Abstract: Methods for automatically providing descriptors for images to a user include providing an image descriptor database having a plurality of image feature vectors, each of the plurality of image feature vectors having an associated descriptor. A specificity value is assigned to each of the descriptors such that the specificity value comprises an estimation of a degree of description specificity. A first image feature vector is determined for a first image, and the first image feature vector is compared with the plurality of image vectors in the image descriptor database. One or more descriptors for the first image vector is identified based on the comparison of the first image feature vector with the plurality of image vectors and the specificity value of the corresponding descriptor.

Abstract: A handwriting processing apparatus and method effective for search of, e.g., a document file including handwriting is provided. When a handwriting characters are input to a coordinate input unit as a search key, a corresponding character in a dictionary is recognized for each of the handwritten characters, a search unit searches for a text code stored in a document file storage unit based on a text code of the corresponding character if the corresponding character is recognized and the search unit searches for handwriting trail data stored in a handwriting trail storage unit based on a handwriting trail of the handwriting character if the corresponding character is not recognized, thereby finding a desired document file.

Abstract: An image evaluation apparatus includes a printing status judgment unit for judging a printing status of a target image assigned with an evaluation value and an evaluation value changing unit for appropriately changing the evaluation value based on information representing the printing status. An image evaluation method and a program for causing a computer to execute the image evaluation method are also provided. Therefore, the evaluation value can be assigned to the image, reflecting an intention and preference of a user at the time of selection of an image for printing.

Abstract: A method of hierarchical image indexing which extracts representative images of a TV program and generates a hierarchical index image guide of the representative images, and an apparatus therefor. Respective images at predetermined times are extracted from a predetermined program. A hierarchical index is allocated to each extracted image and the indexes are mapped to locations of transport stream packets. A tree of images having the allocated indexes is constructed as a hierarchically indexed Image guide (HIIG). When a stored program is to be reproduced, a location desired to be watched is selected in the HIIG, and from the selected location, reproducing and trick plays (FF, rewind) are enabled such that a user easily searches the program and begins viewing the program at the searched for location.

Abstract: The present invention comprises using error propagation for building feature spaces with variable uncertainty and using variable-bandwidth mean shift for the analysis of such spaces, to provide peak detection and space partitioning. The invention applies these techniques to construct and analyze Hough spaces for line and geometrical shape detection, as well as to detect objects that are represented by peaks in the Hough space. This invention can be further used for background modeling by taking into account the uncertainty of the transformed image color and uncertainty of the motion flow. Furthermore, the invention can be used to segment video data in invariant spaces, by propagating the uncertainty from the original space and using the variable-bandwidth mean shift to detect peaks. The invention can be used in a variety of applications such as medical, surveillance, monitoring, automotive, augmented reality, and inspection.

Abstract: In a method for the processing of radiological images for the detection of opacities in a radiography image, an algorithm is proposed for detecting opacities according to their contrasts. To determine a contrast that is not over-estimated in each element, an algorithm is proposed for computing the background intensity by means of a sliding-window filtering. This type of sliding-window filtering computes the set of mean values of grey levels about an immediate environment of a region of elements whose size depends on the size of the opacity to be detected. The algorithm considers a background intensity of the region of elements to be the lowest mean value grey levels among the set of mean values of grey levels.

Abstract: Disclosed is a method for assigning the content of a digital image to a class of a classification system. Said method comprises the following steps: —a predetermined number of F numerical shape characteristics ?m are determined; —the value of each shape characteristic of the F numerical shape characteristics determined for the image is compared to the value filed in a table for the respective shape characteristic, values for the individual numerical shape characteristics being allocated to each class in the table; —the class in which the F numerical shape characteristics determined for said image correspond best to the values of the numerical shape characteristics indicated in the table for said class is output as the class into which the image that is to be recognized is classified.

