Abstract: Techniques for identifying images of a scene including illuminating the scene with a beam of 3 or more wavelengths, polarized according to a determined direction; simultaneously acquiring for each wavelength an image X//(?i) polarized according to said direction and an image X?(?i) polarized according to a direction perpendicular to said direction, X?(?i) being spatially distinct from X//(?i); calculating for each wavelength an intensity image which is a linear combination of X//(?i) and X?(?i), providing an intensity spectrum for each pixel; calculating for each wavelength a polarization contrast image on the basis of an intensity ratio calculated as a function of X//(?i) and of X?(?i), providing a polarization contrast spectrum for each pixel; and calculating a spectro-polarimetric contrast image of the scene, each pixel of this spectro-polarimetric contrast image calculated based on the intensity spectrum and the contrast spectrum of the pixel considered.

Abstract: A method, system and computer program product for detecting presence of an object in an image are disclosed. According to an embodiment, a method for detecting a presence of an object in an image comprises: receiving multiple training image samples; determining a set of adaptive features for each training image sample, the set of adaptive features matching the local structure of each training image sample; integrating the sets of adaptive features of the multiple training image samples to generate an adaptive feature pool; determining a general feature based on the adaptive feature pool; and examining the image using a classifier determined based on the general feature to detect the presence of the object.

Abstract: Disclosed is a driving support display device that composites and displays images acquired from a plurality of cameras, whereby images which are easy for the user to understand and which are accurate in the areas near the borders of partial images are provided. An image composition unit (41) formulates a bird's eye perspective image on the basis of a mapping table reference unit (45) using images of the area around the vehicle acquired from a plurality of cameras (31). An obstacle detection unit (42) detects obstacles in the images acquired by the plurality of cameras (31). On the basis of the obstacle information detected by the obstacle detection unit (42), an obstacle trajectory prediction unit (43) predicts the trajectory of the obstacle and determines the borders so that the borders of the obstacle and the composite image from the plurality of cameras (31) do not overlap.

Abstract: A camera system comprises an image capturing device, object detection module, object tracking module, and match classifier. The object detection module receives image data and detects objects appearing in one or more of the images. The object tracking module temporally associates instances of a first object detected in a first group of the images. The first object has a first signature representing features of the first object. The match classifier matches object instances by analyzing data derived from the first signature of the first object and a second signature of a second object detected in a second image. The second signature represents features of the second object derived from the second image. The match classifier determine whether the second signature matches the first signature. A training process automatically configures the match classifier using a set of possible object features.

Abstract: Systems and methods for tracking human hands by performing parts based template matching using images captured from multiple viewpoints are described. One embodiment includes a processor, a reference camera, an alternate view camera, and memory containing: a hand tracking application; and a plurality of edge feature templates that are rotated and scaled versions of a finger template that includes an edge features template. In addition, the hand tracking application configures the processor to: detect at least one candidate finger in a reference frame, where each candidate finger is a grouping of pixels identified by searching the reference frame for a grouping of pixels that have image gradient orientations that match one of the plurality of edge feature templates; and verify the correct detection of a candidate finger in the reference frame by locating a grouping of pixels in an alternate view frame that correspond to the candidate finger.

Abstract: The position detection system includes a pressure sensor that detects the vertical position of an underwater vehicle, a range sensor unit that detects the relative distances of the underwater vehicle from its surrounding structures; a measurement image acquisition unit that acquires a measurement image of the horizontal plane, an image storage unit that stores images, an image selector that selects one of the stored images that corresponds to the horizontal plane in which the relative distances have been detected, a corresponding-area identification unit that identifies the area in the selected image that corresponds to the measurement image by performing map matching, and a horizontal position calculator that identifies, the pixel that corresponds to the position at which the relative distances have been detected and calculates the horizontal position of the underwater vehicle.

Abstract: A road profile defining apparatus includes an image acquisition unit configured to acquire an image, a lane marking recognition unit configured to extract from the image a left lane marking located at a left end of one lane painted on a road and a right lane marking located at a right end of the lane, and a road profile determination unit configured to output gradient information indicating a gradient change of the road based on directions of the left and right lane markings.

