Abstract: This specification describes technologies relating to directional feathering of an image object. In general, one aspect of the subject matter described in this specification can be embodied in a method that includes obtaining an image object and directional information for the image object, the object including a shape having edges. The method also includes creating an opacity mask for the image object based on the shape of the image object. The method further includes processing the opacity mask such that opacity values in the opacity mask are different going from edge to interior of the shape and the opacity mask is processed in accordance with the directional information of the image object. Additionally, the method includes outputting a composite image based on the image object and the opacity mask.

Abstract: In general, the present invention is directed to systems and methods for finding the position and shape of an animal using video. The invention includes a system with a video camera coupled to a computer in which the computer is configured to automatically provide animal segmentation and identification, animal motion tracking (for moving animals), animal feature points and segments identification, and behavior identification. In a preferred embodiment, the present invention may use background subtraction for animal identification and tracking, and a combination of decision tree classification and rule-based classification for feature points and segments and behavior identification. Thus, the present invention is capable of automatically monitoring a video image to identify, track and classify the actions of various animals and the animal's movements within the image. The image may be provided in real time or from storage.

Abstract: A method for suppressing electromagnetic background radiation in the image according to a precharacterization, a device for carrying out the method according to a precharacterization, a measuring instrument, and a use of the device for identifying a laser signal as a mark for subsequent measuring processes as well as for identifying a source of a laser signal.

Abstract: Embodiments of the invention provide a method and a system for tracking an object from a training image to a target image. The training image and the target image are elements of a sequence of images. The object in the training image is represented by an object state. First, a set of particles is acquired, wherein each particle in the set of particles is associated with a weight, such that the particle represents the object state with a probability equal to the weight. Next, a regression function is applied to each particle in the set of particles based on a target image to determine a set of moved particles and the object state is updated according to the set of moved particles, such that the object state represents the object in the target image.

Abstract: A method of improving the lighting conditions of a real scene or video sequence. Digitally generated light is added to a scene for video conferencing over telecommunication networks. A virtual illumination equation takes into account light attenuation, lambertian and specular reflection. An image of an object is captured, a virtual light source illuminates the object within the image. In addition, the object can be the head of the user. The position of the head of the user is dynamically tracked so that an three-dimensional model is generated which is representative of the head of the user. Synthetic light is applied to a position on the model to form an illuminated model.

Abstract: A circuit arrangement, program product and method utilize the known view orientation for an image frame to be rendered to optimize the generation and/or use of an Accelerated Data Structure (ADS) used in physical rendering-based image processing. In particular, it has been found that while geometry primitives that are not within a view orientation generally cannot be culled from a scene when a physical rendering technique such as ray tracing is performed, those primitives nonetheless have a smaller impact on the resulting image frame, and as a result, less processing resources can be applied to such primitives, leaving greater processing resources available for processing those primitives that are located within the view orientation, and thereby improving overall rendering performance.

Abstract: A circuit arrangement, program product and circuit arrangement utilize the known view orientation for an image frame to be rendered to reposition an Accelerated Data Structure (ADS) used during rendering to optimize the generation and/or use of the ADS, e.g., by transforming a scene from which an image frame is rendered to orient the scene relative to the view orientation prior to generating the ADS. A scene may be transformed, for example, to orient the view orientation within a single octant of the scene, with additional processing resources assigned to that octant to ensure sufficient processing resources are devoted to processing the primitives within the view orientation.

Abstract: The time required to execute processing for detecting the image of a face is shortened. A number of frames of the image of a subject are obtained sequentially by sensing the image of the subject successively. Face detection processing is applied to a first subject image, which is one frame of the image of the subject among the number of frames thereof. The inclination of a face-image area obtained by detection is detected. Among angles of 0°, 90°, 180° and 270°, that close to the detected inclination is decided upon. A face image having an angle within a range of 45° on either side of the decided angle as its inclination is detected from a second subject image, which is obtained following the first subject image. Since face images to undergo detection are thus restricted, detection processing ends in a comparatively short period of time.

