Abstract: A method to arbitrate between results obtained from observationally derived data by several procedures, where the results are items characteristic of the observationally derived data is disclosed. Each procedure is given a ranking according to the confidence on the procedure. The results and characteristics derived from the results are used in a plurality of rules, where the rules are used to arbitrate between the results.

Abstract: A medical image processing apparatus can change an output image in accordance with photographing condition. The medical image processing apparatus for processing image data representing an image obtained by radiography and read by an image reading apparatus, comprises: a unit for receiving the image data and photographing condition when the image is obtained; a unit for executing image processing of the image data; and a unit for reading output-format control information corresponding to the photographing condition from a predetermined set of output-format control information to be used to control an image format when the image is output on a film.

Abstract: A method and apparatus for performing scalar quantization with a power of two step size is described. In one embodiment, the method comprises receiving a sequence of image data to compress and specifying scalar quantization with a power of two step size using three bit values to apply to the sequence of image data.

Abstract: A multimodal biometric method is provided for authenticating at least one user, in which at least one user is assigned to one of at least two classes, in which similarity measures are used that are measures of the similarity of biometric data, determined for the user, to reference data for one of the classes, in which a overall statistical statement on the user's membership of the class is present for each of the at least two classes as a function of the similarity measure, in which the overall statistical statement for each of the at least two classes results from the combination of the statistical statements of at least two individual biometries for the respective class, the statistical statements of the individual biometries being statistical statements on the user's membership of the class as a function of the similarity measures, and in which the user is assigned to one of at least two classes when a specific assignment criterion is fulfilled.

Abstract: A Chinese text entry system and method is provided to allow users to enter a character to a device such as a cellular phone or a PDA by adding a first few strokes required for the character using a joystick or its equivalent. By simply moving the joystick to add one or more strokes which are used to start writing a character, or in some case even before any stroke is added, a user can find a desired character from a displayed selection list. The selection list is context sensitive, varying depending on the last character entered, so that the user can be provided with the most possible candidates of the desired character.

Abstract: A method is described of bit-mapped image analysis comprising division of all analysis means at one's disposal into several groups differing in accuracy and further processing multi-stage analysis. The analysis comprises a primary analysis stage and at least one profound analysis stage, with supplemental data collected at both stages. The primary analysis, includes preliminary recognition of objects with distortion and detection of objects that require more precise analysis means to overcome the distortion. At the primary analysis stage, the analysis means from the group of the most inaccurate group are used. The profound analysis stage includes repeating recognition of objects with a distortion taking into account the supplemental data obtained at the previous stage, detecting objects that require more precise analysis means to overcome the distortion, and collection of newly appeared supplemental data. Each subsequent profound analysis stage uses analysis means from the group of more accurate means.

Abstract: Sprite identification in animated sequences is achieved by performing three separate but related procedures, namely (i) identification of the sprite, (ii) identification of the background and (iii) identification of the identified sprite's translation path. By analyzing a sequence of frames, sprite definition is first established. Then, using the definition of the sprite, a determination is made of the background image and translation path taken. This analysis allows the animated sequence to be compressed in a format in which the background and the sprites are separately identified.

Abstract: An image processing method for facilitating data transmission is provided. An image compression method is performed to convert X-bits binary digital signals to a binary compressed data in a floating-point form of (1.n)*2m. Bit m represents the first bit with logic level “1” of the X-bits binary digital signals, and n represents the bits taken from the X-bits binary digital signals after the bit m. The binary compressed data in the floating-point form of (1.n)*2m is outputted with a sequence of binary number representing a set of (m, n). The latter n bits of the sequence of binary numbers are consisted of the n bits of the X-bits digital signal. Therefore, by the present image compression method, the transmission amount of image data is reduced. The transmission time of image data and the volume of a memory for storing the image data are also reduced.

Abstract: A spatial domain using image produced by a source is combined with watermark data Ri to produce a spatial domain watermarked image. The watermarked image is produced by an embedder according to equation: Ci?=Ci+?. Ri, where Ci and Ci? are wavelet transform coefficients of the image, and ? is a scaling factor. ? is chosen so that the watermark is imperceptible in the image and to resist removal of the watermark by unauthorized processing. It is desirable that a has the smallest value which achieves that. If ? is too big the watermark is perceptible in the image. If ? is too small the mark may not survive processing of the image.

Abstract: An unfair contents appropriation detection system capable of efficiently detecting the contents intentionally changed for the worse with ill will. If a contents browser terminal 130 obtains the contents appropriated by a contents appropriator terminal 120, an unfair contents appropriation detection program is incorporated into the above contents appropriated by the contents appropriator terminal. This program transmits alarm information AL to a watcher terminal 140 when a content browser terminal obtains the unfairly appropriated contents from the unfair contents appropriator terminal 120. Furthermore, in this unfairly appropriated contents, there is embedded a digital water mark for obtaining the changed area information of the contents. The watcher terminal obtains the contents from the contents appropriator terminal based on the alarm information AL and executes verification of the digital water mark with regard to the contents as obtained.

Abstract: The invention relates to an automatic fingerprint identification method. In the method according to the invention, identification is executed in three phases. In coarse alignment (32), fingerprints are compared using a few reference points. In fine alignment (33), coincidence of fingerprint minutiae are used to find the best fingerprint matching transformation. In the matching phase (34), individual minutiae of the fingerprint examined are matched against a known template. Parameters are computed from the matching minutiae and used in making the final identification.

