Abstract: An image processing method includes: a generating step of generating a cumulative histogram of density values of pixels included in at least each of a first image and a second image; a calculating step of calculating a distance between a density value on a first histogram generated from the first image and a density value on a second histogram generated from the second image, in a space defined to include histogram frequencies of the cumulative histograms and the density values; and a determining step of determining a correspondence relation between the density values included in the first image and the density values included in the second image based on the calculated distances between the density values, and determining a conversion function for correcting a density value between the first image and the second image based on the determined correspondence relation.

Abstract: A method of classification and segmentation of an image using modules on a computer system includes receiving a plurality of models having features suitable for classifying each pixel of the image into a respective one of a plurality of categories, using a classifier to provide a score for each pixel in the image for each category and using a segmenter to segment the image into image segments, wherein each image segment is a contiguous set of pixels having at least one common feature. For each image segment a set of average probabilities for each category is determined, and for each image segment, a most likely category to which the image segment belongs is determined by the maximum average probability resulting in a labeled segment image, which is used to identify any empty areas as incorrect holes. Then any empty areas that are identified as incorrect holes are filled.

Abstract: A learning unit generates a function table indicating the relationship between the class number and position information of an object and the probability of appearance of the object for each small area image pattern of a code book, calculates a sharing matrix indicating the commonality of a feature amount between the classes, makes a tree diagram in which the classes with a similar feature amount are clustered, and calculates the weight of each node in the tree diagram for each small area image pattern. The recognition processing unit compares image data captured by a camera with the code book, selects the closest small area image pattern, extracts the class related to the node with the smallest weight among the nodes with a weight equal to or greater than a threshold value, and votes the position information of the small area image pattern for the class, thereby recognizing the object.

Abstract: Blur metrics may be calculated for each of the image pixels of a digital image of a scene captured using an imaging device. The blur metrics may be indicative of the level of blur expressed in the digital image, and a blur image representative of the blur metrics may be generated. Subsequently, when another digital image is to be captured using the imaging device, pixel sensors corresponding to high blur metrics may be digitized at a high level of priority, or at a high rate, compared to pixel sensors corresponding to low blur metrics, which may be digitized at a low level of priority, or at a low rate. The blur images may be updated based on changes in blur observed in subsequent images, and different pixel sensors may be digitized at higher or lower levels of priority, or at higher or lower rates, based on the changes in blur.

Abstract: Systems and methods for providing recommendations to obfuscate an entity context in a sensory environment are described. In one implementation, the method comprises receiving entity data and sensory environment data from a plurality of sources. Further, the method comprises analyzing the entity data to obtain categorized entity data. The categorized entity data comprises a plurality of features indicating characteristics of the entity context. Further, the method comprises analyzing the sensory environment data to obtain categorized sensory environment data. Further, the method comprises determining a correlation score for each of the plurality of features by correlating the categorized entity data and the categorized sensory environment data. Further, the method comprises recommending at least one of the plurality of features, based on the correlation score, to obfuscate the entity context in the sensory environment.

Abstract: In one embodiment, a method for encoding video data is provided that includes receiving an array of transform coefficients corresponding to a luma component or a chroma component of the video data. The method further includes encoding a significance map for the array, where the encoding includes selecting, using a shared context derivation process that applies to both the luma component and the chroma component, context models for encoding significance values in the significance map.

Abstract: Various methods are described for improving the compression of location information during a visual search. One example method may comprise identifying one or more features of a source query image for a visual search. The method may further comprise generating a histogram map comprising one or more histogram bins, wherein the histogram map represents the location of the one or more features of the source query image. Additionally, the method may comprise generating a context for at least one of the one or more histogram bins, wherein the context comprises a first context value based on information related to one or more neighboring bins located a first distance from the respective histogram bin and a second context value based on information related to one or more neighboring bins located a second distance from the respective histogram bin. Similar and related methods, apparatuses, and computer program products are also provided.

Abstract: Subjective user-generated data can be utilized to determine visually similar items. Various item descriptors can be determined for a pair of items, which can provide an objective measure of visual aspects of those items, such as how similar those items are in color, style, material, or texture. The ways in which users interact with data for those items, either explicitly or implicitly through user behavior, can provide a level of perceived visual similarity on behalf of these users. The perceived, subjective visual similarity data from the users can be used to adjust a weighting of the various item descriptor factors for a pair of items, or otherwise adjust a visual similarity score, such that items selected as being visually similar more accurately reflect the subjective opinions of the users.

