Abstract: This application describes grouping variables together to minimize cost or time of performing computer vision analysis techniques on images. In one instance, the pixels of an image are represented by a lattice structure of nodes that are connected to each other by edges. The nodes are grouped or merged together based in part on the energy function associated with each edge that connects the nodes together. The energy function of the edge is based in part on the energy functions associated with each node. The energy functions of the node are based on the possible states in which the node may exist. The states of the node are representative of an object, image, or any other feature or classification that may be associated with the pixels in the image.

Abstract: An image file representing at least a portion of a printed document is processed to highlight the differences between foreground material (e.g., text or other characters) from background. The method includes selecting a neighborhood of pixels, determining a weighted average of an attribute values (e.g., luminance) for each pixel, and modifying each pixel's value based on the weighted average. Graylevel scaling, error diffusion, and a bit level conversion are also performed each pixel ends up with either a first attribute value level (e.g., luminance of 0) or a second attribute value level (e.g., luminance of 255).

Abstract: A method for filtering a digital image, comprising segmenting the digital image into a plurality of tiles; computing tile histograms corresponding to each of the plurality of tiles; deriving a plurality of tile transfer functions from the tile histograms preferably using 1D convolutions; interpolating a tile transfer function from the plurality of tile transfer functions; and filtering the digital image with the interpolated tile transfer function. Many filters otherwise difficult to conceive or to implement are possible with this method, including an edge-preserving smoothing filter, HDR tone mapping, edge invariant gradient or entropy detection, image upsampling, and mapping coarse data to fine data.

Abstract: A computer-implemented method includes segmenting a plurality of video frames of a sequence of video frames into a first portion that includes a selected visual object represented in the video frame and a second portion that includes a background represented in the video frame. The selected visual object is selected by using a selection envelope.

Abstract: An image processing method for removing a moving object includes an input step of inputting input images; a matching step of matching the input images according to corresponding positions; a determining step of determining a background image from the input images; a marking step of marking at least one moving object from at least one of the input images; and a replacing step of replacing a region, occupied by the moving object in at least one of the input images with a corresponding regional background in another input image.

Abstract: Disclosed herein are image color matching and equalization devices and related methods. According to an aspect, a method may include determining overlapping portions of a scene within first and second images of the scene. The method may include generating an array of color channel differences between the overlapping portions. Further, the method may include applying a quantization technique to the array of color channel differences for creating a sparse difference color matrix. The method may also include identifying empty locations of the sparse matrix and computing interpolated difference color values to fill them. Further, the method may include modifying the color of at least one pixel of one of the images by performing an inverse look-up in the sparse table utilizing its color, identifying the color difference on the corresponding entry, and applying the color difference on the target pixel.

Abstract: Methods and apparatus for controlling bleeding at the edges of a blur region or selection. A technique for aesthetically controlling the bleeding of blur introduced by blurring selections, referred to herein as a selection bleed technique, may enable continuous adjustment of the amount of bleeding of image blurs between a selected image region and its complement (the unselected region, e.g. the background). The selection bleed technique may allow selections to go from no-bleed to full-bleed and in between via a percentage indicated by one or more user interface elements, for example a slider.

Abstract: Image classification methods and systems are provided. First, an image is obtained using a computer. The image is then processed using the computer to obtain image information. The image information includes one or any combination of an average color difference between at least one average channel value of pixels in at least one boundary region of the image and a predefined standard value, a gradient variation magnitude difference between at least two regions of the image, and a percentage of the edges of the image to the image. The image is classified using the computer according to the image information.

Abstract: A system and method of color data compression may perform variations of MRC compression including taking into consideration means and/or variances within k×k cells of an image, background texture, background smoothed color, and gradient components for determination of which parts of an image belong to the background and foreground MRC layers, for calculating thresholds values for such determinations, and determining correctness of determined thresholds and polarity, and may determine the background and foreground layer colors based on the variances, the gradient components, and neighboring pixels of non-color-assigned pixels of the background and foreground layers.