Abstract: Methods and apparatus are provided for imposing multiple pages on a printable sheet, in which the multiple pages include pages of varying size. The printable sheet is divided into a plurality of slots, and each of the pages is assigned to a corresponding slot. User-specified settings are received for page alignment, offset and scaling for each of the slots, and the page alignment, offset and scaling of each of the pages are individually adjusted based on the user specifications.

Abstract: There is provided a vision system and a method for scanning a traveling surface such a road to detect surface defects thereof such as cracks. The vision system, which is mountable on a vehicle, is provided with a linear imaging system for imaging successive adjacent transversal linear portions of the traveling surface as the vehicle advances. The vision system is also provided with laser line projecting means angularly projecting a laser line onto the transversal linear portion of the traveling surface to be imaged in a substantially coplanar relationship with the linear imaging system. The present vision system is particularly devised to be immune to surrounding light conditions variations to provide optimum cracks image contrast for both transverse and longitudinal cracks.

Abstract: An imaging apparatus includes face detection means for searching image data from at least one direction to detect a face portion in the image data; capture mode search direction instruction means for instructing a search direction of the image data to the face detection means in a capture mode, which is a mode of recording the image data; and monitoring mode search direction instruction means for instructing the search direction of the image data to the face detection means in a monitoring mode, which is a mode of displaying the image data.

Abstract: A method and an apparatus to obtain a surface mapping of a ballistic piece of evidence (BPOE) under examination, such as a bullet or a spent cartridge case, that can be used thereafter as a 3D signature for identifying purpose during ballistic test comparison. The method comprises providing a measurement unit adapted to acquire a relief map of the surface of the BPOE and acquiring with the measurement unit the relief map of the surface to thereby obtain the mapping of the surface of the BPOE. Preferably, the measurement unit of the present invention comprises a confocal sensor such as confocal microscope. Also, the present invention includes acquiring the relief map of the bullet surface or of the cartridge case surface by acquiring and assembling a mosaic of regional reliefs that are partly overlapping with their surroundings regional reliefs.

Abstract: When a calibration jig (50) which includes an image sensing device (100) is moved within the image sensing range of an image sensing device (200), the image coordinates of an index (110) on the image sensing device (100) in a plurality of images sensed by the image sensing device (200) are extracted. The image coordinates of environment allocation indices (300) in a plurality of images sensed by the image sensing device (100) are extracted. An information processing apparatus (1000) calculates camera parameters of the image sensing device (200) based on the extracted image coordinates of the index (110) and the environment allocation indices (300).

Abstract: A position and an orientation of an image capturing device or an object to which an inclination sensor is mounted are obtained without performing iterative calculation. A position and orientation measuring method includes the steps of: inputting a measured inclination value of an inclination sensor mounted to one of an object and an image capturing device; inputting a captured image from the image capturing device; detecting an index on the object from the captured image; and calculating a position and an orientation of one of the object and the image capturing device to which the inclination sensor is mounted on the basis of the measured inclination value, an image coordinate of the detected index, and already-known positional information on the detected index without performing iterative calculation.

Abstract: A zooming system includes a selecting unit, a detection unit, a calculating unit, a controlling unit, and a zooming unit. The selecting unit is configured for selecting a zoom mode based on the presence of faces in a pre-image. The detection unit is configured for detecting facial areas in the pre-image. The calculating unit is configured for calculating a zoom ratio on the basis of a dimensional ratio of the detected facial areas to the pre-image. The zooming unit is controlled by the controlling unit to zoom the imaging device according to the zoom ratio.

Abstract: A method, computer program product, and apparatus for specifying, making, and displaying measurements within a 3D range data set obtained using a 3D rangefinder device. A 3D range data set is provided and represented as a first 2D displayed image and a second displayed image. The second displayed image is selected from the group consisting of a second 2D displayed image and a 3D displayed image. At least one feature is identified in the first 2D displayed image and the second displayed image, such that a measurement can be computed based on the identified features. The computed measurement is displayed in at least one of the first 2D displayed image and the second displayed image.