Abstract: A displacement detection method includes the steps of: acquiring an image frame; calculating a characteristic index of the image frame; maintaining the image frame when the characteristic index is larger than a threshold value; and adding a fixed pattern to the image frame when the characteristic index is smaller than the threshold value. The present invention further provides a displacement detection device.

Abstract: Provided is a data analysis device for automatically detecting a step on the ground based on point cloud data representing a three-dimensional shape of a feature surface. A space subject to analysis is divided into a plurality of subspaces. A boundary search unit (22) searches for a boundary formed by the step on a horizontal plane for each of the subspaces. The boundary search unit (22) searches for a step neighborhood area having a predetermined width, in which the points projected on the horizontal plane are accumulated equal to or more than a criterion set in advance and a cloud of the points have a difference in height equal to or more than a step threshold set in advance, and searches for a directional line along a distribution of the cloud of points belonging to the step neighborhood area on the horizontal plane as the boundary.

Abstract: A basic idea of the present invention is to selectively employ one of at least two different feature extraction processes when generating a biometric template of an individual. An individual offers a physiological property, such as a fingerprint, an iris, an ear, a face, etc., from which biometric data can be derived, to a sensor of an enrollment authority. In the following, the property to be discussed will be fingerprints, even though any suitable biometric property may be used. From the fingerprint, a positional reference point of the biometric data is derived. The derivation of the positional reference point may be accomplished using any appropriate method out of a number of known methods. Such a reference point could be the location of a core, a delta, a weighted average of minutiae coordinates, or alike. Typically, the reference point includes a core of a fingerprint expressed as a three-dimensional coordinate denoted by means of xr, yr, and angle ?r.

Abstract: A method of authenticating a user by capturing an image of the user, assigning a user-name, selecting an image category, selecting a pattern, selecting locations on the user's face, assigning a rotation-value, logging onto a computer, presenting images, where some images are in the image category, selecting images that are the pattern away from the images in the image category, calculating a rotation angle, projecting the selected images that are rotated by the rotation angle onto the user's face at the locations, capturing an image of the user's face with the rotated images projected thereon, and authenticating the user if the user's facial features match the user's face and the projected images are the pattern away from the images in the image-category, are at the locations, are rotated by the rotation angle, and exhibit distortion consistent with the contour of the user's face.

Abstract: A photo sharing system with a face recognition function is provided. The photo sharing system includes a photo browser and a photo sharing platform. The photo browser has a photo classification module for classifying plural photos according to face image contained in the plural photos, thereby generating a photo classification message. The photo classification message contains the photos having the face image of the same person. Moreover, the photo classification message may be transmitted to the photo sharing platform at the same time to be shared to the receiver corresponding to face image. By the photo sharing system, the efficacy of sharing photos will be enhanced.

Abstract: In a face recognition system, overlapping patches are defined on a canonical face. Random clusters of pixel pairs are defined within each patch, and binary features are determined for each pixel pair by comparing their respective feature values. An inverted index hash table is constructed of the binary features. Similar binary features are then determined on a library of registrable samples of identified faces. A log probability of each registrable sample generating a binary feature from a corresponding cluster of pixel pairs at each specific patch location is determined and stored in the hash table. In a search phase, similar binary features are determined, and a hash key is determined for each binary feature. The log probabilities for each identity found in the hash table are summed for all clusters of pixel pairs and locations and sorted to find the high probability match.

Abstract: A system for processing a video signal into a processed video signal includes a pattern recognition module for detecting a face in the image sequence, based on coding feedback data, and generating pattern recognition data in response thereto, wherein the pattern recognition data indicates the pattern of interest. A video codec generates the processed video signal and generates the coding feedback data in conjunction with the processing of the image sequence.

Abstract: Apparatus for identifying video content includes a scene detector to detect a plurality of consecutive scenes in the video content. The video content may include a video program and an advertisement contained within the video program. A signature generator determines a scene change signature for each of the plurality of consecutive scenes. A content identifier provides the plurality of scene change signatures to a comparator that compares the plurality of scene change signatures to a database having a plurality of video content items to identify the video program. The database includes an ordered plurality of previously determined scene change signatures for each item of video content in the database. The content identifier receives an identification of the video content from the comparator. The content identifier may further receive information related to a position of the plurality of scene change signatures within the ordered plurality of previously determined scene change signatures.