Abstract: A region of interest extracting section automatically extracts regions of interest from within original images by obtaining the degree by which the features of portions within the original images differ from the features of surrounding portions, based on the colors of the original images, brightnesses of the original images, and the orientations of linear components that appear in the original images. A partial image generating section generates partial images by cutting out images of predetermined sizes that include the regions of interest, which are extracted by the region of interest extracting section, from the original images. The generated partial images are displayed.

Abstract: An information apparatus determines whether a number of color pages in a document data has increased, and when it is determined that the number of the color pages has not increased, page reduction processing is executed on specified pages in the document data, and when it is determined that the number of the color pages has increased, page reduction processing is not executed on the specified pages in the document data.

Abstract: A method of labeling pixels in an image that represent human skin are detected and one or more regions in the image are identified. Each region may include all or a portion of a human face. Pixels that represent each face in the image are identified using the pixels that represent skin and the regions that include faces of the people, thereby identifying a position of each face in the image. A face mask and a torso map corresponding to each face is determined using determined face positions. The extracted face masks and the torso maps are used to refine a skin map. A person or people map is determined using the skin map and the rough body map.

Abstract: A representative pixel value calculation unit (120) divides an image represented by raw data (D0) into a plurality of blocks and calculates representative pixel values (S) for each block. A detection unit (130) detects an object block which is a block including a blown-out highlight or a blocked-up shadow, from among the plurality of blocks. A B-detection unit (134) of the detection unit (130) performs a first detection processing for obtaining one of a difference between a color represented by the representative pixel values and gray, and a value that can approximately represent the difference, as a characteristic value, and for detecting, as the object block, a block in which the characteristic value is equal to or smaller than a first threshold. Consequently, blown-out highlights or blocked-up shadows included in an image can be detected at higher speed, and the omission of detection can be prevented.

Abstract: Methods, apparatus, and computer-readable storage media for dynamic color flow modeling. A dynamic color flow model that incorporates motion estimation in a probabilistic fashion is described. By automatically and adaptively changing model parameters based on the inferred local motion uncertainty, the dynamic color flow model accurately and reliably models the object appearance, and improves the foreground color probability estimation when compared to conventional methods. The dynamic color flow model may, for example, be applied to both foreground and background layers (or to additional layers) for video object segmentation, obtaining significantly improved results when compared to conventional methods.

Abstract: An automatic trapping method and system is used for packaging printing. The method includes translating trapping objects into simple color blocks, scanning all of the blocks and attaining trapping boundaries, determining whether the color blocks on both sides of the trapping boundaries meet the trapping rule depending on the ink dots percentage and luminance, connecting trapping boundaries by analyzing regions, creating trapping zone and filling ink into the zone and reducing color.

Abstract: A method and an apparatus for bilayer image segmentation are described. A set of segmentation seeds for the image is generated by analyzing a depth histogram of the image. Then a segmentation map is generated by minimizing an objective function, which models a directed flow from the foreground segmentation seeds towards the background segmentation seeds.

Abstract: Occupancy detection for building environmental control is disclosed. One apparatus includes at least one image sensor and a controller, wherein the controller is operative to obtain a plurality of images from the at least one image sensor, determine a color difference image of a color difference between consecutive images of the plurality of images, determining an area of the color difference image wherein the color difference is greater than a threshold, create an occupancy candidate based on filtering of the determined area if the determined area is less than light change threshold, generate one or more centroids that represent areas of the occupancy candidate that are greater than a second threshold, track the one or more centroids over a period of time, detect occupancy if one or more of the centroids is detected to have moved more than a movement threshold over the period of time.

