Abstract: A pattern evaluation method includes processing image data of at least one pattern serving as an object to be evaluated and detecting coordinates of edge points of the pattern in an image of the image data, making pairs of edge points from the edge points of the pattern, setting an arbitrary axis, calculating a distances between the edge points of each pair of the pairs of edge points and an angle between a straight line connecting the edge points of the pair and the axis, preparing a distance/angle distribution map which represents distribution of the distances and angles of the pairs of edge points, extracting a characteristic point of the distance/angle distribution map and analyzing the pattern on the basis of the extracted characteristic point.

Abstract: Provided is a displacement sensor which allows at least part of the data used from the time of obtaining an image until the time of computing the displacement can be readily verified. In a displacement sensor for automatically extracting a coordinate of a measuring point from an image obtained by using an imaging device according to a prescribed measuring point extraction algorithm, and computing a desired displacement from the automatically extracted measuring point coordinate, the sensor is further provided with the function to edit data used from the time of obtaining the image until the time of computing the displacement for use as display data for an image monitor.

Abstract: An address reading method with processing steps controlled by parameters, in which free parameters which cannot be adapted by learning samples are to be automatically optimized. These parameters are therefore assigned costs. The value of free parameters which are expensive and lie above selectable cost thresholds are maintained and the remaining free parameters are improved by repeatedly modifying their values on the basis of strategies known per se, taking already evaluated parameter settings into account, and training and evaluating the reading method only with these modified values.

Abstract: A rectangle finder tool automatically locates rectangles in an image that correspond to substantially rectangular devices in the image. The rectangle finder tool can be used to automatically locate rectangles in an image or to generate abstract rectangular models of devices in the image. The tool determines the properties of and locates the rectangular device by searching for pairs of lines in an image at a specified orientation.

Abstract: A payment product has a writing area (6c) which is intended for a user's signature. In the writing area there is a first position-coding pattern (5) which makes possible digital recording of the signature. The first position-coding pattern is a subset of a larger second position-coding pattern. The payment product is used in a payment system which is based on electronic payment information, which has been recorded by means of the position-coding pattern, being sent to a server unit, which utilizes the position-coding pattern to check that the payment information is valid.

Abstract: An image processing system detects a plurality of image features in a first image corresponding to a first view of a scene, and a plurality of image features in a second image corresponding to a second view of the scene. The second image deviates from the first image as a result of camera relative motion. The system determines a two-view correspondence resulting in a potential match set having a maximum average strength of correspondence based at least in part on the total number of matching neighbor candidate image features. Additionally, a multiple-view correspondence between images results in a potential match set based at least in part on a computation of reprojection error for matched points that resulted from a projective reconstruction of the potential match set.

Abstract: When application domain structure information is erroneously encoded into parameters for image processing and measurements the accuracy of the result can degrade. A structure-guided automatic alignment system for image processing receives an image input and application domain structure input and automatically creates an estimated structure output having improved alignment. Measurement and image processing robustness are improved.

Abstract: A data processing apparatus for inputting data by writing characters on a touch sensitive display screen. The data processing apparatus comprises a character recognition processor operable to generate an estimate of a character hand written by a user on the touch sensitive screen. The data processing apparatus includes a processing unit operable to receive the estimated character, and a graphics display device operable to receive the estimated character from the processing unit. The graphics display device is operable to display the estimated characters within a text input window of the display screen. The processing unit is operable in combination with the graphics display driver to display the estimated character on the display screen, substantially at a position proximate to a location to where the user has written the character.

Abstract: A method of determining a vanishing point related to an image, the method includes the steps of: detecting line segments in the image; determining intersections from pairs of line segments; assigning a probability to each intersection of the pairs of line segments; determining a local maximum corresponding to a plurality of probabilities; and outputting an estimated vanishing point vector &ngr;E that corresponds to the determined local maximum such that an estimated location of the vanishing point about the estimated vanishing point vector &ngr;E results.

Abstract: A method and apparatus for identifying the orientation and positions of the elements of a grid in an image. The elements of the grid need not be parallel to the sides of the image, and the image need not contain all of the elements of the grid. The method can be applied to multi-sided grids, whether rectilinear or distorted. The addresses of identified grid elements can be used to control, e.g., analytical procedures directed to individual elements.