Abstract: An image processing apparatus includes: an image acquisition section which acquires an original image; a resolution conversion section which converts the resolution of the original image acquired by the image acquisition section and generates a plurality of reduced images having different resolutions; a detection section which processes by template matching the plurality of reduced images generated by the resolution conversion section and detects an area occupied by a picked-up image of a particular object corresponding to the template, from the reduced images; and a detection result processing section which detects the area occupied by the picked-up image of the particular object on the original image, by processing a detection result obtained by the detection section, the detection section detecting the area occupied by the picked-up image of the particular object by processing the plurality of reduced images in an order in which resolution sequentially varies on a step-by-step basis.

Abstract: A method of classifying regions in a digital image or video captured by an image capture device includes providing a geographic location determining device associated with the image capture device; using the location determining device to measure the image capture location at substantially the time that the digital image or video was captured; and classifying regions in the captured digital image or video into one or more classes based on the image-capture location.

Abstract: An information processing apparatus includes a higher node having a plurality of weak discriminators that have learned a learning sample of a first label and a learning sample of a second label that has a predetermined coordinate relationship with the learning sample of the first label, a first lower node having a plurality of weak discriminators that have learned the learning sample of the first label based on the discrimination result of the higher node and a second lower node that has no weak discriminator.

Abstract: A method and system for efficiently detecting faces within a digital image. One example method includes identifying a digital image comprised of a plurality of sub-windows and performing a first scan of the digital image using a coarse detection level to eliminate the sub-windows that have a low likelihood of representing a face. The subset of the sub-windows that were not eliminated during the first scan are then scanned a second time using a fine detection level having a higher accuracy level than the coarse detection level used during the first scan to identify sub-windows having a high likelihood of representing a face.

Abstract: An apparatus, method, and medium for detecting a face in an image using classifiers generated by a boost algorithm are provided. The apparatus may include an image dividing unit dividing a black and white image into sub-windows; a sub-window detection unit detecting a sub-window corresponding to a face image in the divided sub-windows using the front part of the cascade of the classifiers and storing the sub-window detected as the face image; and a post-processing unit verifying whether each of sub-windows included in a group among a plurality of groups into which stored sub-windows are grouped is the face image using the number of child detections in each group and the latter part of the cascade of classifiers, integrating the sub-windows in the group according to the result of the verification, and outputting the result of the integration as the face image included in the black and white image.

Abstract: A method and apparatus for detecting text in real-world images comprises calculating a cascade of classifiers, the cascade comprising a plurality of stages, each stage including one or more weak classifiers, the plurality of stages organized to start out with classifiers that are most useful for ruling out non-text regions, and removing regions classified as non-text regions from the cascade prior to completion of the cascade, to further speed up processing.

Abstract: A system for automatically classifying an object of a target image is provided. A classification system provides a collection of classified images along with a classification of the dominant object of the image. The classification system attempts to classify the object of a target image based on similarity of the target image to the classified images. To classify a target image, the classification system identifies the classified images of the collection that are most similar to the target image based on similarity between salient points of the target image and the classified images. The classification system selects a classification associated with the classified images that are most similar to the target image as a classification for the object of the target image.

Abstract: An analysis system analyzes and measures patterns present in the pixel values of digital images using a computer-implemented network structure that includes a process hierarchy, a class network and a data network. The data network includes image layers, thematic layers and object networks. Various types of processing are performed depending on whether the data of the digital images is represented in the image, thematic or object layers. Pixel-oriented and object-oriented processing is combined so that fewer computations and less memory are used to analyze the digital images. Pixel-oriented processes, such as filtering, are selectively performed only at pixel locations that are assigned to a specified thematic class of a thematic layer or that are linked to a particular object of the object network. Similarly, object-oriented processing is performed at pixel locations linked to objects generated using thematic layers or using image layers on which pixel-oriented processes have already been performed.

Abstract: A plurality of images are classified into hierarchical categories. A degree of importance for each of the categories is set according to at least one of: the number of images therein, the number of images in an upper-level category, the number of categories related to the category, the number of hierarchical levels below the category, and the number of hierarchical levels from the uppermost level to the level of the category.

Abstract: Provided are a method, system, and computer readable medium for training and using classification components on multiple processing units. A plurality of processing units each has a memory including one of a plurality of subsets of a set of data points. At least two of the processing units have different subsets of data points. A plurality of classification components are executed by the processing units. Classification components executing at the processing units are trained, wherein each classification component is trained with the subset of data points in the memory of the processing unit that is executing the classification component. One of the classification components is transferred to an additional processing unit of the processing units to train the transferred classification component using the subset of data points in the memory at the additional processing unit in response to training the classification component with the subset of data points.

