Abstract: Connected components of dark pixels are clustered from across the image. A “most likely” representative image for each cluster of images is determined, with likelihood determined by a probabilistic model of the image capturing process. An a priori (prior) probability distributions on bitmaps may be used to determine the most likely representative images. For example, a priori probability distributions based on so-called chain codes are implemented. The representative images are used to cluster connected components. Clustering may be repeated. The output page is assembled by replacing each member of a cluster of images by that cluster's representative image.

Abstract: The method sets a region to be subjected to red eye correction for red eye designation when a red eye in an image that has been produced in shooting is corrected into an eye having a pupil of a specified color. This setting method automatically extracts only the red eye from a first region including the red eye which has preliminarily been designated by an operator or automatically and sets a second region to be subjected to red eye correction for correcting a color of the pupil into the specified color of the pupil. If only the red eye can not automatically be extracted, the setting method sets the second region by designating only the red eye by the operator manually. The red eye correcting method converts the second region thus set by the above setting method into the eye having the pupil of said specified color to correct the red eye.

Abstract: The present invention relates to methods and systems for an illuminant compensation. In particular, these methods and systems include a method for operations on an image, for example, an image of a human face. In the described methods and systems, it is determined for each pixel in the image whether it is part of the face region. A surface fitting is then determined based on only the pixels that are determined to be part of the face region. Also, described are methods and systems for image normalization wherein the standard deviation and average for the gray levels of the pixels are determined and then used to normalize the image so that the gray level for each of the pixels falls between a particular range.

Abstract: A method of extracting regions of homogeneous color from a digital picture divides the digital picture into blocks and generates a feature vector for each block as a set of moments of the data for the block. The distance between the feature vector of each block and the feature vectors of the nearest neighboring blocks are determined using either a weighted Euclidean distance metric or a probability mass function-based distance metric. The maximum distance is the gradient value for the block, and the set of gradient values over all the blocks form a color gradient field. The gradient field is digitized and smoothed, and then segmented into regions of similar color characteristics using a watershed algorithm.

Abstract: An image processing apparatus collectively manages a variety of image processings. Among the image processings, there are specially designated image processings. If a combination of specially designated image processings which are not mutually compatible is designated, one of the mutually incompatible image processings is rendered invalid. Therefore, the image processing apparatus is able to carry out image processings under appropriate conditions.

Abstract: A system and method for pattern identification are disclosed. According to one embodiment of the present invention, the method for locating a pattern includes the steps of (1) providing a pattern image corresponding to the pattern to be located; (2) extracting a pattern contour from the pattern image; (3) generating vector information for the pattern contours, relative to a reference point; (4) creating a reference table for storing the vector information, the reference tables corresponding to the pattern contour; (5) providing a scene image, which will be searched for the pattern; (6) extracting a scene contour from the scene image; (7) generating vector information for the scene contours; and (8) determining whether the pattern has been located within the scene image using the reference table and the vector information for the scene contour.

Abstract: An apparatus for examining printed characters is disclosed. The apparatus includes a database storing standard characters, means for inputting and storing a reference character, means for acquiring an actual image of a printed character, means for calculating a similarity degree between the actual image of the printed character and an image of a standard character stored in the database, and means for retrieving a first candidate character having a highest similarity degree to the printed character from the standard characters stored in the database and comparing the first candidate character with the reference character. If the first candidate character retrieved from the database equates to the reference character, the printed character is determined to be correct. If the first candidate character retrieved from the database does not equate to the reference character, the printed character is determined to be incorrect.

Abstract: A background removal apparatus comprises a color normalization module that normalizes an original image, and an image segmentation module that segments the normalized image into several segmented regions. A background mesh generation module simulates the variation of pixel colors in background and generates an interpolated background mesh. A comparison module compares the normalized image and the interpolated background mesh to form a background mask by extracting the coherent regions between them. A refinement module uses a refined rule to determine a final background mask, and a background removal module through which a pure foreground image is obtained. The apparatus combines the efficiency of both color and spatial clustering, and improves the capabilities of current image segmentation method to perform background removal.

Abstract: An implementation of a technology is described herein for recognizing the perceptual similarity of the content of digital goods. At least one implementation, described herein, introduces a new hashing technique. More particularly, this hashing technique produces hash values for digital goods that are proximally near each other, when the digital goods contain perceptually similar content. In other words, if the content of digital goods are perceptually similar, then their hash values are, likewise, similar. The hash values are proximally near each other. This is unlike conventional hashing techniques where the hash values of goods with perceptually similar content are far apart with high probability in some distance sense (e.g., Hamming). This abstract itself is not intended to limit the scope of this patent. The scope of the present invention is pointed out in the appending claims.

Abstract: A method for recognizing an object in an image is disclosed wherein a fractal map of the image is generated by estimating the fractal dimension of each pixel in the image. The fractal map is segmented by thresholding and locations of candidate objects are determined. The pixel value of the image pixel corresponding to the same location where the candidate object is found in the fractal map is compared to a threshold value. If the pixel value is greater than the threshold value, the candidate object is recognized as a valid object.