Abstract: A mobile application stored within the mobile device allows for file collection, transfer, and display on a smart device with G3, G4, EDGE, or similar data communication capabilities. The software is used in conjunction with information collection devices synced to a smart device via blue tooth technology, such as a fingerprinting device. The device will transfer a file to the smart device running the mobile application. The application will then encrypt the file and encode the file that is then packaged into a proprietary format. The file is then sent to a web location using the data transmission capability of the mobile device. The device will then monitor a file location of a web address for a return response. It will then get the file from the web location in a standard NIST approved format, extract information and images from the file, displaying them in a GUI.

Abstract: An improved iterative reconstruction method to reconstruct a first image includes generating an imaging beam, receiving said imaging beam on a detector array, generating projection data based on said imaging beams received by said detector array, providing said projection data to an image reconstructor, enlarging one of a plurality of voxels and a plurality of detectors of the provided projection data, reconstructing portions of the first image with the plurality of enlarged voxels or detectors, and iteratively reconstructing the portions of the first image to create a reconstructed image.

Abstract: An information processing apparatus comprises: a spatial filtering unit configured to perform spatial filtering in a frequency range based on a spatial frequency of an object for image data of a current frame; and a recursive filtering unit configured to perform recursive filtering by obtaining image data, which has been processed prior to the current frame, from a memory, multiplying the obtained image data by a coefficient ? (?<1), adding the image data multiplied by the coefficient ? to the image data of the current frame after the spatial filtering, and storing the image data after the addition in the memory.

Abstract: A method for training a classifier to be operative as an epithelial texture classifier, includes obtaining a plurality of training micrograph areas of biopsy tissue and for each of the training micrograph areas, identifying probable locations of nuclei that form epithelia, generating a skeleton graph from the probable locations of the nuclei that form the epithelia, manually drawing walls on the skeleton graph outside of the epithelia to divide the epithelia from one another, and manually selecting points that lie entirely inside the epithelia to generate open and/or closed geodesic paths in the skeleton graph between pairs of the selected points. Data is obtained from points selected from the walls and the paths and applied to a classifier to train the classifier as the epithelial texture classifier.

Abstract: Locations, such as computer assisted detection marks, are presented in medical imaging diagnosis, such as for colon computed tomography diagnosis. To avoid fly-through and/or to assist in more rapid and thorough review of CAD marks, a summary of a plurality of detected candidates is pre-computed and presented to the user in a single collection. For example, a single display or screen view includes a gallery of images for different candidates. These pre-computed images are displayed on the screen, allowing the user to quickly identify locations of interest and rule out other locations. The summary may be used for navigation outside the CAD context, such as presenting different portions of an organ to be viewed.

Abstract: The present invention relates generally to improved methods of defining areas or compartments within which biomarker expression is detected and quantified. In particular, the present invention relates to automated methods for delineating marker-defined compartments objectively with minimal operator intervention or decision making. The method provides for precise definition of tissue, cellular or subcellular compartments particularly in histological tissue sections in which to quantitatively analyzing protein expression.

Abstract: A method for correcting a relaxation map of an object scanned with a magnetic resonance imaging modality the object having a plurality of structure and/or tissue types. The method includes deriving a first relaxation map of a scanned object from at least two three-dimensional scans of the object acquired using a sequence of ultrashort echo time pulses adapted for distinguishing between the various types of a plurality of structure and/or tissue types of the object. Information is obtained on the type of structure and/or tissue type present in voxels in the first relaxation map and binarizing the obtained information. A corrected relaxation map is generated by combining the binarized information with the first relaxation map.

Abstract: The method of segmenting a lesion in an image according to the present invention includes the steps of: obtaining an image of a region of interest from an image of a lesion; adjusting the contrast of the obtained image of the region of interest such that the contrast changes according to the distance from the center of the region of interest; removing noise from the image whose contrast is adjusted; and separating the region of the lesion from the noise-removed image. In the step of adjusting the contrast, the image is adjusted such that the contrast is increased as it goes away from the center of the region of interest, and thus it is possible to segment the lesion while maintaining the original shape of the lesion, thereby reliably segmenting the lesion even in the case where the inside of the lesion is not uniform.