Abstract: The distortion of each node is calculated from the distortions of respective elements in the finite element analysis result of a global model of a structure, and a second-order coefficient of a quadratic function representing the displacement at each node of a micro model is calculated from distortions of respective nodes. In addition, a constant term and a first-order coefficient of the quadratic function are calculated from the displacements of the respective nodes of the global model. Then, the displacement at each boundary node of the micro model is calculated using the obtained quadratic function and a finite element analysis of the micro model is performed.

Abstract: Color conversion apparatus and methods are provided for converting a first color value, such as an RGB, L*a*b* or XYZ color value, to a second color value, such as an RGB, CMY, CMKY, L*a*b*, L,C1C2, or XYZ color value, without saturation error.

Abstract: With little efforts, to estimate the appropriateness of automatic defect classification, and to make classification criteria settable with a guarantee for the better classification performance, defects unknown with their classification classes are classified based on two different classification criteria. Also, defects differed in classification results are collected, and each thereto, a defect classification class is provided by using a manual. Then, the defects provided with the classification classes are divided into two types of groups: one is a setting group for the classification criteria; and the other is an evaluation group. Based on the classification criteria that is so set as to classify the defects included in the setting group with the maximum performance, the classification performance of a case where classification is applied to defects included in the evaluation group is calculated, and the appropriateness of thus set classification criteria is evaluated.

Abstract: Systems and methods are presented that determine components to use as examples to train a component-based face recognition system. In one embodiment, an initial component shape and size is determined, a training set is built, a component recognition classifier is trained, and the accuracy of the classifier is estimated. The component is then temporarily grown in each of four directions (up, down, left, and right) and the effect on the classifier's accuracy is determined. The component is then grown in the direction that maximizes the classifier's accuracy. The process can be performed multiple times in order to maximize the classifier's accuracy.

Abstract: A method and apparatus of tile-based belief propagation are disclosed. An image is split into a number of tiles. Messages are iteratively generated within each of the tiles based on the messages from neighboring pixels to the tile at a previous iteration, wherein each message represents information of a state of the pixel. The generated messages for sending out of the tiles are stored. Labels are then determined based on the stored messages, wherein each label represents the state of the pixel.

Abstract: Systems and methods that exploit unique properties of a language script (e.g., condition joining rules for Arabic language) to enable a two tier text recognition. In such two tier system, one tier can recognize predetermined groups of linked letters that are connected based on joining rules of a language associated with the text, and another tier dissects (and recognizes) such linked letters to respective constituent letters that form the predetermined group of linked letters. Various classifiers and artificial intelligence components can further facilitate text recognition at each level.

Abstract: Provided is an apparatus and method for super-resolution enhancement processing. The apparatus includes a low-resolution image capturing unit, a patch pair generating unit, a patch pair database, and a high-resolution image reconstructing unit. The low-resolution image capturing unit captures a low-resolution image. The patch pair generating unit divides each of a plurality of high-resolution images into patches and generates a plurality of patch pairs composed of medium-frequency patches and high-frequency patches for the patches. The patch pair database stores the plurality of patch pairs generated by the patch pair generating unit.

Abstract: An image indexer for indexing a plurality of images that includes a first data structure for subsequent classification of the one or more images. The first data structure includes characteristics for at least one class. An image classifier classifies one or more individual images found in the plurality of images as classified images according to the first data structure. A second data structure performs subsequent clustering of the plurality of images, wherein the second data structure includes at least two sequential events in a set of known events. The classified images are clustered according to the second data structure, and a representative image is selected from each cluster of classified images.

Abstract: A system (400) for generating training images for a pattern recognition classifier (54) from three-dimensional image data representing an output class comprises a model constructor (404) that constructs an image model in a standard coordinate domain from the three-dimensional image data. A model transformer (406) manipulates the image model in the standard coordinate domain to produce at least one manipulated image model. An image restorer (408) maps the at least one manipulated image model from the standard coordinate domain to a coordinate domain associated with an image source utilized during the run-time operation of the pattern recognition classifier.