Abstract: Long-term understanding of background modeling includes determining first and second dimension gradient model derivatives of image brightness data of an image pixel along respective dimensions of two-dimensional, single channel image brightness data of a static image scene. The determined gradients are averaged with previous determined gradients of the image pixels, and with gradients of neighboring pixels as a function of their respective distances to the image pixel, the averaging generating averaged pixel gradient models for each of a plurality of pixels of the video image data of the static image scene that each have mean values and weight values. Background models for the static image scene are constructed as a function of the averaged pixel gradients and weights, wherein the background model pixels are represented by averaged pixel gradient models having similar orientation and magnitude and weights meeting a threshold weight requirement.

Abstract: A first aspect of the invention relates to a method for creating a binary mask image from an a inputted digital image of a scanned document, comprising the steps of creating a binarized image by binarizing the inputted digital image, detecting first text regions representing light text on a dark background, and inverting the first text regions, such that the inverted first text regions are interpretable in the same way as dark text on a light background. A second aspect of the invention relates to a method for comparing in a binary image a first pixel blob with a second pixel blob to determine whether they represent matching symbols, comprising the steps of detecting a line in one blob not present in the other and/or determining if one of the blobs represents an italicized symbol where the other does not.

Abstract: An apparatus for processing a raster image comprising a plurality of pixels is disclosed. The apparatus comprises a non-transitory computer readable medium for storing steps of a computer program, and a computer processor connected to the non-transitory computer readable medium for executing the steps of the computer program. A plurality of horizontal axial blobs are identified in the raster image, a plurality of vertical axial blobs are identified in the raster image, and the horizontal and vertical axial blobs are linked into a feature blob.

Abstract: A method and system implementing the method for color pixel counting in images by determining color pixels that, upon marking, appear as visibly color or gray to the human eye by applying a 2-dimensional filter and a reverse color lookup process. An area color computation is performed on a window around each pixel of interest to determine average gray level values for output colors (C, M, Y). Gray line values for each output color are determined and the values for at least two output colors are compared to the average gray level values. If average gray level value for both colors is within a tolerance for the gray line, the pixel is determined as visibly gray. Otherwise, if at least one count is not within the tolerance, the pixel is determined as visibly color. The count of visibly color pixels can be used to determine the billable color pixel count.

Abstract: The system and method of use are provided for correcting digital images based on the notions of local black points. The system and method are based on determining local black points in a digital image and correcting the intensities of nearby pixels so as to remove bias in the image introduced by atmospheric conditions. By sampling only localized land imagery the claimed method computes black points that are more robust against the localized effects of atmospheric and land reflection effects than black points computed from wider areas. The result of the method is that images corrected by this method are robust to atmospheric conditions that vary with time and viewing angle.

Abstract: An automated, computerized method is provided for processing an image. The method comprises the steps of providing a sequence of image files depicting images of a same scene, in a computer memory, determining correspondence information relevant to the same scene across the sequence of images, and generating individual intrinsic images for each one of the sequence of images, as a function of the correspondence information.

Abstract: Methods and systems are provided allowing for background identification in video images. A computer-implemented image processing method includes: receiving, using at least one processing circuit, a plurality of image frames of a video; constructing, using at least one processing circuit, a plurality of statistical models of the plurality of image frames at a plurality of pixel granularity levels; constructing, using at least one processing circuit, a plurality of probabilistic models of an input image frame at a plurality of channel granularity levels based on the plurality of statistical models; merging at least some of the plurality of probabilistic models based on a weighted average to form a single probability image; and determining background pixels, based on a probability threshold value, from the single probability image.

Abstract: An image processing apparatus for performing color tone conversion on a predetermined original image data using a predetermined color tone conversion parameter, includes a converted image data generating unit that generates weight-applied converted original image data obtained by performing the color tone conversion by the color tone conversion parameter while applying weighted values, which is determined such that the lower the degree to be changed by the color tone conversion is, the smaller the value becomes, on the original image data for each of a predetermined range.