Abstract: A process for detection of straight-line segments in a stream of digital data that are representative of an image (m, n) in which the contour points of said image each defined by the modulus and the orientation &thgr; of their gradient in relation to a horizontal axis are identified. The stream of digital data is stored in the form of two successive lines, each of n points P(i, j), for each of the lines the n points of said line are extracted successively, and for each point Pc(i, j) extracted, the so-called current point, the three points having inferior coordinates Pa(i−1, j−1), Pb(i, j−1), Pd(i−1, j) directly neighbouring said current point are extracted, successively the value of the orientation of the current point Pc is compared with that of the neighbouring points Pa, Pb and Pd. The values of the orientations of the points Pb and Pd are compared, the result of a comparison is validated if the points are contour points and if the values of their orientation are close.

Abstract: A process for detection of straight-line segments in a stream of digital data that are representative of an image (m, n) in which the contour points of said image each defined by the modulus and the orientation &thgr; of their gradient in relation to a horizontal axis are identified. The stream of digital data is stored in the form of two successive lines, each of n points P(i, j), for each of the lines the n points of said line are extracted successively, and for each point Pc(i, j) extracted, the so-called current point, the three points having inferior coordinates Pa(i−1, j−1), Pb(i, j−1), Pd(i−1, j) directly neighbouring said current point are extracted, successively the value of the orientation of the current point Pc is compared with that of the neighbouring points Pa, Pb and Pd. The values of the orientations of the points Pb and Pd are compared, the result of a comparison is validated if the points are contour points and if the values of their orientation are close.

Abstract: A method of deriving geometrical data of a structure from an image of the structure comprises the selection of marker points in the image. A typical contour is associated with the marker points and the geometrical data is calculated from the typical contour. According to the invention, correctness values are associated with the marker points. The correctness values indicate the reliability of the correspondence of the marker points to typical (anatomical) features in the structure. Furthermore, a cost function is associated with a typical contour, which cost function is dependent on the correctness values of the marker points. The typical contour is arranged in such a manner that the cost function reaches an optimum value, such as a local maximum or minimum value. The method is particularly suitable for deriving geometrical data, such as angles and distances concerning the spinal column of a patient.

Abstract: A process for generating a dynamic and contextualized cartography of information including a step of collaborative filtration which includes determining at least one affinity group formed from a set of profiles, wherein the distance from the profile to a reference object is less than a threshold value, a step of correlation analysis of characteristics of the reference object and objects of the affinity group projecting the characteristics into a multidimensional space, and a step of construction of an image composed of activatable zones representing the characteristics of the reference object and of all or part of the characteristics of the affinity group, and zones representing links among the characteristics, each zone providing direct or indirect access to one or more digital files.

Abstract: A method and apparatus for identifying the orientation and positions of the elements of a grid in an image. The elements of the grid need not be parallel to the sides of the image, and the image need not contain all of the elements of the grid. The method can be applied to rectilinear and curved grids. The addresses of identified grid elements can be used to control, e.g., analytical procedures directed to individual elements.

Abstract: An information code converting apparatus converts object information acquired with respect to an object into a registered information code in association with the object, and transmits or outputs the information code in combination with a configuration code. A reproduced information converting apparatus converts the information code transmitted or inputted from the information code converting apparatus into registered reproduced object information in association with the information code, configuring and reproducing an image of the object. The information code converting apparatus has a parts warehouse database storing object data generated in association various objects, and the reproduced information converting apparatus has a reproduced parts warehouse database storing object data generated in association with the various objects.

Abstract: An image detection method is disclosed wherein local edge information within an image is utilized for accurately detecting lines and curves of the images. The method can more accurately determine the location and orientation of a pattern and thus provides more reliable image match for pattern detection. A detector is trained off-line with example images resulting in a stored template generated by recording a test pattern similar to a pattern to be tested; anchor lines are identified within said template. Long lines are detected in subsequent test patterns using the disclosed method. The template is rotated and shifted before matching it to the test pattern so that anchor lines align with long lines detected within the test pattern. The template and test pattern are then compared to determine whether there is a match.

