Abstract: The interpolation unit comprises: a fetching unit for fetching a first value of a first original pixel of a first input image, based on a first motion vector applicable to the new pixel; a projection unit for projecting a second value of a second original pixel of a second input image, based on a second motion vector applicable to the second original pixel; and a selection unit for assigning the value of the new pixel by means of comparing the first value with the second value.

Abstract: A method for identifying a pathological region of a scan (such as a stroke region within a MRI DWI volume scan) is proposed. A region of the scan which is likely to contain pathological tissue (e.g. infracted tissue) is identified by obtaining a parameter which, for a given slice, or portion of a slice, characterizes the distribution of the intensity of pixels, e.g. the relative proportion of high intensity pixels. In a first case, such a parameter is used to identify those slices of a volume scan which are likely to include infarction. In a second case, such a parameter (hemisphere parameter) is obtained for each of the left- and right-hemispheres of a brain, to estimate which hemisphere contains the stroke. In either case, the parameter may be calculated based on ranges, percentiles and functions of the percentiles of the intensity distribution. These ranges, percentiles and functions of the percentiles are not pre-defined but are selected to maximize sensitivity.

Abstract: An object detection control apparatus includes: a detection processing allocating unit that allocates object detection processing to any one of plural object detecting units that perform the detection processing on the basis of a transfer image area; a size generating unit that generates size of the transfer image area according to a detection area for the object detecting-unit to which the detection processing is allocated; and a reference-position generating unit that generates a position serving as a reference for the transfer image area using the generated size of the transfer image area and a position serving as a reference for the detection area moved by a predetermined number of pixels at a time.

Abstract: A method for performing crowd segmentation includes receiving video image data (S21). Background differencing is performed on the received video image data to identify a foreground silhouette shape (S22). Approximate number and position of human subjects within the received video image data are determined by matching the foreground silhouette shape against a set of predetermined foreground silhouette shapes, for each of which a number and position of human subjects is known (S28). The estimated number and position of the human subjects is refined to determine a final number and position of the human subjects (S27).

Abstract: An apparatus for segmenting a cerebral hemorrhage site in a medical image of a head comprises a means for segmenting an internal region of a skull bone in the medical image of the head, a means for segmenting a possible region in which a cerebral hemorrhage site is contained, out of the region segmented by the means for segmenting the internal region of the skull bone, and a means for determining a cerebral hemorrhage site out of the region segmented by the means for segmenting the possible region in which the cerebral hemorrhage site is contained.

Abstract: Image data representing a captured image of an environment or a portion thereof, and electric-field data representing electric fields present in the environment or a portion thereof are received. The electric-field data and the image data represent information gathered from an overlapping portion of the environment. A candidate object is identified in the overlapping portion and classified as animate or inanimate based at least upon an analysis of the image data and the electric-field data.

Abstract: A mobile object recognizing device comprises a camera (2) for taking time-series images, a feature point extracting unit (23) for extracting the feature points of the individual time-series images taken by the camera (2), an optical flow creating unit (24) for comparing the feature points of the time-series images between different images, to create an optical flow joining the feature points having the same pattern, and a grouping operation unit (25) for selecting that optical flow as one belonging to one mobile object, the prolonged line of which intersects with one vanishing point within a predetermined error range and in which the external ratio of a segment joining the other end point and the vanishing point with one end point of the optical flow being the externally dividing point in the extension of the optical flow.

Abstract: A method of detecting a click event for sensing a motion of a finger corresponding to a click on a sensing area of an optical pointing device, the method including: obtaining an image of the finger from the sensing area; sensing a change in the image of the finger; analyzing a horizontal movement of the finger based on the change in the image of the finger; and generating a click signal when the horizontal movement of the finger is within a predetermined range is provided.

Abstract: A camera tracking apparatus for calculating in real time feature information and camera motion information based on an input image includes a global camera tracking unit for computing a global feature map having feature information on entire feature points; a local camera tracking unit for computing in real time a local feature map having feature information on a part of the entire feature points; a global feature map update unit for receiving the computed feature information from the global and local camera tracking units to update the global feature map; and a local feature selection unit for receiving the updated feature information from the global feature map update unit to select in real time the feature points contained in the local feature map. The local camera tracking unit computes the local feature map for each frame, while the global camera tracking unit computes the global feature map over frames.

Abstract: The present invention discloses apparatus, method, and system for CT image processing; said apparatus comprises: an interface unit for obtaining CT images; a target region separation unit for separating said target region from the CT images by means of judging the CT values of voxel points in said CT images, wherein points with CT values greater than or equal to the threshold for target region separation constitute the target region, and said target region comprises at least one separate region; a false positive region removing unit for removing false positive regions from said target region and obtaining the accurate target region.