Abstract: An image processing system (250) and method (300) are disclosed for correcting a head pose in a video phone image, so that a frontal view is presented on a display. A disclosed head pose corrector (250) estimates the orientation of a head pose and adjusts the orientation of the head pose, if necessary, to present a frontal view. The orientation of the head pose is adjusted by generating a three dimensional model of the face surface and adjusting the orientation of the three dimensional face model to provide the desired frontal view. The head pose corrector (250) may be included in a video phone (100) to correct the head pose of transmitted or received images (or both) or may be included in a server on a network to automatically adjust the head shots of one or more participants to a video phone communication.

Abstract: An image output apparatus for automatically laying out a number of captured images in a composition without uncomfortable feeling and outputting the same is provided. The image output apparatus includes: an object orientation identification section for identifying the orientation of an object in an image; an image layout section for laying out the image within an output region based on the object orientation identified by the object orientation identification section; and an image output section for outputting the image laid out within the output region by the image layout section.

Abstract: Disclosed are systems for and methods of registering a deformable image with a reference image subject to a local or global quality of match. These systems and methods may use the quality of match to reduce effects of artifacts within the deformable image or the reference image on a deformation algorithm. The registration includes modifying the deformable image to match similar features in the reference image. The systems include a deformation engine configured for performing the deformation algorithm subject to the local or global quality of match. The systems and methods allow for correcting for cone beam computed tomography artifacts in one or more of the images.

Abstract: The present invention is directed to various methods and systems for analysis and processing of video and audio signals from a plurality of sources in real-time or off-line. According to some embodiments of the present invention, analysis and processing applications are dynamically installed in the processing units.

Abstract: A camera for aerial photography includes a reference projector that projects collimated beams of light into the optical system entrance aperture which are imaged by an image recording device. The beams produce fiducial images present in every image captured by the imaging recording device. The fiducial images can be used to characterize the performance of the optical system and derive distortion correction coefficients. The distortion correction coefficients can be applied to a portion of an image, such as a group of pixels, or to the entire image, to thereby compensate for distortions in the optical system. In some embodiments, e.g., airborne cameras, the projector is rigidly coupled to an inertial measurement unit. The ability of the airborne camera to perform object geolocation from imagery is improved.

Abstract: An image scanning apparatus moves a copyboard provided between an original document and three line sensors and the three line sensors relatively to each other at a speed different from the speed of conveying an original document. An in-line correction portion allows three data output by three line sensors to be synchronized so that a foreign object on the copyboard appears at different positions of the original document in chromatic color. A foreign object position detection portion detects a position on the copyboard of a foreign object before an original document is conveyed, and a color determination portion decides a determination pixel for use in determination from a plurality of pixels included in each of scanned data obtained while the original document is conveyed, based on the position of the foreign object, and determines whether the original document is color or black-and-white based on the determination pixel.

Abstract: An image processing apparatus detecting, from an input image, a still-zone area disposed at an edge of the input image and displaying a same image continuously, the apparatus including: a difference calculation mechanism calculating a difference between pixel values of pixels located at a same position of the input image in different frames with each other for each pixel of the input image; a pixel-motion-information generation mechanism generating pixel-motion information indicating whether there is a change in the pixel values of the pixels of the input image on the basis of the difference; and a first detection mechanism comparing the pixel-motion information of a plurality of the pixels arranged in a first direction in the input image, and detecting a boundary position in the first direction between the still-zone area and an effective area excluding the still-zone area in the input image.

Abstract: In accordance with an image processing method, a data region is detected from input image data. The input image data is rotated in accordance with inclination of an image. Determination is made whether the data region of rotated image data protrudes from an image area of the input image data. When detection is made of protruding, the smallest area including the data region is extracted from the rotated image data. When the data region of the rotated image data protrudes from the image area of the image data previous to rotation, the smallest region including the data region of the rotated image data is extracted from the rotated image data. Therefore, loosing information from image data subjected to rotational correction can be prevented. Also, the amount of image data subjected to rotational correction can be minimized.

Abstract: A fast decoding technique for decoding a position of a bit in a pattern provided on a media surface that can generate large amounts of solution candidates quickly by switching or flipping bits and utilizing a recursion scheme. The fast decoding technique may be employed to simultaneously decode multiple dimensions of a pattern on the media surface.