Abstract: An image processing apparatus includes a first calculating section that calculates a non-contour degree for pixel of a processing target image data, a second calculating section that calculates a thin line degree for pixel of the processing target image data, a region extracting section that extracts a thin line or non-contour region based on the non-contour degree and the thin line degree and a color value extracting section. The color value extracting section extracts a color value within a range of a predetermined number of colors based on a color used in the thin line or non-contour region extracted by the region extracting section in the processing target image data.

Abstract: A method that includes classifying photographs into categories; generating computational models of image aesthetics, each computational model of image aesthetics associated to one different category of the categories; extracting a plurality of features from a photograph, the plurality of features including simplicity features, global features computed in the whole image and/or low-level features in contrasting regions of an image of the photograph, the contrasting regions being partitions of the image obtained by applying an image segmentation algorithm based on feature contrast to the photograph; and applying a computational model of image aesthetics to at least part of the plurality of features extracted from the photograph, the computational model applied according to the category of the photograph, in order to assess aesthetic quality of the photograph.

Abstract: An image processing apparatus includes: a memory; an obtaining unit that obtains image data representing an image including concatenated pixels; an isolating unit that isolates a rendering element, the rendering element being an image surrounded by border lines of a color in an image represented by the image data; and a classifying unit that, in a case where a plurality of rendering elements has been isolated by the isolating unit, and in a case where color difference between two of the plural rendering elements or the distance between the two rendering elements is less than a threshold, classifies the two rendering elements into the same group, associates pieces of image data that represent rendering elements belonging to the same group with one another, and stores the pieces of image data in the memory.

Abstract: A color converting apparatus includes: an input section acquiring image data of moving image content from each frame; an object detecting section detecting an object from the image in each frame; a confusing region extracting section extracting a confusing region having a confusing hue from the detected object; and a color converting section performing color conversion at least on a region including the confusing region in a particular frame or another frame including the object from which the confusing region has been extracted, when the confusing region is extracted from the particular frame.

Abstract: A system and a method of detecting abnormal or security-significant segments of a video are provided. The video is divided into several segments. A set of features of each segment of the video is extracted in order to calculate a set of factors corresponding to each segment. A value deemed abnormal of each segment is calculated according to the set of factors corresponding to each segment. One or more abnormal segments are determined from the segments based on the abnormal values. The set of features includes a color variable feature, a movement variable feature, a movement variable ratio feature and a background variable feature.

Abstract: A method performed by a software process executing on a computer system includes selecting a first set of pixels in a digital image in the RGB color space. The pixels are selected such that, for each pixel in the set, a red component is a highest value component and a blue component is a lowest value component. The method also includes identifying at least a subset of the first set as a region of orange hue in the digital image.

Abstract: An image processing apparatus includes an extraction unit and a calculation unit. The extraction unit extracts a local color displacement that is a local displacement of color in a region of interest in a given image. The calculation unit calculates a similarity between the local color displacement and an extracted-color displacement that is a displacement of a preset color.

Abstract: Systems and methods systems and methods for calibrating field uniformity are disclosed. An exemplary method includes scanning the imaging area including a first media to obtain optical data for a specular reflectance map. The method also includes scanning the imaging area including a second media to obtain optical data for a diffuse reflective map. The method also includes storing the specular reflectance map and the diffuse reflective map for adjusting actual pixel values during an imaging operation.

Abstract: A color chart having a plurality of mirror-image color blocks arranged in a matrix, with each color block a mirror image of an adjacent color block, each color block having rows and columns of color patches, wherein the concentration of a first color of the multiple colors changes unidirectionally across color patches in any given row of a color block, the concentration of a second color of the multiple colors changes unidirectionally across color patches in any given column of a color block, and the concentration of a third color of the multiple colors remains unchanged across all the plurality of color patches within a color block and differs only between color blocks.

Abstract: An image data processing method comprises: extracting a depth map from an input image; separating the input image into a foreground area and a background area based on grayscales of the depth map; and modulating the maximum depth value of the input image to become smaller and the difference between depth values of a gray grayscale segment of the depth map to be greater than the difference between depth values of the foreground area and background area of the input image.