Abstract: I present a method for matching the spatial relationships between an input set of feature points and a template set of feature points. A feature point consists of a location in space and a label describing the feature at that location in space. A tessellation over the feature point locations is performed. Next, a search identifies polyhedra that have similar contents, the contents being the angles and labels associated with feature points of the polyhedra. Once a match is found, then appropriate adjacent and neighboring polyhedra are examined. Matching the node labels and angular relationships for a set of appropriate adjacent and neighboring polyhedra extends the volume over which matches exist and significantly increases the certainty that a positive match exists.

Abstract: There is disclosed a method of automatically analyzing many repeating patterns. One typical pattern is first selected. On the typical pattern, analysis points, analysis methods and analysis conditions are determined. On the other patterns, automatic analysis is performed at points having the same positional relation as the analysis points determined on the typical pattern with the analysis methods and analysis conditions determined for the typical pattern.

Abstract: The invention is a system and method for deriving a single dimensional representation for a set of points e.g.,minutiae, in image of a two dimensional pattern of lines, e.g. a fingerprint, by creating a one dimensional (string) representation of one or more points (e.g., minutiae) and the respective attributes of each point therein. A landmark point is selected from the two dimensional image, preferably from the set of the points to be represented in single dimension. The relationships of each of the points with reference to the landmark determines a linear order for the points and the attributes associated with each point.

Abstract: The invention is of an “elastic matching” method for comparing one signature against another, comprising the operations of creating a mapping between corresponding points in two signatures to be compared measured at different times after the commencement of writing the signatures which maximizes the correlation between the local spatial neighborhood of the measured points and simultaneously minimizes the curvature of the elastic spatial distortion from the mapping, providing quantitative measures of both the degree of the spatial correlations and the degree of the non-uniform spatial distortions in the mapping, thereby providing measures of the similarity between the signatures.

Abstract: Patterns in an image or graphical representation of a dataset are identified through centroid integration. The image or graphical representation is digitized. A collapsed image is created by identifying the centroid and at least one characteristic value for each feature in the digitized image. A shape, such as a line, curve, plane or hypersurface, is stepped across the image. At each step, the characteristic values are summed for all centroids within a predetermined distance of the shape. Peaks in the resulting summation represent matches between the shape and a pattern in the data, with steeper peaks representing a better match. Different shapes, or the same shape at different angles, can be applied to the data to find better fits with the patterns in the image. The image can be multidimensional, with the shape being one dimension less than the image.

Abstract: A pattern evaluation method includes processing image data of at least one pattern serving as an object to be evaluated and detecting coordinates of edge points of the pattern in an image of the image data, making pairs of edge points from the edge points of the pattern, setting an arbitrary axis, calculating a distances between the edge points of each pair of the pairs of edge points and an angle between a straight line connecting the edge points of the pair and the axis, preparing a distance/angle distribution map which represents distribution of the distances and angles of the pairs of edge points, extracting a characteristic point of the distance/angle distribution map and analyzing the pattern on the basis of the extracted characteristic point.

Abstract: There is disclosed a method of retrieving moving pictures using joint point based moving information for allowing an user to retrieve pictures including a desired person object or the pose of the person object for still picture/moving picture data of multimedia data. The method of retrieving moving pictures using joint points based on pose information according to the present invention comprises the steps of extracting and expressing joint points for the joint point model against pictures where a major person exists from moving pictures or still pictures which are the object of retrival by means of joint points model based indexing apparatus, and retriving the person object according to the joint points extracted by said joint point model based indexing apparatus.

Abstract: A three-dimensional information processing apparatus for obtaining three-dimensional information from an object having a three-dimensional shape, and performing predetermined information processing, comprises: a camera for sensing images of the object from a plurality of coordinate positions using an image sensing system having one or a plurality of optical systems. A plurality of depth information are extracted from image sensing related information sensed by the camera at the plurality of coordinate positions, and the plurality of extracted depth information are converted and unified into depth information expressed by a unified coordinate system.

Abstract: A pattern recognition system which includes an input device for inputting a pattern and a character code, an extractor for extracting feature data from the input pattern, a detector for detecting size data of the input pattern, and a register for registering both the features data and the size data in correspondence with the input character code to a dictionary for character recognition.

Abstract: A method and apparatus is provided for segmenting a binary document image so as to assign image objects to one of three types: CHARACTER-type objects, STROKE-type objects, and LARGE-BITMAP-type objects. The method makes use of a contour tracing technique, statistical analysis of contour features, a thinning technique, and image morphology.