Abstract: Technologies are described herein for generating a diagnostic three dimensional image for a patient. Some example technologies may obtain a sequence of multiple images of the patient using an imaging modality device. The technologies may estimate a registration vector for each image based on a motion function and an image transformation function. Each image may be defined by a measurement noise added to the image transformation function operating on the registration vector with respect to a reference image. The registration vector may be a function of a breathing motion of a prior registration vector added to a transition noise value. The technologies may estimate motion parameters based on the registration vector. The technologies may iteratively refine the registration vector and the motion parameters. The technologies may generate the diagnostic three dimensional image of the patient using the registration vector for each image and the motion parameters.

Abstract: A method for automatically detecting the presence of contrast in an x-ray image includes acquiring an x-ray image prior to administration of contrast. A background image is estimated based on the x-ray image. The contrast is administered. A set of x-ray images is acquired. The background image is subtracted from the set of images. Image intensity is determined for each of the subtracted images. The subtracted images having highest image intensity are selected. A predefined shape model is fitted to the selected subtracted images. The fitting of the predefined shape model is used to fit the shape model to each of the subtracted images. A feature value is calculated for each image frame based on pixel intensities of each pixel fitted to the shape model for the corresponding subtracted image. An image frame of peak contrast is determined by selecting the image frame with the greatest feature value.

Abstract: The invention relates to a method and an apparatus for providing device sizing support, the method comprising obtaining an X-ray image of a vessel, introducing a guide wire having a radiopaque wire tip into the vessel, obtaining an X-ray image of the wire tip, segmenting the wire tip when it passes through the vessel, and providing sizing information relating to the size of the vessel based on the size of the wire tip. The imaging system according to the invention includes means providing functionality for performing the method according to the invention.

Abstract: A method of operating an instrument that is a macroscope, microscope, or slide scanner is provided where the instrument has a larger dynamic range for measurement than a dynamic range required in the final image of a specimen. In the method, data is measured from a specimen using the instrument, and the dynamic range of the measured data is contracted in the final image file during scanning.

Abstract: A method of determining the authenticity of a document of value provided with a watermark pattern, the method comprising determining whether the watermark pattern exhibits a discontinuity.

Abstract: A system for tracking currency bills comprises a currency scanning device. The scanning device includes a sensor that retrieves currency identification characteristic information of each bill processed. The currency identification characteristic information permits the unique identification of each bill processed. The system further comprises a customer identification means and means for associating each processed bill with the customer depositing the bill. Means for identifying the customer (or customer account) associated with a particular processed bill after the deposit transaction has been completed is also included in the system.

Abstract: A batch of bills is received, transported, and imaged to produce image data. The image data is reproducible as a visually readable image of at least a portion of each of the plurality of currency bills. Each currency bill includes a respective serial number. The serial numbers are extracted from the image data for each of the currency bills. Each extracted serial number is compared with a plurality of serial numbers in a currency bill database to determine if a currency bill associated with an extracted serial number is a suspect currency bill. One of the plurality of currency bills is determined to be a suspect currency bill based on the comparing. In response to determining the suspect currency bill, at least a portion of the produced image data is displayed as a visually readable image of the suspect currency bill on a display unit of the document processing system.

Abstract: Systems, methods, and computer program products are provided for automating the separation of a check and a stub attached to the check in a payment receiving and processing system. For example, one embodiment of the invention involves a computerized apparatus configured to receive a document image, locate a Magnetic Ink Character Recognition (MICR) line in the document image, and capture a check image by capturing a portion of the document image based at least partially on the location of the MICR line.

Abstract: A method and apparatus for measuring size distribution of bulk matter consisted of randomly orientated granules, such as wood chips, make use of scanning the exposed surface of the granular matter to generate three-dimensional profile image data defined with respect to a three-coordinate reference system, The image data is segmented to reveal regions associated with distinct granules, and values of the size-related parameter for the revealed regions are estimated. Then, a geometric correction to each ones of estimated size-related parameter values is applied, to compensate for the random orientation of corresponding distinct granules. Finally, the size distribution of bulk matter is statistically estimated from the corrected size-related parameter values.

Abstract: A disclosed method of identifying a data point distribution area having data points on a coordinate plane includes dividing a coordinate plane area into plural divided areas using lines; in each divided area, selecting outermost data point data as representative points, and connecting the selected representative points to define a distribution area; comparing the area with a reference area to determine an overlapping area; and determining that the distribution area is a relevant area based on the existence of the overlapping area.

