Abstract: A method of calculating a transform matrix for transforming points in a model of an object to an image of the object captured by a camera. The camera produces a two dimensional view of the object, the object having a plane, which includes a plurality of lines defining features on the plane of the object. The method comprises identifying a plurality of triplets of points on the lines of the object in the image from the camera, each of the triplets of points providing three or more non co-linear points on the lines of the object.

Abstract: Facial feature point reliability generating means generates a reliability map of each facial feature point from a facial image. Initial facial feature point position calculating means calculates the position of each facial feature point in the facial image based on the reliability map. Off-position facial feature point judgment means judges whether or not each facial feature point is an off-position facial feature point not satisfying a prescribed condition. Facial feature point difference calculating means calculates the difference between the position of each facial feature point, excluding those judged as the off-position facial feature points, and the position of a corresponding point of the facial feature point. Facial feature point position correcting means corrects the determined positions of the facial feature points based on the results of the judgment by the off-position facial feature point judgment means and the calculation by the facial feature point difference calculating means.

Abstract: According to one embodiment, a search skip region setting function generation method includes associating, detecting, and generating. The associating associates a template used to search a model image for an object with a designated search point on the model image, and detects a designated search point similarity between the designated search point and the template. When the designated search point similarity exceeds an object detection determination threshold, the detecting detects surrounding search point similarities between a plurality of surrounding search points around the designated search point on the model image and the template. The generating generates a function required to set a search skip region of the object based on relative positions between the object and the template, which are estimated based on a distribution of the surrounding search point similarities.

Abstract: A pattern information registration device for selecting and registering pattern information data as an object for determination of registration for use in pattern collation includes unit which forms arbitrary pattern information data having a feature point of the same number as pattern information data of an object for determination, unit which calculates an identification accuracy value indicative of the degree of coincidence between the pattern information data as an object for determination and the arbitrary pattern information data and determines whether the pattern information data as an object for determination is adequate or not based on the identification accuracy value.

Abstract: An information processing apparatus includes a storing unit that stores information concerning model feature points and model feature quantities at the model feature points, a first acquiring unit that acquires an input moving image, a first feature-point extracting unit that extracts input feature points for recognizing an action from the input moving image, a first feature-quantity extracting unit that extracts input feature quantities at the input feature points, a feature-quantity comparing unit that compares the input feature quantities and the model feature quantities and generates candidate corresponding feature point pairs, a posture estimating unit that removes outliers from the candidate corresponding feature point pairs, estimates postures of models on the input moving image, and obtains a recognition corresponding feature point pair group corresponding to the postures, and a recognition-result generating unit that generates a recognition result on the basis of the recognition corresponding feature

Abstract: A linear mark detection system includes a filter value calculation unit, a filter value threshold processing unit, and a feature value output unit. The filter value calculation unit calculates, for each pixel of an image, filter values corresponding to each of a plurality of linear mark widths using pixel values inside and outside the linear mark. The filter value threshold processing unit selects the most suitable value from filter values corresponding to each of the plurality of linear mark widths as a feature value. The feature value output unit outputs, for each pixel of the image, the feature value computed by the filter value threshold processing function as the feature value corresponding to the linear mark.

Abstract: An image recognition device includes a processor, and a memory coupled to the processor, wherein the processor executes a process includes detecting a plurality of contour points arranged on a contour line of a given region in an image, detecting a first contour point and a second contour point, in the contour points, the first and second contour points corresponding to respective ends of a first shortcut line formed by connecting portions of the contour line across an external area of the given region, and determining, based on a length of the first shortcut line or a length of a first route that extends along the contour line between the first and second contour points, whether or not a portion surrounded by the first shortcut line and the first route, not contained in the given region, is a depressed portion.

Abstract: An apparatus and method is described for determining the location of a object inside a patient. Three dimensional image data of the patient is obtained by an imaging modality such as a C-arm X-ray device. The same or another imaging modality may be used to obtain projection radiographic images, having a coordinate system that can be registered with respect to that of the three dimensional image. The location of one or more features of a medical treatment device such as a needle or a catheter introduced into the imaging filed of view is determined either by an operator or by image analysis, in at least two projection radiographic images, and the position of the feature located in the three-dimensional volume. The projection on a radiographic image of a vector joining a first and a second position is used to assist in guiding the medical treatment device inside of the patient.