Abstract: A method of identifying at least one particular person of interest in a collection of digital images taken over time, includes providing the collection of digital images with each digital image containing one or more persons; storing in a digital database an appearance model having a set of features associated with the particular person of interest and a prior probability of the person of interest appearing in images by a particular photographer; and comparing the appearance model with features extracted from the digital images in the collection of digital images and using the prior probability and the appearance model to determine if the particular person of interest is in one or more digital images in the collection of digital images.

Abstract: A method and system for learning-based assessment of the quality of an image is provided. An image quality assessment system trains an image classifier based on a training set of sample images that have quality ratings. To train the classifier, the assessment system generates a feature vector for each sample image representing various attributes of the image. The assessment system may train the classifier using an adaptive boosting technique to calculate a quality score for an image. Once the classifier is trained, the assessment system may calculate the quality of an image by generating a feature vector for that image and applying the trained classifier to the feature vector to calculate the quality score for the image.

Abstract: An object recognition apparatus includes an image input unit capturing image data, an object dictionary unit storing conditions to specify a type of an object, and a processor collating the image data with the conditions and specifying the type of the object imaged in the image data, in which the object dictionary unit classifies the conditions into hierarchies and stores the classified conditions, and the processor performs collation while narrowing down object conditions positioned in lower hierarchies based on a collation result of object conditions positioned in upper hierarchies.

Abstract: A method and system for generating a classifier to classify sub-objects of an object based on a relationship between sub-objects is provided. The classification system provides training sub-objects along with the actual classification of each training sub-object. The classification system may iteratively train sub-classifiers based on feature vectors representing the features of each sub-object, the actual classification of the sub-object, and a weight associated with the sub-object. After a sub-classifier is trained, the classification system classifies the training sub-objects using the trained sub-classifier. The classification system then adjusts the classifications based on relationships between training sub-objects. The classification system assigns a weight for the sub-classifier and weight for each sub-object based on the accuracy of the adjusted classifications.

Abstract: A method for classifying features in a digital medical image includes providing a plurality of feature points in an N-dimensional space, wherein each feature point is a member of one of two sets, determining a classifying plane that separates feature points in a first of the two sets from feature points in a second of the two sets, transforming the classifying plane wherein a normal vector to said transformed classifying plane has positive coefficients and a feature domain for one or more feature points of one set is a unit hypercube in a transformed space having n axes, obtaining an upper bound along each of the n-axes of the unit hypercube, inversely transforming said upper bound to obtain a new rule containing one or more feature points of said one set, and removing the feature points contained by said new rule from said one set.

Abstract: A digital image processing method for detecting faces in a digital color image comprises the steps of: providing a distributed face detection system having complementary classifiers, wherein the classifiers are complementary in a frequency domain; selecting classifier parameters for the complementary classifiers from a plurality of different parameter generating sources, at least one of which is controllable by human input; reconfiguring the complementary classifiers in the distributed face detection system according to the selected classifier parameters; and detecting faces using the distributed face detection system.

Abstract: A vehicle restraint system has a vision-based occupant classification system for control of airbag deployment during a crash scenario. The classification system utilizes two imaging sensors which together create a stream of paired images received and stored by an occupant classification controller. A computer program product of the controller utilizes the paired images to extract disparity/range features and stereo-vision differential edge density features. Moreover, the controller extracts wavelet features from one of the two paired images. All three features or maps are classified amongst preferably seven classifications by algorithms of the computer program product producing class confidence data fed to a sensor fusion engine of the controller for processing and output of an airbag control signal input into a restraint controller of the vehicle restraint system.

Abstract: A speech recognition method includes several embodiments describing application of support vector machine analysis to a mouth region. Lip position can be accurately determined and used in conjunction with synchronous or asynchronous audio data to enhance speech recognition probabilities.