Abstract: A method is described for processing an image to generate a signature which is characteristic of a pattern within the image. The method includes: receiving an image; overlaying a window at multiple locations on the image to define a plurality of sub-images within the image, with each sub-image each having a plurality of pixels having a luminance level; determining a luminance value for each sub-image where the luminance value is derived from the luminance levels of the plurality of pixels; and combining the luminance values for each of the sub-images to form the signature. The combining is such that the signature is independent of the location of each sub-image. A method of creating a database of images using the method of generating signatures is also described.

Abstract: Provided are an apparatus and method for aligning color channels on the basis of depth information on an image taken by an imaging device equipped with a multiple color-filter aperture (MCA). An image divider generates a binary image including edge information on an original image, and generates a label map including a plurality of regions of interest (ROIs) respectively corresponding to a plurality of objects included in the original image. A shifting vector estimator estimates color shifting vectors (CSVs) indicating shift directions and distances of color channels of the original image according to the respective ROIs. A channel matching unit obtains a plurality of matched images corresponding to the respective ROIs included in the label map by shifting the color channels according to the respective CSVs, and generates a multi-focus image by combining the matched images.

Abstract: A method (199) is disclosed for modifying colors of text extracted from an image of a document page to improve visual quality. The document image comprises text and image regions of various colors. The method divides (100) the document image into a plurality of connected components, the connected components comprising pixels of similar color that are closely spatially located, and selects from the image a plurality of the connected components that represent text. The plurality of text connected components are grouped to form a logical structure grouping of said page comprising one or more text blocks wherein each said text block represents a text line or paragraph. One or more logically consistent color sets are then formed (120, 699, 399, 499) for each text block based on the colors of the connected components in the text block. The logically consistent color sets represent all text of visually similar color of the corresponding text line or paragraph.

Abstract: A method and system for mapping colors of an image. The method including the steps of: segmenting an image into a plurality of regions, so that pixels in each of the plurality of regions have a characteristic that meets a predetermined similarity; determining an initial contrast between adjacent regions in the plurality of regions; and transforming an initial color of each of the plurality of regions into a target color; where a target contrast between the adjacent regions in the plurality of regions is equal to or greater than the initial contrast or a difference between the target contrast and corresponding initial contrast is less than a predetermined threshold; and where at least one of the steps is carried out by a computer device.

Abstract: Systems and methods are described for learning visual object cutout from a single example. In one implementation, an exemplary system determines the color context near each block in a model image to create an appearance model. The system also learns color sequences that occur across visual edges in the model image to create an edge profile model. The exemplary system then infers segmentation boundaries in unknown images based on the appearance model and edge profile model. In one implementation, the exemplary system minimizes the energy in a graph-cut model where the appearance model is used for data energy and the edge profile is used to modulate edges. The system is not limited to images with nearly identical foregrounds or backgrounds. Some variations in scale, rotation, and viewpoint are allowed.

Abstract: A method and system for mapping colors of an image. The method including the steps of: segmenting an image into a plurality of regions, so that pixels in each of the plurality of regions have a characteristic that meets a predetermined similarity; determining an initial contrast between adjacent regions in the plurality of regions; and transforming an initial color of each of the plurality of regions into a target color; where a target contrast between the adjacent regions in the plurality of regions is equal to or greater than the initial contrast or a difference between the target contrast and corresponding initial contrast is less than a predetermined threshold; and where at least one of the steps is carried out by a computer device.

Abstract: A method and system for mapping colors of an image. The method including the steps of: segmenting an image into a plurality of regions, so that pixels in each of the plurality of regions have a characteristic that meets a predetermined similarity; determining an initial contrast between adjacent regions in the plurality of regions; and transforming an initial color of each of the plurality of regions into a target color; where a target contrast between the adjacent regions in the plurality of regions is equal to or greater than the initial contrast or a difference between the target contrast and corresponding initial contrast is less than a predetermined threshold; and where at least one of the steps is carried out by a computer device.