Abstract: A method for detecting facial feature is provided. Concerned points and concerned areas are extracted from a captured image, and an area which is considered to highly possibly include a facial area is selected. It is determined whether the selected area corresponds to a facial feature by using a support vector machine.

Abstract: The present invention relates to a face detection system for a vehicle. At least one first lighting unit is configured to radiate infrared light onto a left side of a driver's face. At least one second lighting unit is configured to radiate infrared light onto a right side of the driver's face. An image capturing unit separately captures the driver's face onto which the infrared light is radiated from the first and second lighting units. A control unit acquires left and right images of the face from the image capturing unit, and obtains a difference image between the acquired left and right images, thus determining whether the driver is inattentive in looking ahead. The system stably performs the face detection function with no or less effect by external optical environments as well as reduced computational load.

Abstract: Automatic conflation systems and techniques which provide vector-imagery conflation and map-imagery conflation. Vector-imagery conflation is an efficient approach that exploits knowledge from multiple data sources to identify a set of accurate control points. Vector-imagery conflation provides automatic and accurate alignment of various vector datasets and imagery, and is appropriate for GIS applications, for example, requiring alignment of vector data and imagery over large geographical regions. Map-imagery conflation utilizes common vector datasets as “glue” to automatically integrate street maps with imagery. This approach provides automatic, accurate, and intelligent images that combine the visual appeal and accuracy of imagery with the detailed attribution information often contained in such diverse maps. Both conflation approaches are applicable for GIS applications requiring, for example, alignment of vector data, raster maps, and imagery.

Abstract: Methods are disclosed for finding images from a large corpus of images that at least partially match a query image. The present method makes use of feature detectors to bundle features into local groups or bundles. These bundled features are repeatable and much more discriminative than an individual SIFT feature. Equally importantly, the bundled features provide a flexible representation that allows simple and robust geometric constraints to be efficiently enforced when querying the index.

Abstract: A document alteration detection method compares a target image with an original image by comparing character shape features without actually recognizing the characters. Bounding boxes for the characters are generated for both images, each enclosing one or more connected groups of pixels of one character. The bounding boxes in the original and target images are matched into pairs. Addition and deletion of text is detected if a bounding box in one image does not have a matching one in the other image. Each pair of bounding boxes is processed to compare their shape features. The shape features include the Euler numbers of the characters, the aspect ratio of the bounding boxes, the pixel density of the bounding boxes, and the Hausdorff distance between the two characters. The two characters are determined to be the same or different based on the shape feature comparisons.

Abstract: The invention relates to an image processing apparatus by means of which an image identification can take place in real time and which, with only very little or even no a priori information, is capable of carrying out a pertinent and reliable identification of objects. To this end, the connection probabilities are determined between two contour points in each case, taking into account the distance between the points, by means of a computation mechanism.

Abstract: An image processing apparatus includes a characteristic point extracting section and a signal processing section. The characteristic point extracting section is configured to analyze a frequency component of an inputted image and extracts a plurality of pixels characterizing the image as a plurality of characteristic points. The signal processing section is configured to perform signal processing for the plurality of extracted pixels corresponding to the respective characteristic points, produces an interpolated value based on a result of the signal processing, and rewrites a value of at least one pixel other than the characteristic points to the interpolated value.

Abstract: A method for mapping images having a common landmark or common reference point, in order to enable the creation, location and/or mapping of pixels, coordinates, markings, cursors, text and/or annotations across the images The method includes selecting at least two images having the common landmark or common reference point, mapping the selected images so as to generate mapping parameters that map a first location on a first image to the corresponding location of the first location on a second image, and identifying at least one pixel on the first image and applying the mapping parameters to the at least one pixel on the first image to identify the corresponding pixel or pixels in the second image The mapping parameters then may be used to locate or reproduce any pixels, coordinates, markings, cursors, text and/or annotations of the first image at the corresponding location of the second image.

