Abstract: A method and apparatus for determining a texture which best matches an image (100) are provided. The method involves determination of an image feature value from the image (100), and determination of corresponding texture feature values for each member of a set of textures (106, 112). Thereafter the image feature value is compared with the corresponding texture feature values of each member of the set of textures (106, 112) to produce comparison values, from which the best comparison value is selected according to a rule. This best comparison value determines the coresponding desired matching texture.

Abstract: A fast temporal alignment estimation method for temporally aligning a distorted video with a corresponding source video in order to perform accurate video quality measurements transforms each video into a corresponding signature curve by calculating a data point for each frame. The data points are calculated as a cross-correlation between a present frame and a subsequent frame in the video. The signature curves are then matched using a normalized cross-correlation algorithm to compute the temporal misalignment between the videos so that the video signals may be temporally aligned. Fast spatial alignment may then be performed and accurate video quality measurements made.

Abstract: In a method for varying one or more print attributes of a print made from a digital image, a print attribute value is computed for the digital image based on a determination of the degree of importance, interest or attractiveness of the image and the resulting print attribute value is used to control a print attribute of the print made from the image. The print attribute value is derived from an emphasis or appeal value, wherein an appeal value is an assessment of each image taken by itself and the emphasis value is an assessment of each image in relation to other images in a group. In a typical embodiment, the print attribute is either the number of prints made of the image, the size of a print made from the image, or the magnification factor used for the image.

Abstract: A method and apparatus is described for detection and deterrence of counterfeiting that permits one to make legitimate color copies without introducing visual artifacts or experiencing substantial processing delays. The scheme detects a characteristic color of frequently counterfeited documents and alters the characteristic color in the copies. The scheme exploits the fact that conversion between color spaces is frequently accomplished using a color Look-Up Table (LUT). Because it represents no computational burden to alter the values in the LUT, we can efficiently modulate slightly the value of the characteristic color in a way that depends on the value of the counter. For example, at first even banknote green is rendered correctly. However, as the page uses more and more of the ink, a modulation function causes conversion to a darker value than the actual green requested.

Abstract: A computer-aided image diagnosis system having (1) a detection section for detecting an abnormal shadow candidate in a radiation image, based on image information representing the radiation image, (2) a switching section for switching a destination where a diagnostic image is output, in accordance with a result of the detection of the abnormal shadow candidate, and (3) an outputting section for outputting the diagnostic image which corresponds to the detection result.

Abstract: The present invention is directed to an imaging apparatus for examining an object having smooth surfaces to determine shape deformations in the surface of object. The imaging apparatus includes an imaging device for obtaining a scanned image of the object to be examined. A reference image of the object is stored in a memory. An image register is coupled to the imaging device and to the memory containing the reference image of the object. The image register stores patch information corresponding to both the reference image and the scanned image. A transformation estimator is coupled to the image register, and provides a transform for registering the scanned image to the reference image. A deformation estimator is coupled to the transformation estimator and to the image register. The deformation estimator is configured to utilize the transform and the patch information to determine shape deformations of the object.

Abstract: An image data compression method invention for compressing image data including pixel data having a specific number of bits per pixel obtained at an image input device, includes: a step in which the pixel data having the specific number of bits are divided into higher-order bit data having a first number of bits and lower-order bit data having a second number of bits; and a step in which at least the higher-order bit data resulting from division are compressed through a specific compression method.

Abstract: Encoding efficiency is enhanced by actively combining a plurality of encoding methods. A picture element memory receives and stores picture elements within a predetermined range. A mode determinator selects an encoding mode based upon a reference picture element. A first encoding section and a second encoding section predicts the value of the encoded picture element, determines whether the prediction is correct, and then encodes the value of the encoding picture element based on the determination result and outputs a codeword for the encoding picture element. An encoding controller selectively operates the first encoding section and the second encoding section based upon one of the specific encoding modes and the other encoding mode other than the specific encoding mode selected by the mode determinator.

Abstract: Apparatus and method for concealing an error in a frame of a video sequence. The video sequence includes a plurality of frames and being encoded as at least two independently-coded signals, each of which represents a sub-set of frames of the video sequence. The method includes receiving said at least two signals, identifying an error in a frame of the video sequence, and concealing the error by predicting corresponding data using at least one frame which is encoded in a signal 10b other than that in which the error is identified.

Abstract: A footwear database system receives precision three dimensional foot image data of imaged feet of human beings, and receives precision footlast 3D image data from footwear manufacturers for storage into databases for automatic determination of footwear sizes. After imaging feet and storing the footlast data, a customer can thereafter remotely select, for example, through catalogs and web sites, a style of footwear and the footwear database system then determines the acceptable and best fitting size for the desired style of the footwear of a respective footwear manufacturer, and in so doing provides an automatic best fitting size selection for reducing footwear returns due to poor and uncomfortable fits.

Abstract: In a method of histogram equalization of an image signal (Y), a histogram of the image signal (Y) is determined (HC), the image signal (Y) is processed (HP) in dependence upon the histogram to obtain a modified signal (Ymod), and the modified signal (Ymod) is enhanced (E) where said modified signal (Ymod) falls below said image signal (Y). Preferably, the enhancement (E) is color saturation (RGBsat) dependent.