Abstract: The invention described herein provides a method and apparatus for document processing that efficiently separates and interrelates single modalities, such as text, handwriting, and images. In particular, the present invention starts with the recognition of text characters and words for the efficient separation of text paragraphs from images by maintaining their relationships for a possible reconstruction of the original page. The text separation and extraction is based on a hierarchical framing process. The process starts with the framing of a single character, after its recognition, continues with the recognition and framing of a word, and ends with the framing of all text lines. The method and apparatus described herein can process different types of documents, such as typed, handwritten, skewed, mixed, but not half-tone ones.

Abstract: A method beginning at (10) is provided for creating high-fidelity visible coloring from infrared images of a scene under surveillance. The infrared images captured at (12) are analyzed at (14) to determine if an object, such as a face, is identifiable within the image. If an object is identifiable at (16) the object features are compared to a plurality of stored object features at (20). If there is a match at (22), the color characteristics of the object are obtained at (24) and the object is colored at (26) based on the stored database feature information. If there is no match at (22) or identifiable object at (16) and object color cannot be identified at (23), the image is analyzed at (28) to determine if a pattern, such as clothing, is identifiable within the image. If a pattern is identifiable at (30), the color characteristics of the pattern are obtained at (34) and the pattern is colored at (36) according to infrared reflectance characterization in conjunction with the stored pattern information.

Abstract: An image processing apparatus detecting a wheelchair passenger for an elevator. This image processing apparatus includes a TV camera installed so that a hail where passengers are waiting is monitored through the camera, which is located in the vicinity of the elevator door, a face area extraction unit, a wheelchair area estimation unit, and a wheelchair existence judgment unit. The image processing apparatus is operated in the following manner, one or a several passengers' face area is identified by the face area extraction unit in the image picked-up by the TV camera, and the position of the face area in the image is detected. If the passenger is sitting in a wheelchair, the area where the wheelchair is likely to be in the image is estimated by the wheelchair area estimation unit from the position of the face area. The estimated area where the wheelchair is likely to be is checked by the wheelchair existence judgment unit, to determine whether the wheelchair is present.

Abstract: In a computer processing apparatus 2, a 3D computer model comprising a polygon mesh 500 representing the visual hull of an object 300 is generated by processing images of the object recorded at different positions and orientations to back-project the silhouette of the object in each image to give a respective cone which constrains the volume of 3D space occupied by the object. To remove concave and convex artefacts 510 in the polygon mesh 500, the polygon mesh is projected into each image to give a respective reference silhouette for each image. A change is made to at least one edge or vertex in the polygon mesh to give a refined polygon mesh, which is then projected into each image. The resulting silhouette in each image is tested against the corresponding reference silhouette.

Abstract: While both a forehead and a cheek are made in close contact to an eye position fixing portion, a space defined between a left eye and a pinhole is shielded from extraneous light. When light derived from a room lighting device is entered into a right eye, a diameter of a pupil of the left eye is defined in a self-definition manner due to an interlocking characteristic of a living body, and then, a width of an iris 3 may be secured under stable condition. While infrared light is illuminated from a button lighting device onto this left eye, this illuminated left eye is photographed by an image pick up unit, and then, feature information as to the photographed left eye is compared with registered feature information in an individual identifying unit so as to execute an individual identification operation.

Abstract: The invention concerns an image information apparatus that stores a plurality of image information corresponding to a plurality of images. The apparatus includes a receiving section to receive the plurality of image information, an ID information reading section to read the ID information corresponding to the image information, a memory section to store the plurality of image information and the ID information read by the ID information reading section relatively, and an ID distinction section that outputs an ID distinction-signal when detecting that the second ID information does not coincide with the first ID information. The memory section stores a plurality of the image information corresponding to the first ID information, before the second ID information is received, when the memory section receives the ID distinction-signal outputted by the ID distinction section.

Abstract: An image decoding method is utilized to obtain high quality images without error accumulation. Such an image decoding method comprises receiving an encoded bitstream including information of P and B frames, and executing motion compensation by synthesizing a predicted image of a current frame using motion vector information included in the encoded bitstream and a reference image which is a previously decoded image. The motion compensation includes calculating intensity values at points where no pixels actually exist in the reference image by interpolation performed according to information specifying a positive rounding method or a negative rounding method when the current frame is a P frame, and using a predetermined rounding method which is a positive rounding method or a negative rounding method when the current frame is a B frame.

Abstract: The color of a color image having at least two sets of image pixels having color and luminance values is corrected by generating, for each color value, a low spatial frequency monochrome image including a set of smoothed image pixels. A color correction function is applied to selected sub-sets of the image pixels to generate corresponding sub-sets of corrected, smoothed image pixels. The correction includes a contribution from smoothed image pixels in one or more of the other sets of smoothed image pixels. The sub-sets of corrected smoothed image pixels is completed by interpolation or extrapolation. Each set of corrected smoothed image pixels is used to generate a corrected color image by generating one or more high spatial frequency luminance images that are combined with each of the corrected low spatial frequency monochrome images to form a full color image.

Abstract: A method of defining deviations from a desired position of pixel sites in an image recording matrix by defining the 3-D data of a planar or slightly curved surface. To the extent necessary the results are smoothed, and the measuring points, utilizing the 3-D data, are projected back the sensors of the image recording matrix for determining the difference between the two points projected onto the sensors and associated with a 3-D measuring point. After selectively shifting the surface relative to the 3-D measuring system, prior method steps are repeated until the desired accuracy in the definition of the deviation has been has been attained.