Abstract: A system for matching nodes of different spatial data sources. The system matches or ties a node from one data source to another data source. The present invention uses a highly accurate exclusion technique for excluding potential corresponding nodes of a second data source. Potential corresponding nodes are selected using node degree comparison, adjacency, fourier descriptor analysis of shapes formed by connecting paths between a node under investigation and a secondary node associated with that node under investigation, an analysis of total length to secondary nodes, an angle analysis to eliminate reflections, a secondary node degree analysis, and a surrounding node correlation coefficient comparison. Using one or more of these techniques in various combinations can result in the selection of a potential corresponding node with a high degree of accuracy.

Abstract: An apparatus and method for image registration of a template image with a target image with large deformation. The apparatus and method involve computing a large deformation transform based on landmark manifolds, image data or both. The apparatus and method are capable of registering images with a small number of landmark points. Registering the images is accomplished by applying the large deformation transform.

Abstract: An apparatus for recognizing visual patterns of handprinted characters or the like by means of an optical character reader which reads the patterns by a so-called outermost point method. While tracing the contour of the pattern stored in a two-dimensional memory, the distances from the starting point of the tracing and the integrated values of the coordinates of the points traced on the contour from the starting points are simultaneously obtained successively to extract outermost points for the series of contours. According to this outermost points, the contours are segmented into the convex line segments, concavity line segments and hole segments and the corresponding parameters of features of each segment are detected. Simultaneously, the convex line segment of which is shorter than a predetermined length is rejected and the remaining segments are subjected to matching operation in accordance with preliminarily prepared dictionary, thereby making a decision as to the pattern's identity.

Abstract: The invention provides boundary-based machine vision methods and systems for comparing a pattern to an image. These contemplate the steps of finding boundary points in the image and comparing those points with boundary points in the pattern to identify “matches,” i.e., boundary points in the image that match boundary points in the pattern, “extras,” i.e., boundary points in the image which do not match boundary points in the pattern, and “missing,” i.e., boundary points in the pattern that do not match boundary points in the image. Quantitative and/or qualitative information about matching, extra or missing points can be generated, including the differences between the locations, angles and contrasts of each boundary point in the image and the corresponding boundary point in the pattern.

Abstract: A handwritten pattern and stroke recognition system recognizes an input pattern by comparing it to reference patterns. The input and reference patterns comprise at least one stroke, where a stroke is defined as the series of sample points from pen-down to pen-up. The system of the present invention compares the input and reference patterns stroke by stroke and thus, compares input strokes to reference strokes. The system includes a tangent and stable point extractor, a reference database and a stable-point-based classifier. The extracter determines tangents to at least some of the sample points of the input stroke and determines, which of the sample points between a beginning and end of the input stroke are stable points. The reference database stores tangents of sample points of the reference strokes and stable points of the reference strokes.

Abstract: A parameter input section inputs, as parameters, the luminance difference between a to-be-recognized object P and its background, conveyance conditions, etc. An endpoint detecting section uses the parameters to detect a plurality of border points between the object P and the background, without being influenced by external noise or a stain on the object P. A length detecting section detects vertical and horizontal lengths of the object P, and an entire-object position determining section determines the position of the entire object P using the lengths detected by the section. A recognizing section compares an image input by an image input section, with reference image data stored in a reference image data memory, thereby recognizing the object P.

Abstract: Contour line data is easily extracted from line drawing data (binary image data). Run data representative of a line drawing compressed according to run-length coding is read line by line, and run data connected to a preceding line is inspected. The run data is divided into blocks in each of which present and preceding lines of run data are connected to each other in one-to-one correspondence, based on an inspected result. Connection information between the blocks of run data is added to the blocks, and the blocks of run data and the connection information are managed as run block data. Left-hand coordinates of the run block data are traced downwardly, and right-hand coordinates of the run block data are traced upwardly for thereby extracting coordinate data (contour line data).

Abstract: A vertex processing unit is provided for labeling each selected vertex dot as a vertical or horizontal attribute dot based on the direction of a side vector terminating at the vertex dot selected and that of a side vector originating from the vertex dot selected. A contour line segment processing unit is also provided for labeling each edge dot selected from a side as a vertical attribute dot if the y coordinate of the edge dot selected is different from that of a previous edge dot and that of a vertex dot located at the terminal point of the side, or otherwise, as a horizontal attribute dot. Contour color data are defined for the dots labeled as vertical or horizontal attribute dots. And inner color data are defined for the dots existing between an odd-numbered vertical attribute dot and an even-numbered vertical attribute dot, which is next to the former dot, on a selected scan line parallel to the axis of x coordinates, except for horizontal attribute dots.

