Abstract: A history image generating system and method is configured such that a determining unit determines whether or not an image to be combined with a history image stored as a history is a digital watermark image, and when it has been determined by the determining unit that the image is a digital watermark image, a processing unit performs, with respect to the digital watermark image, processing including enhancement processing such that the visibility of the image is increased, and a generating unit generates an image in which the digital watermark image with respect to which processing has been performed by the processing unit and the history image are combined. Further, a recording medium stores a computer program that causes a computer to execute operation of each unit of this history image generating system.

Abstract: Methods and apparatus to specify regions of interest in video frames are disclosed. An example disclosed method comprises determining an initial template region to represent a region of interest whose location is based on a first point selected in a graphical presentation, determining a first modification to perform on the initial template region in response to a second point selected in the graphical presentation, detecting the second selected point in the graphical presentation, and reshaping the initial template region toward the second selected point, the reshaping corresponding to the first modification, the reshaping being performed in response to detecting the second selected point without also requiring the user to select any point substantially on the boundary defining the initial template region to initiate the reshaping.

Abstract: The present invention discloses an object image detection method, which uses a coarse-to-fine strategy to detect objects. The method of the present invention comprises steps: acquiring an image and pre-processing the image to achieve dimensional reduction and information fusion; using a trained filter to screen features; and sequentially using a coarse-level MLP verifier and a fine-level MLP verifier to perform a neural network image detection to determine whether the features of the image match the features of the image of a target object. The present invention simultaneously uses three mainstream image detection methods, including the statistic method, neural network method and adaboost method, to perform image detection. Therefore, the present invention has the advantages of the rapidity of the adaboost method and the accuracy of the neural network method at the same time.

Abstract: According to an apparatus and method for detecting an obstacle through stereovision, an image capturing module comprises a plurality of cameras and is used for capturing a plurality of images; an image processing module edge-detecting the image to generate a plurality of edge objects and object information corresponding to each edge object; an object detection module matches a focus and a horizontal spacing interval of the camera according to the object information to generate a relative object distance corresponding to each edge object; a group module compares the relative object distance with a threshold distance and groups the edge objects with the relative object distance smaller than the threshold distance to be an obstacle and obtains a relative obstacle distance corresponding to the obstacle.

Abstract: A method of determining in a three-dimensional operating space a measure of super-imposition M between at least a portion of an object and a spatial representation of the object or part thereof in an object representation pose, a pose comprising position and orientation in operating space includes: a) obtaining a similarity measure S between a first imaging data taken in first respective imaging poses of the object or part thereof and second imaging data comparable to the first imaging data taken in second respective imaging poses that are digitally reconstructed from the spatial representation of the object or part thereof in the object representation pose; b) locating in operating space at least one point on the surface of the object or part thereof; c) obtaining a distance measure D between the at least one point in operating space and a surface of the spatial representation of the object or part thereof in the object representation pose; and d) obtaining a measure of superimposition M by a combination

Abstract: An object recognition apparatus sets each of keypoints extracted from a typical image and one of keypoints extracted from an input image, having scale invariant features similar to each other, as a typical corresponding point and an object corresponding point matching with each other, and produces a position vector directed from a typical reference point of the typical image to each typical keypoint. The apparatus determines a position of an object reference point in the input image from a position of each object corresponding point and the vector of the typical corresponding point matching with the object corresponding point. When the positions of the object reference point are concentrated, the apparatus judges that an object picture having the object corresponding points in the input image matches with the typical image, and the apparatus recognizes the picture as an object indicated by the typical image.

Abstract: An image processing apparatus includes a moving image file storage unit operable to store a moving image file; an area selection receiving unit operable to receive a selection of a predetermined area corresponding to one of a plurality of frames forming the moving image file; a template image generating unit operable to generate as a template image an image of the selected area; an image matching unit operable to obtain the frames from the moving image file storage unit, and to match each of the frames against the template image to search for a matching frame containing an image similar to the template image; and a scene change detecting unit operable to detect a start and an end of a scene containing the matching frame.

Abstract: Pixel data is stored in a field corresponding to a pixel of an image. The field includes an opacity bit and a set of bits. If the opacity bit is set to an opaque-indicating state, then a color level value of a component color of the pixel is stored in the set of bits. However, if the opacity bit is set to a transparency-indicating state, then a transparency level value is stored in at least one bit of the set of bits and the respective color level value is stored in other bits of the set of bits.

Abstract: A method for classifying a handwritten input character is disclosed. Character models are used. Each character model is associated with an output character and defines a model specific segmentation scheme for that output character and an associated segment model. The model specific segmentation scheme defines a minimum length corresponding to a number of points in a stroke of the output character and a minimum length threshold. Using each of the character models, the input character is decomposed into segments and the segments are evaluated against the segment model of the respective character model to produce a score indicative of the conformity of the segments with the segment model. The character model that produced the highest score is selected and the input character is classified as the output character associated with the character model that produces the highest score.

Abstract: A vehicle surrounding monitoring system includes an image-capturing portion, a storage portion, a display-image generation portion, and a display portion. The image-capturing portion captures the image of at least a portion of the area surrounding the vehicle. The storage portion stores the image captured by the image-capturing portion. The display-image generation portion generates a synthesized image for display using a latest image that is captured by the image-capturing portion, and a previous image that is captured before the latest image is captured. The synthesized image is displayed on the display portion. The display-image generation portion decreases the level of visibility of an image region relating to the previous image, according to an elapsed time after the previous image is captured.