Abstract: According to one embodiment, an image processing apparatus connectable to a main memory in which a plurality of pixel values of unconverted image is stored and a cache memory including a plurality of cache blocks. The apparatus includes a counter, a coordinate determination module, a memory controller, a cache access module, a pixel value calculator, and an output module. The counter determines a coordinate within converted image according to a predetermined execution sequence. The coordinate determination module determines a plurality of coordinates within unconverted image of the pixel values of unconverted image necessary to calculate a pixel value of converted image corresponding to the coordinate within converted image. The memory controller transfers the pixel values of unconverted image stored in the main memory to the cache blocks corresponding to each of the coordinates within unconverted image.

Abstract: The image analysis system includes a processor and memory and displays an image to a first user. The image analysis system tracks gaze of the first user and collects initial gaze data for the first user. The initial gaze data includes a plurality of gaze points. The image analysis system identifies one or more ignored regions of the image based on a distribution of the gaze data within the image; and displays at least a first subset of the image. The first subset of the image is selected so as to include a respective ignored region of the one or more ignored regions and the first subset of the image is displayed in a manner that draws attention to the respective ignored region. In some embodiments, the ignored region is visually emphasized within the image. In some embodiments, only the first subset of the image is displayed.

Abstract: An image processing device including: a motion amount estimating unit estimating a first motion amount indicating a displacement amount of a second picture with respect to a first picture; an area setting unit setting an area to be used for post-processing for each of the first picture and a first corrected picture, using feature points extracted from the first picture and the second picture; a post-processing unit generating a first cutout picture from the first picture and a second cutout picture from the first corrected picture and estimating a second motion amount indicating a displacement amount of the second cutout picture with respect to the first cutout picture; and a motion amount obtaining unit obtaining a motion amount indicating a displacement amount between the first picture and the second picture, from the estimated first motion amount and the estimated second motion amount.

Abstract: There is provided an image processing device including a feature point detection processing unit configured to detect a feature point from an image, and a feature quantity generation processing unit configured to compare a pixel difference value of two pixels in an image region having a position of the detected feature point as a reference with a threshold and generate binary information indicating a result of comparison as a component of feature quantities corresponding to the feature point.

Abstract: There is provided an image processing device including a feature point detection unit configured to detect feature points from an image, a reliability computation unit configured to compute reliability of each of the detected feature points, and a removal processing unit configured to remove some of the detected feature points on the basis of the reliability so that the number of the remaining feature points in each split region that is obtained by splitting the image into a plurality of regions is within a restricted number of feature points.

Abstract: A large number of stable local regions can be set with low calculation cost. In a face recognition apparatus which discriminates similar face images using feature amounts extracted from local regions included in an image to be discriminated, a moving destination of a feature point extracted from the image to be discriminated, and the size of an image to be clipped at the moving destination are calculated based on a table which defines information required to designate a moving destination of each feature point, and information required to designate the size of an image to be clipped at the moving destination, and an image with the calculated size is clipped at the calculated moving destination as the local region.

Abstract: Provided is a marker generation device which has a singular feature selection means and a marker pattern generation means. The singular feature selection means extracts feature points from an image, indicates the feature points in a predetermined space, and selects, as singular features, a plurality of parts where the number of feature points is equal to or less than a predetermined number in the space. The marker pattern generation means generates a plurality of kinds of marker patterns with the use of the whole or a part of the plurality of singular features.

Abstract: A feature used in face detection can be applied to an image portion and can be scaled to fit differently sized image areas. If a feature is positioned with respect to an image area such that a vertex of the feature is aligned with a non-integer pixel location, at least one dimension of the filter can be rounded. A dimension to be rounded further can correspond to a directional component of the feature. For instance, contrast regions within the feature can be arranged horizontally, such that the vertical dimension represents a directional component. A rounding rule associated with the feature can be used in rounding a dimension corresponding to a directional component, such that a size ratio between the contrast regions is maintained. In some instances, the rounding rule can specify a factor that is a positive integer determined based on the number of contrast regions in the feature.