Abstract: A method for deriving character features in a character recognition system for recognising characters, such as letters and digits, and a character recognition system in which the method is applied. After deriving a character description from a recorded image pattern of a character to be recognised, a unique point Pi of said character description is appended, by means of a distance transform, to each point Ri of a subset R={Ri/i=1, - - - ,N} of preselected reference points in the plane of said image pattern. Feature values of one or more features possessed by the character description in each appended unique point are then determined. If the character description is a contour description, the unique point Pi appended to each reference point Ri is that point where the reference point Ri has the shortest distance D(Ri) to the contour of the character, and the character features selected are the shortest distance D(Ri) and the direction H(Ri) of the contour in the appended point Pi.

Abstract: A method and apparatus for reconstructing legible characters from stored data are described. For each of a plurality of handwritten strokes, a set of endpoint conditions is received. The endpoint conditions are data which define an initial and a final tangent angle for the stroke, and positions of an initial and a final endpoint of the stroke. For each set of endpoint conditions, an artificial stroke is constructed such that it satisfies the corresponding endpoint conditions and consists of at most three segments. Each of these segments consists of a straight line or an arc. Within any given stroke, lines and arcs may occur only in alternation.

Abstract: A main image pickup section for picking up a main image and two or more subsidiary image pickup sections for picking up subsidiary images are used to shoot a subject. A 3-D object on which a predetermined pattern is drawn is also shot by the same image pickup sections, and parameters representing relative positions and attitudes of the image pickup sections are obtained using the images of the 3-D object. Feature points are extracted from each of the main image and the subsidiary images, and 3-D coordinates corresponding to each feature point in the main image are determined using the parameters and the feature points in the subsidiary images. Subsequently, judgment is executed and each of the feature points in the main image is accepted if the 3-D coordinates is within a designated 3-D space domain. Subsequently, subject outline points are extracted from the accepted feature points. Then, part of the main image which is surrounded by the subject outline points is extracted as the subject image.

Abstract: In a point symmetry shaping apparatus used for a curved figure, any unrecorded symmetry figures are automatically shaped in a point symmetry manner within a short processing time. In the point symmetry shaping apparatus, a coordinate point which constitutes a symmetry center candidate point is calculated from all of center points of symmetry-judging-line-segments. A first point symmetry center candidate point/center point distance calculating unit calculates a distance between the symmetry center candidate point and the center point of the symmetry-judging-line-segment. A point symmetry judging unit judges a point symmetry of the input curved figure in such a manner that a calculation is made of a fluctuation value contained in the distances between the symmetry center candidate point and the center points of the symmetry-judging-line-segments, and the calculated fluctuation value is compared with a threshold value.

Abstract: An arrangement of discrete objects is analyzed. First and second characteristics are extracted from an actual arrangement of discrete objects and an ideal arrangement of points, respectively, and the first and second characteristics are compared. Each characteristic is invariant, to within an angular constant, under an affine transformation of the respective arrangement.

Abstract: When an image file is to be registered, image data is read from a document; features of the document are extracted from the read-out image data; the image file is generated with the extracted document features correlated to the read-out image data for storage, and further in a case wherein an image file is to be retrieved, image data is read from a document, features of the document are extracted from the read-out data, and an image file having document features identical to the information for retrieval from image files with the extracted document features stored therein as information for retrieval.

Abstract: A figure location detecting system is designed to detect a location of a first figure (e.g., search fingerprint) within a second figure (e.g., file fingerprint). Particularly, the system is designed to correctly match characteristic points of the first figure with characteristic points of the second figure even if the first figure contains a plenty of similar characteristic points whose characteristics are similar with each other and/or even if the first figure is (partially) distorted. Herein, the system creates matching characteristic point pair candidates each corresponding to a pair of a first characteristic point, corresponding to one of the characteristic points of the first figure, and a second characteristic point corresponding to one of the characteristic points of the second figure.