Abstract: A pattern identification apparatus for identifying one of a plurality of classes defined in advance, to which data of a pattern identification target belongs, comprises a read unit adapted to read out, from a storage unit in correspondence with each of the plurality of classes, a projection rule to a hyperplane which approximates a manifold corresponding to the class in a feature space an input unit adapted to input identification target data; a calculation unit adapted to calculate, for each class, a projection result obtained by projecting the input identification target data to the hyperplane which approximates the manifold corresponding to each of the plurality of classes, on the basis of the projection rule; and an identification unit adapted to identify, on the basis of the projection result of each classes calculated by said calculation unit, one of the plurality of classes to which the identification target data belongs.

Abstract: A data visualization device comprises a focal attribute calculator for calculating a focal value of each of a plurality of data input items, and for categorizing each of the plurality of data input items as either non-cluster blobs or second blobs. The device further comprises a clustering attribute calculator for clustering data input items categorized as second blobs into one or more cluster blobs, and a visualization shaper for arranging the one or more cluster blobs and the non-cluster blobs, relative to each other in accordance with the focal values of the data input items of respective blobs.

Abstract: Systems and methods for object or pattern detection that use a nonlinear support vector (SV) machine are described. In the illustrated and described embodiment, objects or patterns comprising faces are detected. The decision surface is approximated in terms of a reduced set of expansion vectors. In order to determine the presence of a face, the kernelized inner product of the expansion vectors with the input pattern are sequentially evaluated and summed, such that if at any point the pattern can be rejected as not comprising a face, no more expansion vectors are used. The sequential application of the expansion vectors produces a substantial saving in computational time.

Abstract: The present invention relates to a method for classifying a signal (S). An intermediate signal (IS) is decimated in order to obtain a processed signal (PS) as a new intermediate signal (IS). A processed signal (PS) is compared with respect to the signal (S) to be classified to thereby generate comparison data (CompDAT). Based on the comparison data (CompDAT) said signal (S) to be classified is classified, thereby classification data (ClassDAT) are generated.

Abstract: A method is provided for evaluating identity of an object, the method including: converting feature information representing the object to a plurality of mathematically defined components; grouping the components into multiple modalities; producing respective first prediction information for each respective modality wherein the respective prediction information for each respective modality is based upon respective components grouped into that respective modality; and producing second prediction information based upon the respective first prediction information produced for the multiple respective modalities.

Abstract: An image detecting device for executing a pattern matching using a pyramid structure search, comprises an input image storing portion for receiving image data from an external device to hold, a pattern image storing portion for storing a pattern image to be searched, an image processing portion connected to the input image storing portion and the pattern image storing portion, for reading respective image data to execute the pattern matching, a control portion for calculating set parameter candidates that are required to execute the pyramid structure search in the image processing portion, a display portion for displaying groups of set parameter candidates that are calculated in the control portion as a table, and an input portion used to change set parameter values based on the groups of set parameter candidates displayed on the display portion.

Abstract: In computerized recognition having multiple experts, a method and system is described that obtains an optimum value for an expert tuning parameter in a single pass over sample tuning data. Each tuning sample is applied to two experts, resulting in scores from which ranges of parameters that correct incorrect recognition errors without changing correct results for that sample are determined. To determine the range data for a given sample, the experts return scores for each prototype in a database, the scores separated into matching and non-matching scores. The matching and non-matching scores from each expert are compared, providing upper and lower bounds defining ranges. Maxima and minima histograms track upper and lower bound range data, respectively. An analysis of the histograms based on the full set of tuning samples provides the optimum value. For tuning multiple parameters, each parameter may be optimized by this method in isolation, and then iterated.

Abstract: An image of an inspection object is picked up by an image pickup unit, and a characteristics quantity of an image of a defect part extracted by a defect extracting unit is extracted and digitized by a characteristics extracting unit. A database preparing unit regroups defects having similar characteristics by a defect classification unit on the basis of the characteristics information digitized by the characteristics extracting unit with respect to defects belonging to a defect group selected and designated by an operator via a display/input unit, and prepares on a database memory a database in which the defects of the inspection object are hierarchically classified.

Abstract: A digital watermark information detection method for detecting information inserted as a digital watermark from contents has a step of converting the contents into a predetermined specified format and a step of trying to detect information from the converted contents. When a plurality of specified formats are defined in advance and detection of the information from the converted contents fails, the contents is converted into another specified format and the information detection is retried.