Abstract: Subject matter disclosed herein relates to electronic image object or logo recognition.

Abstract: Disclosed is a method for determining and displaying at least one piece of information on a target volume, especially in a human body, the information being obtained from an image record. At least one embodiment of the method includes: a first and at least one second image record of a target zone encompassing the target volume are recorded, the first image record having a higher contrast regarding the boundaries of the target volume, and the first and the second image record being registered together; the target volume is segmented in the first image record; target volume image data.

Abstract: A controller and navigation system to implement beacon-based navigation and scene-based navigation is described. The navigation system may generate position data for the controller to compensate for a misalignment of the controller relative to the coordinate system of the navigation system. The navigation system may also distinguish between a beacon light source and a non-beacon light source.

Abstract: Analyzing an input image, the input image being one of a digitized image stored in a memory or a scanned image from a scanner. Forming a feature image from the input image by dividing the input image into a plurality of blocks of pixels, thus associating each block of pixels in the input image with a single pixel in the feature image, and outputting the feature image for further analysis or storage in a memory. Example embodiments extract and analyze features from a document image to detect particular characteristics associated with the page area, the distortion area, and the book spine area. Extracted features can be further analyzed to detect document characteristics at the paragraph, line, word, and character levels.

Abstract: The invention provides a method and apparatus for acquiring descriptive information of a plurality of images and an image matching method. The method for acquiring descriptive information of a plurality of images includes: performing a feature point detection with respect to each image of the plurality of images, so as to obtain a plurality of feature points of each image; acquiring 0-level descriptive information of the plurality of images; and the following steps are performed for each image: performing a division of the image for the (n?1)th time, so as to obtain a plurality of (n?1)-level sub-images of the image; and (n?1)-level descriptive information of the image is generated in accordance with a plurality of nth local feature descriptors for the image and a plurality of nth visual words, where n=2,3, . . . , K+1, and K is a positive integer.

Abstract: Disclosed is an image processing device which is provided with a determination section which determines texture direction in a vicinity region based on the vicinity region which includes a target pixel in the image and a detection section which detects whether or not the target pixel is a defective pixel based on a plurality of pixels including the target pixel which are lined up in the texture direction in the image.

Abstract: A method for determining feature point locations in an image performs a first search in a predetermined first search area to search for locations of plural feature points in the image, corrects the locations of the plural feature points based on a geometric layout relationship among the plural feature points searched for, sets a second search area based on the corrected location of each of the feature points, and performs a second search in the second search area to search for the location of each of the feature points. Then, the method determines reliability of the location of each feature point searched for by the second search and selects one of the corrected location and the location searched for by the second search, as a location of the feature point.

Abstract: A method for delivering an electronic coupon to a mobile device includes receiving via a wireless network a target image or icons extracted from the target image, comparing the target image to a plurality of learned images using a processor in communication with the wireless network, determining whether the target image matches at least one learned image, and when the target image matches at least one learned image, delivering to the mobile device over the wireless network an electronic coupon associated with the at least one learned image. The target image can represent at least a portion of a front surface of a consumer packaged good located in a retail outlet. The target image can be generated using a mobile device configured to communicate with the wireless network.

Abstract: According to one embodiment, an image processing apparatus includes a feature based detector, an optimum area generator, and an output data generator. The feature based detector detects a feature portion of input image data and generates feature area information showing a position of a feature area comprising the detected feature portion. The optimum area generator generates optimum area information showing a position of an optimum area in accordance with a size of the feature area based on the feature area information. The output data generator extracts pixels of the optimum area among the input image data based on the optimum area information and generates output image data based on the extracted pixels.

Abstract: The present invention provides a true-scale, coordinate-matched, linked in real-time, dual three-dimensional/two-dimensional visual display/viewer. The display simultaneously shows a 3D digital image and an associated 2D digital image of a selected drawing. The display of the present invention allows a user to visualize an asset's location, surrounding environment and hazards and true scale structural details for interior or external structural scenes. Using the display and associated tools, the user can obtain real-time information of an environment, true-scale measurement, plan ingress/egress paths, shortest paths between points and the number of doorways in a structure and track objects within the displayed environment. The intelligence gained using the tools and 3D/2D display may be used and further manipulated by a single user or may be distributed to other users.