Abstract: Systems, methods and apparatus are provided through which carotid plaque is classified in an image and visually displayed using an iterative adaptive process, such as an expectation maximization process.

Abstract: Stored documents can be recognized and identified by evaluation of their attributes from an image file to obtain a document specification that can be compared to stored data about the document. The image file is analyzed using a processor in order to discern image attributes. The image attributes form a specification for the document in the image. This specification can then be compared to the stored criteria using the processor, and the result of the comparison can be stored. An embodiment of the invention can be used in a check processing system for a financial institution. In such a case, the corresponding data comprises MICR data and the image attributes can, for example, indicate whether the document is a check in a carrier, as opposed a paper check by itself.

Abstract: An embodiment of the present invention includes: a tracking object image extracting section that extracts a tracking object image, which represents a tracking object, from an image captured by a monocular camera; a two-dimensional displacement calculating section that calculates, as actual movement amounts, amounts of inter-frame movement of the tracking object image; a two-dimensional plane projecting section that generates on a two-dimensional plane a projected image of a three-dimensional model, which represents in three dimensions a capturing object captured by the monocular camera; a small motion generating section that calculates, as estimated movement amounts, amounts of inter-frame movement of the projected image; and a three-dimensional displacement estimating section that estimates amounts of three-dimensional motion of the tracking object on the basis of the actual movement amounts and the estimated movement amounts.

Abstract: A monitoring station for monitoring distribution of media content, on basis of a watermark, comprises: receiving means for receiving an information signal representing the media content to which the watermark is added; extracting means for extracting perceptual features, identifying the information signal; first retrieving means for retrieving a supporting signal on basis of the perceptual features; second retrieving means for retrieving the watermark on basis of the supporting signal; and comparing means for comparing the watermark with predetermined information.

Abstract: A method for determining the distance of visibility for a driver of a vehicle in the presence of an element disrupting the visibility of the driver, the method comprising the following steps: determining the luminosity of the pixels of a region of an image taken in the field of vision of the driver, resulting in a luminosity curve, determining a first tangent to the curve of luminosity, determining a second tangent to the curve of luminosity, determining a sweep-line according to the first tangent and second tangent, the sweep-line being representative of the distance of visibility.

Abstract: A system for recognition of a target three-dimensional object is disclosed. The system may include a photon-counting detector and a three-dimensional integral imaging system. The three-dimensional integral imaging system may be positioned between the photon-counting detector and the target three-dimensional object.

Abstract: A method of processing vegetation data including the steps of identifying data relating to an agricultural field, segregating areas of predetermined development patterns and prescribing application rates of an agricultural compound. The identifying step includes identifying data relating to an agricultural field representative of areas of predetermined development patterns of vegetation in the field. The segregating step includes segregating the areas of the predetermined development patterns thereby defining segregated areas, other areas in the field being non-segregated areas. The prescribing step including prescribing application rates of an agricultural compound to the non-segregated areas dependent on at least one attribute determined from the data.

Abstract: An apparatus, system and method for mapping information. The apparatus for mapping information includes an information input unit providing image information and position-view information in a specified area, a three-dimensional model database storing three-dimensional model data of a structure within the specified area and generating a two-dimensional image from the three-dimensional model data using the position-view information, an image processing unit comparing the two-dimensional image with the image information to analyze the image information, a related information acquiring unit acquiring structure related information within the specified area with reference to the analyzed image information, and an information mapping processing unit mapping the structure related information on the image information and outputting a mapping result.

Abstract: A water marking system generates a water marked image frame for conveying a payload data word by combining at least one image frame with a two-dimensional water mark pattern. The water mark pattern includes plural regions each associated with one of symbols of the payload data word. Each region includes a two dimensional water mark block selected from a predetermined set of possible water mark blocks to represent the value of the payload data symbol associated with that region. A detecting data processing apparatus detects the payload data from a received water marked image frame, which has been encoded by utilizing a block match prior probability calculator operable to correlate each region of the water marked image frame with each of the water mark blocks from the set of water marked blocks) to form for each region a set of probability surfaces.

Abstract: A method, data processing facility and program storage device for quantitative analysis on medical image data of a bifurcated tubular organ which involves processing the medical image data to identify contours of the bifurcated tubular organ. The contours are used to determine a Polygon of Confluence amongst the bifurcated tubular organ. The Polygon of Confluence is used to determine at least one parameter value characterizing geometry of the bifurcated tubular organ. The at least one parameter value is outputted to a user for angiography purposes. The at least one parameter value can include at least one diameter value of the bifurcated tubular organ, at least one angle value between parts of the bifurcated tubular organ, and at least one reference diameter value for the bifurcated tubular organ, the at least one reference diameter value compensating for damage to the bifurcated tubular organ.