Abstract: Although there has been a method for evaluating pattern shapes of electronic devices by using, as a reference pattern, design data or a non-defective pattern, the conventional method has a problem that the pattern shape cannot be evaluated with high accuracy because of the difficulty in defining an exact shape suitable for the manufacturing conditions of the electronic devices. The present invention provides a shape evaluation method for circuit patterns of electronic devices, the method including a means for generating contour distribution data of at least two circuit patterns from contour data sets on the circuit patterns; a means for generating a reference pattern used for the pattern shape evaluation, from the contour distribution data; and a means for evaluating the pattern shape by comparing each evaluation target pattern with the reference pattern.

Abstract: In the case where a three-dimensional image including a plurality of element figures, such as a quadrangular plane, a curved surface including a ridgeline, and a cylinder, is edited using an image editing apparatus such as a three-dimensional CAD apparatus, when one element figure in the three-dimensional image is specified, other element figures connected to the specified one element figure, or other element figures included in the same three-dimensional image in which the one element figure is included are retrieved, and other element figure in the retrieved other element figures, which was determined as having a specific relationship to the one element figure is related to the one element figure. Then, the specified element figure and the element figures related to the specified element figure are edited collectively based on an operation received by the image editing apparatus.

Abstract: The present invention is directed to a system and method for interactive and iterative reconstruction in a manner that helps to reduce computational requirements by generating a model from a subset of the available data, and then refining that model using additional data. Example embodiments directed to scene reconstruction, reconstruct a 3D scene from a plurality of 2D images of that scene by first generating a model of the 3D scene from a subset a of the 2D images. The model can then be refined using specific characteristics of each image in the subset that are calculated using the other images in the subset. The model is further refined using images not in the original subset.

Abstract: Methods and systems for generating a size measurement of a body part of person for fitting a garment include providing photographic data that includes images of the body part and using feature extraction techniques to create a computer model of the body part.

Abstract: A system and method of correcting a depth map for 3D image is disclosed. A spatial spectral transform unit extracts pixels of object boundaries according to an input image, wherein the spatial spectral transform unit adopts Hilbert-Huang transform (HHT). A correction unit corrects an input depth map corresponding to the input image according to the pixels of object boundaries, thereby resulting in an output depth map.

Abstract: For converting a two-dimensional visual image into a three-dimensional visual image, the two-dimensional visual image is segmented into regions, including a first region having a first depth and a second region having a second depth. The first and second regions are separated by at least one boundary. A depth map is generated that assigns variable depths to pixels of the second region in response to respective distances of the pixels from the boundary, so that the variable depths approach the first depth as the respective distances decrease, and so that the variable depths approach the second depth as the respective distances increase. In response to the depth map, left and right views of the three-dimensional visual image are synthesized.

Abstract: A computer implemented method for matching video data to a database containing a plurality of video fingerprints of the type described above, comprising the steps of calculating at least one fingerprint representing at least one query frame from the video data; indexing into the database using the at least one calculated fingerprint to find a set of candidate fingerprints; applying a score to each of the candidate fingerprints; selecting a subset of candidate fingerprints as proposed frames by rank ordering the candidate fingerprints; and attempting to match at least one fingerprint of at least one proposed frame.

Abstract: In an information processing apparatus that processes data using cascade-connected weak classifiers, processing specification information specifying the processing content of each of the weak classifiers is stored. The weak classifiers to be used in processing the data are selected from the weak classifiers by referring to a table in which is specified information for determining the weak classifiers to be used based on a condition for processing the data. The data is then processed by the selected weak classifiers based on the processing specification information that corresponds to those weak classifiers, and an object is extracted from the data using an obtained evaluation value. Through this, a combination of extraction process speed and extraction accuracy can be changed in a flexible manner when extracting a specific object from image data.

Abstract: A system and method of adaptive edge detection and noise reduction in an image where edge information is detected for each color component of each pixel, whether sensed or synthesized. In some embodiments, the filter applied to a selected non-edge pixel may be determined by the ultimate size of a region around the selected pixel, where the size of the region may be increased if a count of the non-edge pixels in the region is less than a threshold value.