Abstract: An image processing apparatus includes a plurality of color conversion tables for color-separation-limited printing. Using a color conversion table for color-separation-limited printing, a piece of input color information corresponding to chromatic color within a specified applying hue range for a specified specific color is converted into a piece of output color information corresponding to color having hue of the specific color, and a piece of the input color information corresponding to color outside of the applying hue range is converted into a piece of the output color information corresponding to gray color. Each of the color conversion tables for color-separation-limited printing is configured such that color values corresponding to saturation and lightness of the output color information are reduced at least in gradation from a maximum saturation part to a low-lightness part.

Abstract: Some embodiments of the invention provide an image-editing process that non-uniformly adjusts at least one particular color attribute (e.g., saturation level) of pixels in an image. Specifically, the image-editing process of these embodiments uniformly adjusts the particular color attribute of pixels with a first set of values and non-uniformly adjusts the color attribute of pixels with a second set of values. In this manner, some embodiments provide a vibrancy process that non-uniformly adjusts saturation levels of an image in order to preserve saturation levels for skin tones appearing within the image.

Abstract: Systems for segmenting human hairs and faces in color images are disclosed, with methods and processes for making and using the same. The image may be cropped around the face area and roughly centered. Optionally, the illumination environment of the input image may be determined. If the image is taken under dark environment or the contrast between the face and hair regions and background is low, an extra image enhancement may be applied. Sub-processes for identifying the pose angle and chin contours may be performed. A preliminary mask for the face by using multiple cues, such as skin color, pose angle, face shape and contour information can be represented. An initial hair mask by using the abovementioned multiple cues plus texture and hair shape information may be created. The preliminary face and hair masks are globally refined using multiple techniques.

Abstract: A character string detection device for detecting a character string including at least one character in an image has a clustering unit that defines at least one cluster including at least a fixed number of pixel groups having similar colors based on color information included in each pixel configuring the image, a clipping unit that partitions the image into a region of the pixel groups belonging to the cluster defined by the clustering unit and a region of other pixel groups, and generating a clipped image excluding the region of the other pixel groups from the image, and a character string detection unit that detects each character in a detection target character string to be detected so as to detect the detection target character string in the clipped image generated by the clipping unit.

Abstract: A system and method to identify pictures in documents. An image representing a page of a document is received. The image is analyzed to identify text objects in the page. A masked image is generated by masking out regions of the image including the text objects in the page. Groups of pixels in the masked image are identified, wherein a respective group of pixels corresponds to at least one picture in the page. When there is one or more groups of pixels, regions for pictures are identified based on the one or more groups of pixels. Metadata tags for the pictures are stored, wherein a respective metadata tag for a respective picture includes information about a respective bounding box for the respective picture.

Abstract: A method and system for generating a cartoon are provided. The method includes receiving a digital image for processing at a server. The method includes executing a color structure coding procedure to produce a color-reduced interim image. The method includes generating a sketch from the digital image with a Difference of Gaussian procedure. The method includes combining the color-reduced interim image and the sketch into a processed cartoon.

Abstract: Various arrangements for recognizing gestures are presented. A plurality of color images of a scene may be received. Each image of the plurality of color images may be processed to make a plurality of processed color images. The processing of an image of the plurality of color images may comprise grouping pixels of the image being processed into a plurality of clusters based on color. A gesture may be identified using movement of a cluster of the plurality of clusters in the plurality of processed color images.

Abstract: The present invention provides an apparatus for processing image data obtained by reading a document and a background image outside the document, and method of controlling the apparatus. The apparatus determines a degree of similarity between a color of the background image and a color of a marginal region of the document from the image data, sets a region extraction parameter based on the determined degree of similarity, and determines a document region by using the region extraction parameter from the image data.