Abstract: A method of determining an anatomically consistent scan protocol for an object of interest includes obtaining a volumetric image of an object of interest to be imaged, transforming the volumetric image, estimating the position and orientation of the object using the volumetric image and the transformed volumetric image, and modifying the imaging scan protocol using the estimated object position and orientation.

Abstract: According to an aspect of the present invention, once the mode setting device sets the person photographing mode, and the face zooming instruction device instructs zooming to the face of the person, the detection device detects the size of the face of the person captured by the image pickup device. The zooming device is then controlled so as to capture the face of the person in a certain size based on the detection result. Subsequently, the photographer instructs zooming in the telephoto direction or the wide direction using the zooming instruction device as necessary to thereby allow fine adjustment of the composition. This enables to easily and quickly photograph an image in a user's desired composition.

Abstract: A system for automatically identifying sperm cells in a smear on a microscope slide, including capturing a first digital color image within an area of interest on the microscope slide. The first digital color image is split into a plurality of component color space images and stored into a plurality of memory spaces. The component color space images are manipulated by mathematical functions, logical functions or a combination of mathematical and logical functions to produce a result image. Thresholding the result image creates a binary result image which is processed with particle (blob) analysis. A set of blob factors is applied to identify probable sperm cells; and a list of universal coordinate system positions of the probable sperm cells is created.

Abstract: An image processing apparatus includes a detecting unit for detecting a position of a facial feature in a face image, a principal component analysis performing unit for performing principal component analysis on the position of the facial feature in a registered image that is a pre-registered face image, a first evaluation value calculating unit for calculating a first evaluation value that is used to evaluate the position of the facial feature in a target image relative to the principal component obtained in the principal component analysis operation, and a first determining unit for determining whether or not the face in the target image matches the face in the registered image by comparing the calculated first evaluation value with a first threshold set based on the position of the facial feature in the registered image relative to the principal component.

Abstract: An image calibration method according to the present invention includes the steps of receiving a calibration request from a user terminal 20 (S10), receiving a calibration object image photographed in accordance with an instruction for calibration photographing (S16), performing a calibration process using the received calibration object image (S26), receiving a subject image photographed by a camera 2, transmitted from the user terminal 20 (S30), performing a calibration on the subject image in accordance with results of the calibration process to form a corrected subject image (S32), and returning the formed corrected subject image to the user terminal 20 (S34). The present invention allows acquisition of a corrected subject image in which distortions due to the camera lens are corrected even for those unskilled in three-dimensional measurement based on the principle of the stereo method, precise two-dimensional measurement, and precise image formation.

Abstract: The present invention provides a three-dimensional image measuring apparatus and method capable of measuring projections and depressions on a surface of an object with fine precision, as well as ensuring stable convergence, even for stereo images with significant project distortion.

Abstract: A part object which forms a model object is disposed by receiving a disposition direction for the part object and updating slot state data indicating a disposition state of the part object in a part slot provided corresponding to each part object. When another part object has been disposed in the part slot corresponding to the part object for which the disposition direction has been issued, whether or not to update the slot state data of the part slot is determined based on priority information which determines disposition priority of each part object, and the slot state data is updated based on the determination result.

Abstract: A printing device enables a laid-out and drawn image to be printed without any subsequent rotation. In the presence of an orientation change command, the printing device receives a layout command and determines a layout of each object in each page on the medium with conversion of coordinates at a preset position of the object. In the presence of the orientation change command, the printing device receives a print command, fetches an object according to the stored layout in each page, and draws and prints the fetched object in response to the orientation change command. No subsequent orientation change of drawing data, which has been laid out and drawn in the specified page orientation, is required to generate converted drawing data corresponding to the specified medium orientation.

Abstract: An ultrasonic image processing apparatus allowing an ultrasonic image of an inspection object to be displayed using echo data in a three-dimensional region constructs a three-dimensional image based on the echo data by an image construction section. An image operation input section performs an operation for changing the display state of the constructed three-dimensional image. Also, an image display state changing section changes the display state of the three-dimensional image that has been constructed based on input information inputted by the image operation input section.

