Abstract: A method for representing images for pattern classification extends the conventional Isomap method with Fisher Linear Discriminant (FLD) or Kernel Fisher Linear Discriminant (KFLD) for classification. The extended Isomap method estimates the geodesic distance of data points corresponding to images for pattern classification, and uses pairwise geodesic distances as feature vectors. The method applies FLD to the feature vectors to find an optimal projection direction to maximize the distances between cluster centers of the feature vectors. The method may apply KFLD to the feature vectors instead of FLD.
Abstract: A method for recognizing a pattern of an alignment mark on a wafer includes positioning the wafer on an adjustable wafer stage in an alignment apparatus; capturing images of a key alignment mark by magnifying an alignment mark region of the wafer; deleting image data from a region where the alignment pattern does not exist between the captured images; and extracting an alignment mark pattern by a pattern recognition of the remaining image data after the deletion of the image data. Thus, an alignment failure can be reduced because a particle on the wafer is not mistaken as an alignment mark.
Abstract: The contrast of an image is enhanced by converting pixels of detail images of a multi-scale digital representation of the image according to a conversion function. The conversion function is monotonically increasing and odd and provides maximum amplification within an intermediate subrange of pixel values of a detail image which subrange is considered critical. Furthermore the slope of the conversion function has a local maximum and a local minimum.
Abstract: A “composite signal generator” automatically combines two or more copies of a signal to produce a composite that is better than the individual copies. For example, given two or more input images of a scene captured from approximately the same viewpoint, the composite signal generator automatically produces a composite image having reduced or eliminated areas of occlusion with respect to any occlusions existing in the input images. First, the input images are registered using conventional image registration techniques. Differences between the registered images are then used to identify regions of potential occlusion in one or more of the images. A determination of which image is actually occluded is made by identifying which image has a larger discontinuity along a border of the potentially occluded regions. A composite image is then created by choosing one image and mosaicing parts from the other images where it is occluded and they are not.
Abstract: Techniques and tools for analyzing and adjusting the exposure of digital images are described. For example, a computer processes a digital image by analyzing exposure data, assigning an image classification (e.g., StretchNeeded, UnderExposed, OverExposed, or Normal) based on the analysis, and selecting an exposure compensation technique (e.g., histogram stretch, positive gamma curve, negative gamma curve, or no adjustment) based on the image classification. The exposure data can be luminance values for pixels in the digital image represented in a histogram. The computer can produce transform data comprising a transformation of the exposure data according to the selected exposure compensation technique. The computer can store transform data in a look-up table and can store the look-up table in the digital image file. The described techniques and tools can be implemented as a feature of an operating system environment and can be activated responsive to user action via a user interface.
March 26, 2003
Date of Patent:
April 15, 2008
Min Liu, Po Yuan, Richard S. Turner, Jr.
Abstract: The present invention is operative to determine the angular pose of an assumed object. A series of projection sums are calculated from a sample image of the known object where the known object has an assumed pose. The projection sums are calculated for a range of angles and organized into a two dimensional array coined a projection image. An image of the known object is provided where the known object has an unknown pose. A projection sum is calculated across the image of the known object with the unknown pose preferably at the perpendicular. This projection sum is correlated against the projection image and the angle corresponding to the best correlation is determined to be the angular pose of the known object.
Abstract: A portable data carrier comprises a communication means and a memory with a biometric template which is intended to be compared with a biometric sample for identity check. The biometric template is divided into a private part which is adapted to be exclusively used in the portable data carrier and a public part which is adapted to be transferred, with the aid of the communication means, to an external processing unit and be used therein. Moreover, a processing unit for use in biometric identity check is described, comprising a processor and a communication means. The communication means is adapted to receive a biometric sample and a public part of a biometric template, the processor is adapted to compare the received public part of the template with the biometric sample, and the communication means is further adapted to transfer, when a comparison criterion has been satisfied, at least part of the biometric sample to the portable data carrier for further comparison on the data carrier.
Abstract: The present invention pertains generally to the field of computer graphics user interfaces. More specifically, the present invention discloses a video image based tracking system that allows a computer to robustly locate and track an object in three dimensions within the viewing area of two or more cameras. The preferred embodiment of the disclosed invention tracks a person's appendages in 3D allowing touch free control of interactive devices but the method and apparatus can be used to perform a wide variety of video tracking tasks. The method uses at least two cameras that view the volume of space within which the object is being located and tracked. It operates by maintaining a large number of hypotheses about the actual 3D object location.
July 12, 2003
Date of Patent:
January 8, 2008
Nils Krahnstoever, Emilio Schapira, Rajeev Sharma
Abstract: A recognition device which judges whether a target is identical with a predetermined reference, comprising: a holding unit which holds the target; multiple deformation units which deform the target held by the holding unit with at least one degree of flexibility in deformation; multiple deformed amount estimation units which correspond to the deformation units in a one-to-one relationship and estimate a deformed amount of the target from the reference with respect to the flexibility in deformation according to the corresponding deformation unit; an estimated error evaluation unit which evaluates an estimated error of the deformed amount estimated by the deformed amount estimation unit; an adjustment unit which operates any of the deformation units with precedence according to the estimated error evaluated by the estimated error evaluation unit; a similarity calculation unit which calculates a similarity between the reference and the target which is deformed by the deformation unit operated with precedence by
Abstract: The present invention is a method, system, and apparatus for visually displaying digitized information in a way that allows a human operator to detect easily patterns and characteristics within the data. A personal computer is used to extract raw data and to perform Boolean operations at the behest of a user. The manipulated information is then converted into a false-color image and displayed on an output screen. The images can be compared to known data, called a template, in order to detect one or more desired patterns that indicate desirable features. The desired pattern can then be highlighted (or illuminated) and be displayed to the operator so that the location of the desired feature may be identified.