Abstract: Disclosed is a processor-implemented method for processing image data representing a document using an image processing apparatus. The processor receives image data, determines the classifications and counts of pixels in the image data (e.g., in device independent space that are degrees of neutral or non-neutral), classifies the image data into a category based on the determination of classifications and counts of pixels, and determines a billing structure for the image data based on the classification of image data. The classification can be performed on the image data prior to conversion into device dependent space. For example, by counting true color, neutral, non-neutral, and fuzzy color pixels and comparing them to thresholds, the image data can be classified as neutral or very colorful. By classifying images earlier in the image path, errors in billing are reduced.

Abstract: A night-scene light source detecting device includes a pixel value obtaining unit configured to obtain a pixel value of each pixel in an input image; a night-scene-feature extraction unit provided for extracting a zone area of a mean corrected-brightness value and a high corrected-brightness value of the input image as two night-scene features based on the pixel value of each pixel in the input image; a night-scene image detection unit provided for determining whether the input image is a night-scene image or not based on the two night-scene features; a specific color detection unit provided for detecting whether each pixel belongs to specific color or not; and a night-scene light source determining unit provided for determining whether the night-scene image is picked up under irradiation by the specific light sources in a night scene or not based on the result of the specific color detection.

Abstract: Provided is an image processing apparatus including an extraction unit that extracts a color boundary of an image, an influence value calculation unit that calculates an influence value indicating the degree of an influence of the color boundary on a pixel, a setting unit that sets a coefficient for a convolution operation in accordance with the color boundary extracted in the extraction unit and the influence value calculated in the influence value calculation unit, and an operation unit that performs a convolution operation using the coefficient set by the setting unit.

Abstract: An image processing apparatus comprises a central processing unit (CPU) and a graphics processing unit. The CPU comprises a first table specification, a second table specification, an inverse discrete cosine transformation (IDCT) module, a color space converting module and an entropy decoder. The entropy decoder retrieves and processes input image data as a two-dimensional quantized image according to the first table specification. The graphics processing unit performs a dequantizing operation on the input image data according to the second table specification. The IDCT module performs an IDCT operation on the dequantized input image data. The color space converting module performs a color space conversion on the input image data and outputs the converted image data.

Abstract: A decoding system comprises N different decoders each having a unique circuitry that is different from the circuitries of the other N?1 decoders. The decoders each generate at least one texel value based on an input encoded texel block. A value selector is configured to selectively output at least N texel values from at least one of the decoders based on the position of the at least N texels relative a boundary of a texel block comprising at least one of the at least N texels. A pixel calculator calculates a pixel value of a decoded pixel based on the at least N selected texel values from the value selector.

Abstract: An image compression circuit 1 of the invention includes an image calculator 11, a compression-system decision unit 12, and a compression processor 13. Compression-system decision unit 12 determines a variation of pieces of image data of four adjacent pixels based on a variation in luminance Y of the pixels or variations in luminance Y and saturation S of the pixels, compresses the image data into image data of one or two pixels (RGB 888 and RGB 787) expressed by an RGB color space having a small quantization error in the case of the small variation of the pieces of image data, and compresses the image data into data including the luminances of the four adjacent pixels and pieces of image data of two pixels (YCbCr 422) expressed by a YCbCr color space having a large quantization error in the case of the large variation of the pieces of image data.

Abstract: An image processing device comprises a halftoning module, a detecting module, an adding module and a generating module. The halftoning module generates an error diffusion value. The detecting module detects an edge at a current pixel of an image. The adding module adds the pixel value of the current pixel with one of the error diffusion value and an error diffusion seed according to the detecting result. The generating module generates the error diffusion seed. The error diffusion seed is generated by a formula based on the pixel value and location of the current pixel and a random number, or the error diffusion seed is generated by a formula based on the pixel value of the current pixel and a looked-up value, and the looked-up value is retrieved from a look-up table according to the location of the current pixel.

Abstract: A color constancy method and system include dividing an image into a plurality of sub-images and applying a plurality of color constancy algorithms to each of the sub-images. The outputs of each of the color constancy algorithms are analyzed for each of the sub-images to determine which of the color constancy algorithms give inconsistent results across the sub-images. The influence of the outputs of the algorithms providing inconsistent results is adjusted to decrease their influence (e.g. effect or weight) with respect to the outputs of algorithms providing consistent results. The outputs from the plurality of color constancy algorithms are combined based upon the adjustment of the outputs.