Abstract: A pattern recognition apparatus including: an extracting section for extracting, from a query image that is composed of at least one piece of pattern component and previously undergoes a geometric transformation, the pattern component; a feature acquiring section for acquiring a geometric invariant feature of the pattern component as a query feature, the query feature being represented by at least three feature points including first, second and third feature points, each feature point locating on the pattern component and being retrieved from the pattern component based on a predetermined rule; a comparing section for comparing the query feature with a plurality of reference features, each reference feature representing different reference patterns prepared as candidates for pattern recognition; and a pattern determination section for determining, as a recognition result, a specific reference pattern out of the candidates based on a similarity of features therebetween and, wherein: each reference feature is

Abstract: An image processing apparatus acquires and transforms first plane image data representing a space having a depth. The apparatus includes a vanishing point estimating unit for estimating a vanishing point of the first plane image data, an angle-of-view estimating unit for estimating an angle of view of the first plane image data, and an image generating unit for generating, on the basis of the vanishing point estimated by the vanishing point estimating unit and the angle of view estimated by the angle-of-view estimating unit, second plane image data corresponding to a case in which the first plane image data is projected onto a portion corresponding to the angle of view on a curved surface of a cylinder having a predetermined radius by using, as a reference point, a position being the center of the cylinder and being equal in height to the vanishing point.

Abstract: An information processing apparatus that compares a query image and a model image and provides support information for discriminating a subject of the model image from a subject of the query image is disclosed. The information processing apparatus includes: a feature point extracting unit extracting one or more feature points from the model image; a feature describing unit describing features of the one or more feature points extracted by the feature point extracting unit; and a discrimination capability value calculating unit generating correlation images among the features described by the feature describing unit, the extracted model image, and one or more other model images for the one or more feature points extracted by the feature point extracting unit, and calculating a discrimination capability value indicating the degree of contribution to discriminating the subject of the model image on the basis of the correlation images.

Abstract: A method for following hand movements in an image flow, includes receiving an image flow in real time, locating in each image in the received image flow a hand contour delimiting an image zone of the hand, extracting the postural characteristics from the image zone of the hand located in each image, and determining the hand movements in the image flow from the postural characteristics extracted from each image. The extraction of the postural characteristics of the hand in each image includes locating in the image zone of the hand the center of the palm of the hand by searching for a pixel of the image zone of the hand the furthest from the hand contour.

Abstract: A method for determining a centroid coordinate of an image spot, comprising a) determining a search region having a border, b) identifying a first pixel in the search region, the pixel having a first intensity value, c) determining an upper threshold intensity value greater than the first intensity value, d) searching for a second pixel within the search region having an intensity value that is greater than the upper threshold intensity value, and e) upon finding the second pixel, designating the first pixel an integration region center.

Abstract: A computer-implemented method for extracting boundary elements of an object includes receiving measurement data of a point cloud of the object and a maximum edge length of a triangle to form a triangular mesh surface of the point cloud, constructing the triangular mesh surface, and extracting boundary points from the triangular mesh surface. The method further includes grouping the boundary points according to a vector relationship between one boundary point and each of the other boundary points, fitting each group of boundary points to an appropriate boundary characteristic. The appropriate boundary characteristic may be output to a storage system.

Abstract: A portion within an image in which a plurality of objects (21 and 22) overlap is partitioned into a plurality of small regions (31a and 32a), each of the units of identification information for the plurality of objects (21 and 22) is assigned to a dot corresponding to the plurality of small regions (31a and 32a), and thereby, identification-use layout information is generated, whereby even if a plurality of objects overlap upon a layout of a display image, upon the layout of the identification-use layout information that has been generated for identifying the plurality of objects, the identification information corresponding to each object does not overlap. As a result, it is possible to identify each object by the identification information acquired from the position of the cursor.