Abstract: This disclosure provides method and systems of classifying a digital image of an object. Specifically, according to one exemplary embodiment, an object classifier is trained using a constrained MI-SVM (multiple instance-support vector machine) approach whereby training images of objects are sampled to generate a collection of image regions associated with an object type and viewpoint, and the classifier is trained to determine an appropriate mid-level representation of the training image which is discriminative.

Abstract: Techniques describe submitting a video clip as a query by a user. A process retrieves images and information associated with the images in response to the query. The process decomposes the video clip into a sequence of frames to extract the features in a frame and to quantize the extracted features into descriptive words. The process further tracks the extracted features as points in the frame, a first set of points to correspond to a second set of points in consecutive frames to construct a sequence of points. Then the process identifies the points that satisfy criteria of being stable points and being centrally located in the frame to represent the video clip as a bag of descriptive words for searching for images and information related to the video clip.

Abstract: In some implementations, legibility of an image may be automatically determined based, at least in part, on text contained in the image. For example, image analysis techniques may be used to identify text components in an image. One or more features of each text component may be determined for use in assessing the legibility of the text component. For example, a classifier trained on the one or more features may provide a confidence level indicative of the legibility of each text component. The confidence level for each of the text components may be compared to a legibility threshold for determining whether the text component is legible or illegible. Based, at least in part, on the determination as to how much of the text in the image is legible or illegible, an overall legibility of the image may be assessed.

Abstract: Techniques are provided in which a plurality of edges are detected within a digital image. An anchor point located along an edge of the plurality of edges is selected. An analysis grid associated with the anchor point is generated, the analysis grid including a plurality of cells. An anchor point normal vector comprising a normal vector of the edge at the anchor point is calculated. Edge pixel normal vectors comprising normal vectors of the edge at locations along the edge within the cells of the analysis grid are calculated. A histogram of similarity is generated for each of one or more cells of the analysis grid, each histogram of similarity being based on a similarity measure between each of the edge pixel normal vectors within a cell and the anchor point normal vector, and a descriptor is generated for the analysis grid based on the histograms of similarity.

Abstract: Provided are a method and device for measuring a critical dimension of a nanostructure.

Abstract: A method for recognizing and locating an object includes an offline mode process and an online mode process. In the offline mode process, plural sampled edge points of a template image of the object and respective gradient angles and a gravity position of the plural sampled edge points are obtained, and plural similarity score tables are obtained according to the plural sampled edge points, a predetermined detecting distance range and a predetermined gradient angle difference range. In the online mode process, plural edge points of a live image and respective gradient angles are obtained, plural predictive gravity positions are calculated, and plural similarity scores corresponding to the plural predictive gravity positions are summed up. The predictive gravity position with the local maximum of the similarity scores higher than a threshold value is correlated with the gravity position of the template image so as to recognize and locate the object.

Abstract: Systems, methods, and device are provided to perform refresh-rate dependent dithering. One embodiment of the present disclosure describes a computing device that includes an image source that generates spatially dithered image data and an electronic display communicatively coupled to the image source. More specifically, the electronic display receives the spatially dithered image data from the image source and determines a refresh rate with which to display an image by comparing a local histogram and an artifact histogram, in which the local histogram describes pixel grayscale distribution of a portion of the image and the artifact histogram describes a pixel grayscale distribution that when displayed will cause a perceivable artifact.

Abstract: A feature value extraction apparatus is provided that includes an image input unit that receives an input of an image, and a converted image generation unit that converts the input image into at least one converted image for extracting a feature value of the input image. The apparatus also includes a feature value generation pattern holding unit that stores a feature value generation pattern that includes a plurality of comparison pairs of unit positions for comparing unit values in a unit array in which units, each of which includes at least one pixel, are arranged. The apparatus further includes a feature value extraction unit that compares the unit values corresponding to the comparison pairs through referring to the feature value generation pattern with respect to the converted image, and extracts the feature value of the converted image.

Abstract: An interaction method for a dynamic wallpaper and a desktop component including providing a dynamic wallpaper and at least one desktop component on a terminal screen, with the dynamic wallpaper including a background image, providing a screen capturing module for capturing the background image loaded with the desktop component to form a buffer file, and coding the same as an image format for storage, providing an edge detection module for performing pixel scanning comparison on the background image and the stored buffer image, circularly performing pixel scanning comparisons until the edge location data of the at least one desktop component is obtained, and changing the dynamic effects of the dynamic wallpaper according to the data obtained.

Abstract: A near-touch interface is provided that utilizes stereo cameras and a series of targeted structured light tessellations, emanating from the screen as a light source and incident on objects in the field-of-view. After radial distortion from a series of wide-angle lenses is mitigated, a surface-based spatio-temporal stereo algorithm is utilized to estimate initial depth values. Once these values are calculated, a subsequent refinement step may be applied in which light source tessellations are used to flash a structure onto targeted components of the scene, where initial near-interaction disparity values have been calculated. The combination of a spherical stereo algorithm, and smoothing with structured light source tessellations, provides for a very reliable and fast near-field depth engine, and resolves issues that are associated with depth estimates for embedded solutions of this approach.