Abstract: A method classifies images of faces according to gender. Training images of male and female faces are supplied to a vector support machine. A small number of support vectors are determined from the training images. The support vectors identify a hyperplane. After training, a test image is supplied to the support vector machine. The test image is classified according to the gender of the test image with respect to the hyperplane.

Abstract: A system and method for color correcting electronically captured input images utilizes a correlation between the input images and a set of reference images of different media types to determine the media type of the input images. In an exemplary embodiment, the correlation is performed using a media correlation matrix, which includes media characteristic data from the different media types. The media type information of the input images is used to customize the color correction procedure being performed on the input images. The customization of the color correction procedure ensures that the final images are accurate reproduction of the input images with respect calorimetric or perceptual match.

Abstract: A method and structure for determining overall lightness contrast of an image includes extracting pixel values from the image, calculating an image edge contrast based on the pixel values, calculating an area contrast based on the pixel values, calculating image range based on the pixel values, calculating relative average lightness based on the pixel values, and calculating the overall lightness contrast of the image by combining the edge contrast, the image range, the area contrast and the relative average lightness.

Abstract: Disclosed is a method of segmenting one or more objects from one or more backgrounds in an image, the method comprising defining a plurality of image nodes, each said image node corresponding to one or more pixels of said image, connecting pairs of adjacent nodes with n-links, each said n-link weighted with an n-link cost, defining a source node, defining a sink node, defining one or more object seeds, said object seeds corresponding to image nodes within said objects, defining one or more background seeds, said background seeds corresponding to image nodes within said backgrounds, connecting said source node with each said object seed with a plurality of t-links, connecting said sink node with each said background seed with a plurality of t-links, wherein each said t-links is weighted with a t-link cost, and calculating a segmentation cut having the smallest total cost of all cuts separating said source from said sink, wherein said total cost of each said cut is defined as the sum of the costs of all said n-

Abstract: Duplicate images are detected by providing at least two images captured at determinable times; computing an indication of the image content for each image; determining the time of capture of each of the images; and evaluating the indication of image content and the time of capture to determine whether the images are duplicate images. The images are divided into blocks and the indication of image content is computed from a histogram for each block. Thereafter, the step of evaluating the indication of image content and the time of capture comprises comparing one or more blocks of one image, using a histogram intersection metric, to corresponding blocks of another image and using the time difference between capture of the two images to determine whether the images are duplicate images.

Abstract: A low-frequency, temporally asymmetric watermark carrier decreases in luminance in a substantially monotonic manner more slowly than it increases in luminance in a substantially monotonic manner. This allows the carrier to be embedded, e.g., in a video work, in a manner that renders the resultant watermark virtually invisible to the human eye. Moreover, the resultant watermark is substantially resistant to common pirating techniques in which high-frequency visual content (e.g., including watermark data) tends to be filtered out by piracy processing.

Abstract: A technique is provided that identifies screen and non-screen regions of a projected or displayed image to smooth and selectively remove moiré from the screen regions while maintaining sharpness in the non-screen regions. Each pixel in the image is classified as a screen or non-screen pixel and then pixels in a predetermined surrounding area of each pixel are examined to check the classification of that pixel. A low pass filter is applied to pixels in the image, such that, when the low pass filter is applied, one or more pixels covered by the low pass filter are respectively replaced by one or more other pixels covered by the low pass filter based on the examination.

Abstract: A technique is provided that identifies screen and non-screen regions of a projected or displayed image to smooth and selectively remove moiré from the screen regions while maintaining sharpness in the non-screen regions. Each pixel in the image is classified as a screen or non-screen pixel and then pixels in a predetermined surrounding area of each pixel are examined to check the classification of that pixel. A low pass filter is applied to pixels in the image, such that, when the low pass filter is applied, a center of the low pass filter is selectively shifted relative to a current pixel based on the examination.

Abstract: The contrast determination apparatus includes illuminating light sources (12) and an image sensor (26) i.e. a 2-D CCD for capturing a two-dimensional image of e.g. a marked wire (32) on a holder (34), the captured image being sampled over the mark and the background. The sensor includes an optical system (28,30) selected with regard to the character size of the marks so that the field of view of the sensor is sufficient to contain a complete character as well as a generous amount of background. An area of the captured image is designated in which the samples for contrast measurement are to be taken, and a processor (24) takes the pixel values from the image which are thresholded to categorise them as “black”, or “white”, thus forming a two-tone image.

Abstract: A method for performing color dropout on a digitized document, the method comprises the steps of obtaining color values from the digitized document for background and a color of interest to a user; transforming the color values of the background and the color of interest into an identical gray scale value according to a color dropout function; obtaining a gray scale value different from the identical gray scale value for the remaining portion of the image according to the color dropout function; and thresholding the gray scale values obtained from the previous steps to obtain a binary image that substantially eliminates color classification error which retains character integrity by a functional transformation that suppresses color fringe artifacts.