Abstract: A method and system for categorizing non-textual subject data, such as digital images, content-based data and meta-data to determine outcomes of classification tasks. The meta-data is indicative of the operational conditions of a recording device during the capturing of the content-based data. For example, the non-textual subject data may be a digital image captured by a digital camera, and the meta-data may include automatic gain setting, film speed, shutter speed, aperture/lens index, focusing distance, date and time, and flash/no flash operation. The subject image is tagged with selected classifiers by subjecting the image to a series of classification tasks utilizing both content-based data and meta data to determine classifiers associated with the subject image.

Abstract: An image processing device of the present invention is provided with four kinds of sub masks in total including two kinds in a main scanning direction and two kinds in a sub scanning direction, in a main mask constituted by a plurality of pixels including a target pixel. In the image display device, when determining a target pixel of an inputted image data, a difference in a total density of the two kinds of sub masks in a main scanning direction is added to a normalized difference in total density of the two kinds of sub masks in a sub scanning direction, and a resultant value is compared with a threshold value so as to determine if the target pixel is an edge area or not.

Abstract: In computerized recognition having multiple experts, a method and system is described that obtains an optimum value for an expert tuning parameter in a single pass over sample tuning data. Each tuning sample is applied to two experts, resulting in scores from which ranges of parameters that correct incorrect recognition errors without changing correct results for that sample are determined. To determine the range data for a given sample, the experts return scores for each prototype in a database, the scores separated into matching and non-matching scores. The matching and non-matching scores from each expert are compared, providing upper and lower bounds defining ranges. Maxima and minima histograms track upper and lower bound range data, respectively. An analysis of the histograms based on the full set of tuning samples provides the optimum value. For tuning multiple parameters, each parameter may be optimized by this method in isolation, and then iterated.

Abstract: Complex multidimensional datasets generated by digital imaging spectroscopy can be organized and analyzed by applying software and computer-based methods comprising sorting algorithms. Combinations of these algorithms to images and graphical data, allow pixels or features to be rapidly and efficiently classified into meaningful groups according to defined criteria. Multiple rounds of pixel or feature selection may be performed based on independent sorting criteria. In one embodiment sorting by spectral criteria (e.g., intensity at a given wavelength) is combined with sorting by temporal criteria (e.g., absorbance at a given time) to identify microcolonies of recombinant organisms harboring mutated genes encoding enzymes having desirable kinetic attributes and substrate specificity. Restriction of the set of pixels analyzed in a subsequent sort based on criteria applied in an earlier sort (“sort and lock” analyses) minimize computational and storage resources.

Abstract: A resolution converting method, device, and computer-readable recording medium that can improve the quality of a resolution-converted image. The invention converts resolution based on a model generated through statistical processing of multiple images by preparing multiple models having different resolutions, receiving an input image, projecting the input image onto a model to obtain characteristic parameters of the input image, and generating an image by applying the obtained characteristic parameters to a model. Using the characteristic parameters thereby obtained, an image is generated using a model different from the model used to obtain the characteristic parameters. Consequently, the invention is capable of improving the quality of the resolution-converted image.

Abstract: A method and apparatus for searching, browsing and summarizing moving image data using the fidelity of a tree-structured moving image hierarchy, which can partition or divide the moving image data into predetermined units, extract a key frame or representative frame from each of the partitioned or divided units, create a tree-structured key frame hierarchy on the basis of the extracted key frames, store a fidelity value of each of the key frames, which is a sub-tree information representative degree of each key frame, at an edge between adjacent ones of the key frames and effectively and efficiently search, browse and summarize the moving image data using the stored fidelity values. Therefore, a query image or video clip desired by a user can more rapidly and accurately be searched for.

Abstract: A method for classification of objects in video image data. The method including the steps of: detecting moving objects in the image data; extracting two or more features from each detected moving object in the image data; classifying each moving object for each of the two or more features according to a classification method; and deriving a classification for each moving object based on the classification method for each of the two or more features. Also provided is an apparatus for classification of objects in video image data.