Abstract: An index detection unit (2030) detects indices allocated or set on an object from a sensed image. An evaluation amount calculation unit (2060) calculates evaluation amounts of the indices using two-dimensional geometric features of the indices on the image and/or three-dimensional geometric features that represent relationships between an image sensing device (2010) and the indices on a three-dimensional space. A reliability calculation unit (2070) calculates reliabilities of the indices according to the calculated evaluation amounts of the indices. A position and orientation calculation unit (2080) calculates the position and orientation of the object or the image sensing device (2010) using at least the calculated reliabilities of the indices and information associated with the image coordinates of the detected indices.

Abstract: A parameter selection section selects one from a plurality of image synthesis parameters stored in a parameter storage section according to an output of a vehicle state detection section. An image synthesis section reads out camera images from frame memories according to the selected image synthesis parameter and generates a synthesized image showing a state of vehicle surroundings. A moving object detection section detects, in the thus generated synthetic image, a moving object region estimated as a region where a moving object is present.

Abstract: Upon synthesizing objects, information bits indicating the types of objects are lost. To solve this problem, this invention provides an image processing apparatus having discrimination means for discriminating a type of object to be rendered, determination means for determining the presence/absence of synthesis of the discriminated object, synthesis means for synthesizing an object and information of the type of object in accordance with the determination result, and processing means for appending information indicating the type of synthesized object to a rendering result obtained by rendering the object to be rendered in units of pixels.

Abstract: The invention concerns a method of detecting from a vehicle a bend in a road comprising a surface and road edges, comprising the following operations: taking an image of a road scene unfolding in front of the vehicle and at least partly illuminated by the vehicle, determining, for each pixel in the image a light decrease gradient, analysing these decrease gradients and determining an image of the road edges, mathematically discriminating the gradients from the image of the road edges, analyzing this discrimination and determining edges, and an angle of the bend. The invention also concerns a system for implementing the method of the invention comprising a camera mounted in the vehicle, an image processing unit and a neural network.

Abstract: Method of determining the position of a target drill hole to be drilled in a target corrective lens having an expected target outline after shaping, the position being determined from a reference lens having a reference outline and at least one reference drill hole, the method includes: acquiring an image and at least one characteristic of the curvature of the reference lens; determining, in the acquisition plane, the reference distance in projection between the projection of a reference anchor point of the associated reference lens and the reference lens and the projection of a reference drilling point of the reference drill hole calculating the three-dimensional reference distance between the reference anchor point and the reference drilling point as a function of the characteristic of the curvature of the reference lens (100) and of the determined reference distance; and determining the position of the target drilling point for the target drill hole of the target corrective lens as a function of the calcul

Abstract: An image processing device has: a main subject detector that detects the position of a main subject in an input image; a clipping region setter that determines a clipping region including the position of the main subject detected by the main subject detector; and a clipper that generates a clipped image by cutting out the clipping region from the input image. The clipping region setter determines the clipping region such that the position of the main subject detected by the main subject detector coincides with a predetermined position in the clipping region.

Abstract: A method and an apparatus estimate an object part location in a digital image using feature value analysis. The method according to one embodiment accesses digital image data representing a region including an object part of a digital image; accesses reference data including class data of classes relating to predetermined positions of the object part in predetermined regions, and features that discriminate among the classes; calculates feature values for the features in the region using pixel values within the region; and determines a location estimate of the object part using the feature values and the reference data.

Abstract: Voting of coordinates of each pixel in a photographic image onto a Hough space corresponding to annuluses is performed, the Hough space being defined by a coordinate system having an X coordinate axis for a circle center point, a Y coordinate axis for a circle center point, and an r coordinate axis for a radial direction. A vote value of each voting position having coordinates (X, Y, r) on the Hough space is acquired as a first integrated vote value from a calculation of a number of votes given to each voting position. A position represented by (X, Y) coordinate values of a voting position associated with the largest first integrated vote value is detected as the position of the center point of the circular pattern.

Abstract: A CPU 411 splits selected image data GD into a plurality of areas (pixel data groups), analyzing the image data GD in terms of area units, and determining the color range of the areas of the image data GD. The CPU 411 determines color range areas by associating adjacent areas of the same color range when determining the color range for the areas of the image data GD. The CPU 411 acquires position data for the color range areas that have been determined. The CPU 411 acquires position conditions (main object conditions) from ROM/HDD 413, and narrows the main object candidates to “sky, green, or people.” The CPU 411 finally determines the main object by comparing the position data of the color range color areas and the position conditions corresponding to the narrowed main object.