Abstract: The present invention relates to a Visual Background Extractor (VIBE) consisting in a method for detecting a background in an image selected from a plurality of related images. Each one of said set of images is formed by a set of pixels, and captured by an imaging device.

Abstract: An image generation apparatus includes an information acquisition section and an image generation section. The information acquisition section acquires record information to be recorded on a recording medium. The image generation section generates an image including code pattern images and a latent image for checking paper being copied. The code pattern image represents the record information acquired by the information acquisition section. The latent image for checking paper being copied includes second pattern images each having a smaller size than each code pattern image and third pattern images each having a larger size than each code pattern image.

Abstract: A method of classifying a character string formed from a known number of hand-written characters is disclosed. The method starts by determining character probabilities for each hand-written character in the character string. Each character probability represents a likelihood of the respective hand-written character being a respective one of a plurality of predetermined characters. Each predetermined character has a respective character type. Character templates having the known number of characters are next identified. Each character template has a respective predetermined probability and represents a respective combination of character types. Character sequence probabilities corresponding to each of the character templates having the known number of characters are next determined. The character sequence probabilities are a function of the predetermined probability of the respective character template and the character probabilities of the hand-written character in the character string.

Abstract: An occupant information detection system for detecting information related to a vehicle occupant on a vehicle seat includes an image detecting section facing a vehicle seat area taking a three-dimensional image, an information extraction processing section extracting information related to a three-dimensional surface profile of a detected object on a vehicle seat based on the three-dimensional image, a conversion processing section converting the extracted information into numerical coordinates of a predetermined coordinate system, and a computation processing section deriving a computed value by subtracting an average of first and second coordinate components from a third coordinate component. The system also includes a determination processing section determining the concave-convex shape of a line segment formed by the arbitrary points and the middle point based on the computed value.

Abstract: A system and method of processing infrared imagery to determine projectile trajectories and the locations of shooters with a high degree of accuracy. The method includes image processing infrared image data to reduce noise and identify streak-shaped image features, using a Kalman filter to estimate optimal projectile trajectories, updating the Kalman filter with new image data, determining projectile source locations by solving a combinatorial least-squares solution for all optimal projectile trajectories, and displaying all of the projectile source locations. Such a shooter-localization system is of great interest for military and law enforcement applications to determine sniper locations, especially in urban combat scenarios.

Abstract: A method and system for determining the three-dimensional location and orientation of a medical device distal end using a single-plane imaging system, using a computational model of the medical device and a transfer function for the medical device describing local device shape and orientation in response to user or computer determined inputs. The method allows guidance of an interventional medical system to a set of target points within the patient using a single-projection imaging system.

Abstract: The likelihood of a particular type of object, such as a human face, being present within a digital image, and its location in that image, are determined by comparing the image data within defined windows across the image in sequence with two or more sets of data representing features of the particular type of object. The evaluation of each set of features after the first is preferably performed only on data of those windows that pass the evaluation with respect to the first set of features, thereby quickly narrowing potential target windows that contain at least some portion of the object. Correlation scores are preferably calculated by the use of non-linear interpolation techniques in order to obtain a more refined score. Evaluation of the individual windows also preferably includes maintaining separate feature set data for various positions of the object around one axis and rotating the feature set data with respect to the image data for the individual windows about another axis.

Abstract: An X-ray CT apparatus includes a unit which extracts a high-contrast region of comparatively high X-ray attenuation coefficient from an image, units which generate an unsharp image concerning the high-contrast region, on the basis of the position of the extracted high-contrast region and a point spread function peculiar to the apparatus, and a unit which subtracts the unsharp image from the original image in order to generate a low-contrast image concerning a low-contrast region of comparatively low X-ray attenuation coefficient.

Abstract: In a digital camera 1, when the shutter button is half-depressed (Y at S2), face detection processing is performed on periodically imaged frame images until it is fully depressed (N at S6), and face detection results are cyclically stored in a detection result storage area. When it is fully depressed (Y at S6), a still-image imaging operation is performed, and the face detection processing is performed on an imaged still-image (S7 and S8). Then, after a high priority face detection result is selected from among the face detection results currently stored in the detection result storing area (S9), the selected face detection result and the face detection result of the still-image are compared and judged whether or not they are in agreement (S10 and S11). When judged not to be in agreement, the still-image is displayed as a preview and the warning display of this disagreement is performed (S13).

Abstract: A system and method is provided for using backscattered data and known parameters to characterize vascular tissue. Specifically, in one embodiment of the present invention, an ultrasonic device is used to acquire RF backscattered data (i.e., IVUS data) from a blood vessel. The IVUS data is then transmitted to a computing device and used to create an IVUS image. The blood vessel is then cross-sectioned and used to identify its tissue type and to create a corresponding image (i.e., histology image). A region of interest (ROI), preferably corresponding to the identified tissue type, is then identified on the histology image. The computing device, or more particularly, a characterization application operating thereon, is then adapted to identify a corresponding region on the IVUS image. To accurately match the ROI, however, it may be necessary to warp or morph the histology image to substantially fit the contour of the IVUS image.