Abstract: Data compression apparatus for compressing an input data unit to produce an output data unit in accordance with a desired data quantity of the output data unit, the degree of compression being determined by a compression control variable having a range of possible values. A trial encoder compresses successive sections of the input data unit in accordance with a trial group of two or more values of the compression control variable. A data quantity detector detects the trial data quantities generated by each trial encoding of the sections of the input data unit and selects a base compression control variable applicable to the input data unit on the basis of the detected trial data quantities, in order to comply with the desired data quantity.

Abstract: An IR image input section inputs an IR image of printed matter P1, using IR light having a near-infrared wavelength. An edge emphasizing section executes edge emphasizing processing on the IR image. A fold/wrinkle extracting section extracts pixels corresponding to a fold or a wrinkle from the edge-emphasized image, counts the number of the extracted pixels, measures the average density of the extracted pixels obtained when the IR image is input thereto, and outputs the number and the average density of the extracted pixels as feature quantity data. A determining section determines the soil degree of the printed matter P1 due to a fold or a wrinkle on the basis of the feature quantity data.

Abstract: The invention relates to a coding apparatus for embedding second data into first data without deteriorating the first data and a decoding apparatus for decoding coded data into original first data and second data without deteriorating those data. The invention provides a coding apparatus which determines change subject data and change data based on a frequency distribution of at least partial data of the first data, and embeds second data into first data by causing a change from the change subject data to the change data according to the second data. The invention also provides a decoding apparatus which determines change subject data and change data based on a frequency distribution of at least partial data of coded data in which second data is embedded, and decodes the coded data into original first data and restores the second data that is embedded in the coded data by restoring the change subject data from the change data.

Abstract: A binarization method for gray address images which combines high quality and high speed. The method is designed specifically for efficient software implementation. Two binarization approaches, localized background thresholds and Laplacian edge enhancement, are combined into a process to enhance the strengths of the two methods and eliminate their weaknesses. The image is divided into tiles, making binarization decisions for each tile. Tile decisions are modified based on adjacent tile decisions and then the tiles are binarized. Binarization of pixels is performed by performing background thresholding and edge detection thresholding. Only pixels exceeding both thresholds are selected as “on”.

Abstract: This invention converts given color label string data into another color label string data without requiring any original image. For this purpose, a color label string input unit (2010) inputs a color label string pertaining to a source image and its attribute information. A conversion attribute destination unit (2020) designates attribute information that pertains to labeling of a destination color space. A color space matching unit (2030) attains color space matching for the input color label string on the basis of the input attribute information and designated attribute information. A block segmentation matching unit (2040) attains matching with destination blocks, and a color label string output unit (2050) generates and outputs a destination color label string.

Abstract: A scalable predictive coder in which the current frame of data is predicted at the enhancement-layer by processing and combining the reconstructed signal at: (i) the current base-layer (or lower layers) frame; and (ii) the previous enhancement-layer frame. The combining rule takes into account the compressed prediction error of the base-layer, and the parameters used for its compression.

Abstract: An object recognition system includes one or more image sensors and a controller that is adapted for measuring the distance from the system to objects with respect to respective windows of an image captured by the sensors is provided. The controller forms clusters by uniting adjacent windows that have similar measured distances. The system includes a memory for storing data on a previously recognized physical object. The controller infers a physical object based on data on a previously recognized physical object and the speed of the vehicle relative to a previously recognized physical object and wherein said controller determines a combination of clusters that best matches the inferred physical object. The controller groups the clusters into one or more cluster groups according to the distance from the vehicle carrying the system and selects from one of the cluster groups those clusters that overlaps with a physical object inferred by the controller.

Abstract: Improved data embedding and extracting techniques provide a way to embed and extract messages in text sections of documents during copying. Extracted text pixels are grouped together to form text lines of the document. From this formation, a document layout is constructed that is used to embed the message in the text pixels. Each text line is partitioned into blocks, and those of which contain a certain threshold percentage of text pixels are identified as valid. Each valid block is used to embed one bit of information by labeling text pixels of that block with a certain predetermined color. The embedding of bits in valid blocks in a particular text line is done in a column-wise raster order. Only one message character (which may be comprised of multiple bits) is embedded in a particular text line, although that character may be embedded multiple times in the same line if there are enough valid blocks.

Abstract: An apparatus for embedding information in a signal includes a band dividing device, a block divider, a quantization portion, a signal replacement portion, a mean difference addition portion, a mean calculation portion, and a band synthesis portion. The band dividing device divides the signal into transform coefficients over a plurality of frequency bands. The block divider divides one frequency band into a plurality of blocks in accordance with a previously determined block size. The quantization portion calculates for each block, a mean value M of the transform coefficients in the block, and subjects the mean value M to linear quantization, using a previously determined quantization step-size Q to calculate a quantization value. The signal replacement portion replaces the quantization value for each block, on the basis of the quantization value and the value of the information to be embedded.