Abstract: The present invention is directed to diagnostic imaging of small pulmonary nodules. There are two main stages in the evaluation of pulmonary nodules from Computed Tomography (CT) scans: detection, in which the locations of possible nodules are identified, and characterization, in which a nodule is represented by measured features that may be used to evaluate the probability that the nodule is cancer. Currently, the most useful prediction feature is growth rate, which requires the comparison of size estimates from two CT scans recorded at different times. The present invention includes methods for detection and feature extraction for size characterization. The invention focuses the analysis of small pulmonary nodules that are less than 1 centimeter in size, but is also suitable for larger nodules as well.

Abstract: A method characterizes field of view (FOV) distortion and may correct machine vision measurements accordingly. A plurality of distorted images are acquired, with the same calibration target at a plurality of spaced-apart locations within the FOV. The method analyzes the images and determines distortion parameters that can correct the distortions in the images such that features included in the calibration target pattern achieve a sufficient or optimum degree of congruence between the various images. Distortion parameters may be determined for each optical configuration (e.g., lens combination, magnification, etc.) that is used by a machine vision inspection system. The distortion parameters may then be used to correct FOV distortion errors in subsequent inspection and measurement operations.

Abstract: A video game device calculates an angle of an object 71, etc., with respect to a reference direction in a three-dimensional space. Then, the video game device defines a virtual cylindrical model for each of a plurality of objects so that a bottom surface of each cylindrical model is perpendicular to the reference direction. The video game device defines a plurality of cylindrical models for a predetermined object if the inclination of the predetermined object with respect to the reference direction is greater than a predetermined angle. Then, the video game device determines whether or not the defined cylindrical models have an overlap therebetween. Then, if two cylindrical models are determined to be overlapping with each other, the video game device gives a predetermined change to the two objects in response to the collision.

Abstract: A method of reconstructing an original image from a printed and scanned image includes determining a plurality of transformation parameters including a rotation a of the scanning grid relative to the printing grid and a plurality of translation parameters which give a position of a node on the printing grid in a coordinate system of the scanning grid, generating a system of non-linear equations which represent the printed and scanned image as a function of: (i) the original image, (ii) a print/scan resolution ratio, and (iii) the transformation parameters, computing coefficients for the system using the resolution ratio, the transformation parameters and the given print model, generating a linear approximation of the system of non-linear equations, solving the linear approximation for an approximation of the original image, and creating a reconstructed original image from the approximation of the original image.

Abstract: A method of detecting an image of a reflective object includes the steps of capturing an exposure image by an image sensor while an LED is ON; down-sampling an exposure image signal from the image sensor to a first digital signal by comparing the level of the first image signal with a first reference level; capturing a dark image by the image sensor while the LED is OFF; down-sampling a second image signal from the image sensor to a second digital signal by comparing the level of the second image signal with a second reference level; computing an image of difference between the exposure image and the dark image by subtracting the second digital signal from the first digital signal; and specifying the image of the reflective object by locating a bright point remaining in the image of difference.

Abstract: A vehicle drive assist system comprises a camera for picking up an image of an area existing in an advancing direction of a vehicle; display means for displaying the image picked up by the camera; steering angle detecting means for detecting a steering angle for steering the vehicle; traveling path predicting means for predicting a traveling path of the vehicle on the basis of the steering angle detected by the steering angle detecting means; and drive assist means for overlaying on the display means drive assist information containing the vehicle predictive traveling path predicted by the traveling path predicting means and guide lines prolonged from the lines defining the width of the vehicle body on the image of the area existing in the vehicle advancing direction.

Abstract: A test image has a black bias on a white background. The black bias is a line set at about 45 degrees to the scan lines of a scanner. Boundary points of the scanned bias are found. A regression line is calculated from the positions of the boundary points. Differences in the positions of adjacent boundary points, together with the slope reciprocal of the regression line, are used to determine error values. The error values are compared with a gate value to determine if there are any occurrences of scan line misalignment.