments.

Abstract: A method for calibrating an image printing system that includes a plurality of parallel printers, where each of the printers has a minimum luminance value. The method includes determining a minimum luminance value profile for the plurality of printers in the image printing system based on the minimum luminance values of each of the printers in the image printing system, determining a black point compensation function based on the determined minimum luminance value profile, adjusting luminance values for input image data using the determined black point compensation function, and rendering, with the parallel printers, images on substrates based on the input image data with adjusted luminance levels.

Abstract: What is disclosed is a filter array for a multi-resolution multi-spectral camera system which not only captures 2D images at multiple wavelength bands simultaneously but also at a spatial resolution that meets the demand for spatial feature extraction. The present system optimizes filter bands that provide high image contrast at the highest possible resolution to enable spatial feature extraction and other wavelength bands at lower resolution to achieve maximum number of wavelength bands (e.g. spectral resolution) for multi-spectral analysis. After determining the required spatial resolution and number of wavelength bands for spectral analysis, multiple filters are arranged in a geometric pattern with each filter being designed to have specified wavelength and spatial resolution. Physical sizes of filters differ within each filter group. This maximizes the detector usage while optimizing the trade-off between spatial resolution and spectral resolution.

Abstract: A system and methods for performing UCR/GCR (Undercolor Removal/Gray Color Replacement) in a digital printer is provided. The system and method includes maximizing the linearity of a color mapping function mapping a device independent color to a device dependent color through linear filtering and performing color projection to ensure color accuracy. The system and methods reduces graininess of printed color images having flesh-tones by also including a K reduction step.

Abstract: What is disclosed is a novel system and method for determining the number of objects in an IR image obtained using an IR imaging system. In one embodiment, a total of N intensity values are collected for each pixel in an IR image using a IR imaging system comprising an IR detection device and an IR Illuminator. Intensity values are retrieved from a database which have been estimated for a plurality of known materials, such as skin and hair. A classification is determined for each pixel in the IR image using either a best fitting method of a reflectance, or a correlation method. Upon classification, a total number of objects in the IR image can be determined. The present system and method finds its intended uses in of real world applications such as, determining the number of occupants in a vehicle traveling in a HOV/HOT lane.

Abstract: What is disclosed is a system and method which addresses the problem of inheriting color management intent from a fleet color management LUT into a 4-to-4 cascaded LUT. The LUT is updated starting with a set of patches that either form a uniform or non-uniform grid of the LUT or can be projected to a uniform or non-uniform grid in the CMYK space using, e.g., PCA. A set of L*a*b* target as well a variable that serve the purpose of preserving the GCR and other elements of the fleet profile is obtained by passing the set of CMYK values to the fleet printer model. By iterating on the printer (or the printer model of the engine at the moment of the engine change) and converging to the targets, the engine device cmyk set is obtained. The LUT is formed between the uniform or non-uniform gridded CMYK and the device cmyk.

Abstract: A projection system includes a display apparatus comprising a plurality of tunable Fabry-Perot filters, each of the filters being configured for shifting between a state in which the filter transmits radiation in a bandwidth in the visible range of the electromagnetic spectrum and a state in which the filter transmits radiation in a bandwidth outside the visible range of the electromagnetic spectrum. An illuminator provides light to the plurality of Fabry-Perot filters. A control system receives image data and controls the display apparatus to project an image onto an associated display surface. The control system includes a modulator which provides wavelength modulation signals to the plurality of Fabry-Perot filters to modulate a color of pixels in the image and causes selected ones of the Fabry-Perot filters to shift into the bandwidth outside the visible range to modulate the brightness of pixels in the image.

Abstract: What is disclosed is a novel system and method for determining the number of persons in an IR image obtained using an IR imaging system. The present method separates a human from the surrounding background via a set of particularly formed intensity ratios. Quantities derived from these ratios and threshold values are used to selectively classify whether a pixel in the IR image is from a human or from a non-human. Based upon the classification of the various pixels in the IR image, the number of persons is determined. Various embodiments are disclosed.

Abstract: What is disclosed is a novel system and method of inheriting intended GCR into a 4-to-4 cascaded LUT to compensate for print engine change which adversely impacts the quality of an output print. The present method uses the device link transform to generate starting LUT nodes for the inverse of the drifted printer model by inputting CMYK or RGB into the device link at a color space resolution sufficient to determine output nodes. Lab and CMYK values are derived from this process, and a “starting LUT” is created at the Lab destination nodes. An iterative control begins with this starting LUT to produce an inverse drifted printer model that contains a close approximation for the intended GCR. The present method captures the intent of the device link transform for a fleet and generates the cascaded LUT for drift, specific printer, side1/side2, and other modes leading to special GCRs.

Abstract: This disclosure provides methods, apparatus and systems for generating a profile for a printing device. Specifically, the profile is generated by selecting a GCR profile from a plurality of GCR profiles and mapping a plurality of nodes to a device dependent color space as a function of the selected GCR profiles.

Abstract: Disclosed is an image color management system and method for controlling an image output device. The method for controlling the image output device comprises generating an image output device profile LUT (look-up-table) characterizing the color profile of the image output device for a plurality of drift states associated with the image output device; generating a set of basis vectors representing the LUT; storing the set of basis vectors in an image output device controller; and generating an image output device active profile associated with a current drift state of the image output device to convert image color data for display or printing by the image output device, wherein the image output device active profile is generated from the set of basis vectors.

Abstract: Disclosed is an image color management system and method for controlling an image output device. The method for controlling the image output device comprises generating an image output device profile LUT (look-up-table) characterizing the color profile of the image output device for a plurality of drift states associated with the image output device; generating a set of augmented basis vectors representing the LUT, L*a*b* values and multiple GCR/UCRs; storing the set of basis vectors in an image output device controller; and generating an image output device active profile associated with a current drift state of the image output device to convert image color data for display or printing by the image output device, wherein the image output device active profile is generated from the set of basis vectors.