Abstract: An automated symbolic recognition system and method includes pre-stored symbolic models to support a three phase symbolic reshaping scheme during the symbol recognition process that includes: (a) deriving dissimilarity level from alphanumeric ID's net topological variation and the integration of each of its arcpoly structural variation(s) signifying a reasonably accurate confidence level for the goodness of recognition, thus establishing a mechanism that derives dissimilarity level (or cost value) between image and database features including shape, size and topological relationship, (b) determining the reshaping or transformation of an arcpoly to another arcpoly by introducing variations to the original arcpoly to alter its shape and orientation and deriving at each step, the new cost value as a function of variation(s) present and imposed, and (c) determining the equivalent representation of an arcpoly by a succession of smaller and adjoining arcpoly(s) in order, or vice versa.
Abstract: A plurality of binary images are obtained from one form image by binarizing the form image by using a plurality of binarizing threshold values for improving a recognition rate of the search form irrespective of a state of the inputted form image, and format data are generated respectively from the plurality of binary images. If the inputted form image is a binary image, after the binary image has been converted into a multi-valued image by a vignetting filter, a plurality of binary images are obtained from one form image by executing the binarizing processing by use of the plurality of binarizing threshold values respectively, thereby generating the format data respectively from the plurality of binary images.
Abstract: A method for controlling the display of a screen which allows the user to discriminate the scanned image and the recognition result intuitively and easily. The display control method for allowing the user to verify the recognition result of a character on a verification screen form is implemented as follows. First of all, the standard pattern of a specific character stored in a memory and a plurality of character images recognized as the specific character are read, upon an operation of the user. And the read standard pattern is displayed in a recognition result character display portion within the verification screen form and a plurality of the read character image are displayed sequentially in a character image display portion adjacent or proximal to the recognition result character display portion and at a predetermined position of a character image list display portion.
May 8, 2003
Date of Patent:
August 21, 2007
International Business Machines Corporation
Abstract: A device for carrying out of biometric identification by comparing face shapes. The shape of a first face is compared with the shape of a second face. A map can be determined between the shapes. The shape, and hence the comparison, is substantially independent of pose and lighting, and also parts of the shape can be removed from the comparison in order to enhance the recognition.
May 5, 2003
Date of Patent:
July 10, 2007
Fish & Richardson P.C.
Roman Waupotitsch, Mikhail Tsoupko-Sitnikov, Gerard Medioni
Abstract: An automated symbolic recognition system and method includes multiple representations of each alphanumeric in the database by identifying a finite number of exemplars per alphanumeric symbol.
Abstract: An automated symbolic recognition system and method includes the basis for the system's feature set by identifying (a) a set of logical symbols comprising a finite class of arcpolys (lines and arcs and a point) that to the exclusion of the point, each member class has a unique (distinct) orientation, and (b) a set of subclass symbols per logical class of symbol representing a finite subclass of arcpolys (lines and arcs and a point) that to the exclusion of the point, each subclass member has a unique (distinct) extreme points size and/or depth size.
Abstract: A surface inspection apparatus in accordance with the principles of the invention includes an optical system having a plurality of time delay integration (TDI) sensors. The plurality of TDI sensors are arranged to generate a plurality of images of an object so that the images are offset a sub-pixel distance from each other. A scanning element enables the TDI sensors to scan the object so successive images of the object can be generated. Image processing circuitry is used to process the plurality of successive images together to produce a reconstructed image of the object having increased pixel density. The embodiments of the invention also include methods for generating reconstructed images from a plurality of TDI images obtained from at least two offset TDI sensors.
Abstract: An automated symbolic recognition system and method includes a hierarchical hypothesis-and-verification technique during various stages of the handwriting recognition process, whereby a series of initial assessments are made based on the information availed upon them, and later during processing, they are validated or rejected depending on the degree in which preset milestones were satisfied and are followed by a sequence of alternative hypotheses in the event of failure of the latest hypothesis until they are satisfied.
Abstract: A character input device for inputting a character by extracting and recognizing a character from an image of a document photographed by a photographing device includes the photographing device. The device stores a recognition result of the extracted and recognized character and reliability indicative of sureness of recognition. Character recognition results of images obtained by repeatedly photographing a whole or part of the document are compared in order to input the character based on a character recognition result whose reliability is highest. A character input method includes adding reliability indicative of sureness of recognition to a recognition result of the character, comparing character recognition results of images obtained by repeatedly photographing the whole or part of the document, and selecting a character recognition result whose reliability is highest.