Abstract: Provided is an image quality evaluation method for evaluating image qualities of a second image by using a difference from a first image. In the image quality evaluation method, a representative pixel component value indicating a pixel component value that represents pixels in the image frame of one of the images, and pixel position information indicating a pixel position where the representative pixel component value appears are extracted as a feature quantity. By using the representative pixel component value and the pixel position information, which are the image feature quantity, and based on a difference between a pixel component value at the pixel position indicated by the pixel position information in the image frame of the other image and the representative pixel component value, a difference of the entire second image from the first image is estimated.

Abstract: The invention in particular relates to the hybrid tracking of representations of objects in a sequence of images using at least one key image. After acquiring a first and second images including a representation of the tracked object, a first image portion is identified in the first image, and a second image portion is retrieved from the key image. A relative pose of a first image portion of said second image, similar to the first image portion of the first image, is estimated. A second image portion of the first or second image, similar to the second image portion of the key image, is sought. The relative pose of the object is then estimated according to the relative poses of the first image portions and the second image portions.

Abstract: A system and method for automatically transposing an image from a circular image space to another image space, for example, horizontal. Examples of applications include a mail piece a roundel on a mail piece. On a mail piece, company name, city and state, or zip code information can be contained in the roundel instead of, for example, in the permit block. The system implements the methods electronically. Control and data information is electronically executed and stored on computer-readable media.

Abstract: A document processing apparatus includes a marking detection part that detects a marking written on the form from data read by a first reading part, an attribute name extraction part that extracts a character string described beforehand within or near a marking area of the detected marking as an attribute name, an attribute name detection part that detects the attribute name, extracted by the attribute name extraction part, stored in an attribute information memory and specifies the descriptive position of the detected attribute name from the data read by a second reading part that reads the form on which the attribute values are entered, and an attribute value extraction part that extracts the character string around the detection position of the attribute name detected from the read data, and registers the extracted character string as the attribute value of the attribute associated with the attribute name in the attribute information memory.

Abstract: A fast matching system for a digital video is provided. The fast matching system includes a video feature point extractor for extracting feature points of video frames of a digital video when the digital video is input, a feature point index mapper for receiving the video feature points from the video feature point extractor and configuring an index table by mapping the video feature points to a plurality of indices, a video feature point database (DB) for storing the index table, and a video feature point comparator for outputting video information corresponding to matched indices by comparing the video feature points extracted by the video feature point extractor with the indices of the index table stored in the video feature point DB.

Abstract: In the present invention, processing for setting a parameter expressing a measurement condition of three-dimensional measurement to a value necessary to output a proper recognition result is easily performed. The three-dimensional measurement is performed to stereo images of real models WM1 and WM2 of a workpiece using a measurement parameter set by a user, and positions and attitudes of the workpiece models WM1 and WM2 are recognized based on the measurement result. An image expressing the recognition result is displayed, and numerical data indicating the selected recognition result is set to sample data in response to a user manipulation for selecting the recognition result.

Abstract: A method of generating output image data representing a view from a specified spatial position in a real physical environment. The method comprises receiving data identifying the spatial position in the physical environment, receiving image data, the image data having been acquired using a first sensing modality and receiving positional data indicating positions of a plurality of objects in the real physical environment, the positional data having been acquired using a second sensing modality. At least part of the received image data is processed based upon the positional data and the data representing the specified spatial position to generate the output image data.

Abstract: Provided is an image matching method of matching at least two images. The image matching method extracts feature points of a reference image and feature points of a target image, changes a feature point, selected from among the feature points of the reference image, to a reference point in the target image, sets a matching candidate region on the basis of the reference point, in the target image, and performs a similarity operation between the selected feature point in the reference image and a plurality of feature points included in the matching candidate region among the feature points of the target image. The image matching method decreases the number of similarity operations performed in the image matching operation, thereby guaranteeing a high-speed operation.

Abstract: A method of processing a video sequence is provided that includes receiving a frame of the video sequence, identifying a plurality of blobs in the frame, computing at least one interior point of each blob of the plurality of blobs, and using the interior points in further processing of the video sequence. The interior points may be used, for example, in object tracking.