Abstract: A system and method for determining offset errors between line and pixel coordinates of landmarks in a digitized image generated by an imaging system disposed on a spacecraft and line and pixel coordinates predicted by a mathematical model of the imaging system using landmark geodetic coordinates on the Earth. The system and method use landmarks in symbolic form, such as perimeters of lakes and islands that are stored in a database. A digitized image generated by the satellite-based imaging system is processed to extract a patch of the image containing a landmark. The image patch is then upsampled (magnified). The landmark boundary is processed using a mathematical model of the imaging system to generate absolute coordinates of the boundary pixels, which are upsampled and rasterized to produce a landmark mask.

Abstract: A computer-implemented process identifies an unknown language used to create a document. A set of training documents is defined in a variety of known languages and formed from a variety of text styles. Black and white electronic pixel images are formed of text material forming the training documents and the document in the unknown language. A plurality of line strokes are defined from the black pixels and point features are extracted from the strokes that are effective to characterize each of the languages. Point features from the unknown language are compared with point features from the known languages to identify one of the known languages that best represents the unknown language.

Abstract: An image and method for determining the location of known pattern of data in a captured object image wherein a feature analysis is performed upon a representative plurality of pixel locations to determine a starting point, feature analysis is performed upon the pixel locations surrounding the starting point and the starting point is moved to another pixel location if provides a higher feature analysis value than the starting point; the process of performing feature analyses and moving the starting point continues until the starting point is moved a maximum feature analysis value; once the maximum feature analysis value is found, the object image data may be rotated incrementally with respect to the reference pattern data and the feature analysis/starting point selection process repeated for each iteration until an object image location and object angle of rotation is determined.

Abstract: An apparatus for and a method of reliably detecting a very small dotted image area. A detecting area including a predetermined number of pixels including a target pixel is set. It is judged whether or not the detecting area belongs to a dotted image area on the basis of the number of edge pixels in the detecting area. Consequently, it is possible to detect edges peculiar to the dotted image area on the basis of a leading edge portion or a trailing edge portion of the density, to detect the very small dotted image area on the basis of the number of edge pixels.

Abstract: A method of 3-dimensional structure estimation of the invention, making use of a plurality of stereo-pictures, repeats, for each pixel of a first picture, a step of extracting corresponding small regions (4) corresponding to the concerning pixel according to a depth estimation (68); a step of calculating a neighboring correspondence value for each of the corresponding small regions representing correspondence among neighboring corresponding small regions of picures taken by neighboring cameras; a step of obtaining a sum (61) of the neighboring correspondence values; and selecting a value of the depth estimation (68) which gives a singular value of the sum (61) representing correspondence among the corresponding small regions.

Abstract: An image processing coefficient determination method of an image processing system comprising the steps of:outputting test pattern image data having one or more image areas each containing a specific spatial frequency component by an image output device for preparing an output image,measuring a spatial frequency characteristic of the output image, anddetermining spatial frequency processing coefficients from a relationship between a spatial frequency characteristic of the test pattern image data and the measurement result of the spatial frequency characteristic of the output image.

Abstract: An image detection method is disclosed wherein local edge information within an image is utilized for accurately detecting lines and curves of the images. The method can more accurately determine the location and orientation of a pattern and thus provides more reliable image match for pattern detection. A detector is trained off-line with example images resulting in a stored template generated by recording a test pattern similar to a pattern to be tested; anchor lines are identified within said template. Long lines are detected in subsequent test patterns using the disclosed method. The template is rotated and shifted before matching it to the test pattern so that anchor lines align with long lines detected within the test pattern. The template and test pattern are then compared to determine whether there is a match.

Abstract: A method and apparatus for controlling the operation of a video system such as a video game system through recognition of a visual pattern, or key. For example, the presence and identity of the key may enable and select certain features of the video system. The key may suitably consist of a locator mark and a selector mark, the presence of the locator mark identifying the presence of a key, and the selector mark identifying the particular key present. A key recognition system detects the presence of the locator mark and identifies the selector mark. The locator mark is preferably of a distinctive color, readily distinguishable from background objects. The selector mark may be chosen from a limited set of colors, and the color of the selector mark may then be identified to identify the key being used. In another aspect, the locator mark is located in a particular position within the selector mark and is then used by the key detector to scale a rectangle about the selector mark.