Abstract: Various embodiments provide a method for determining color information for an image. For example, a color descriptor for an image can be determined and compared against color descriptors stored for each of a number of sample images, which each represent a color in a color space. Upon comparison, matching scores can be generated for a color match between the image and each respective sample image. In this example, the number of sample images with a matching score above a threshold value can be summed and the image can be assigned to a color associated with a highest frequency of the number of sample images. Accordingly, the assigned color of the image can then be used in a “query by color” search or a browse-by-color capability.

Abstract: For testing an object recognition device for a motor vehicle at reasonable costs for different routes, image data for testing the object recognition device may be generated with a camera simulation device. Because the image data of a camera simulation device are artificially generated, it must be made certain that they have a realistic effect on the object recognition device. Reference image data are generated with a camera and simulation image data are generated with the camera simulation device for at least one route. The simulation image data and the reference image data are compared with each other based on at least two comparison measures. A value which is independent of the object recognition device to be tested can be determined for each of the comparison measures. It is then checked if the totality of the generated comparison values satisfies a predetermined validation criterion.

Abstract: A linear function describing a framework for identifying an object of class k in an image sample x may be described by: wk*x+bk, where bk is the bias term. The higher the value obtained for a particular classifier, the better the match or strength of identity. A method is disclosed for classifier and/or content padding to convert dot-products to distances, applying a hashing and/or nearest neighbor technique on the resulting padded vectors, and preprocessing that may improve the hash entropy. A vector for an image, an audio, and/or a video may be received. One or more classifier vectors may be obtained. A padded image, video, and/or audio vector and classifier vector may be generated. A dot product may be approximated and a hashing and/or nearest neighbor technique may be performed on the approximated dot product to identify at least one class (or object) present in the image, video, and/or audio.

Abstract: A method for processing an image including: identifying a first group of keypoints in the image; for each keypoint of the first group, identifying at least one corresponding keypoint local feature related to the each keypoint; for the at least one keypoint local feature, calculating a corresponding local feature relevance probability; calculating a keypoint relevance probability based on the local feature relevance probabilities of the at least one local feature; selecting keypoints, among the keypoints of the first group, having the highest keypoint relevance probabilities to form a second group of keypoints, and exploiting the keypoints of the second group for analyzing the image. The local feature relevance probability calculated for a local feature of a keypoint is obtained by comparing the value assumed by the local feature with a corresponding reference statistical distribution of values of the local feature.

Abstract: Methods and apparatuses for updating and searching a graphic code are disclosed. The apparatus receives a graphic query request carrying a graphic code to be searched. If no product information corresponding to the graphic code is found in the graphic code database, the apparatus searches for a recommended keyword corresponding to the graphic code; if a recommended keyword is found, the apparatus returns the recommended keyword, and if not found, returns a search box. The apparatus then receives a search request carrying a search keyword, and, if the search keyword is determined to be inconsistent with the recommended keyword corresponding to the graphic code, makes the search keyword a new recommended keyword of the graphic code, and adds the new recommended keyword to the graphic code database.

Abstract: Desired portions of a background image presented on a mobile device are prioritized for viewability. The background image, which includes desired feature components, is presented on a display of the mobile device. The desired feature components are identified from the background image. A user interface that includes an indication of the desired feature components is presented on the mobile device. One or more icon images are presented in accordance with the prioritization of the desired feature components of the background image.

Abstract: In one exemplary aspect, a sorted list of scored media content episodes is received with a computing device of a user. Each respective media content episode is scored by an iterative autotuning prediction algorithm, and wherein each element of the sorted list of scored media content episodes comprises a value that represents a likelihood of a user listening to the respective media content episode and a reference to a location of the respective media content episode. A number of bytes of a download iteration for each media content episode is determined based on value that represents a likelihood of the user listening to the respective media content episode and an index of the respective media content episode in the sorted list. It is detected that a mobile device is in the preferred network. The download iteration is implemented for each media content episode when it is detected that the mobile device is in the preferred network.

Abstract: A method, apparatus, and article of manufacture provide the ability to generate a damage proxy map. A master coherence map and a slave coherence map, for an area prior and subsequent to (including) a damage event are obtained. The slave coherence map is registered to the master coherence map. Pixel values of the slave coherence map are modified using histogram matching to provide a first histogram of the master coherence map that exactly matches a second histogram of the slave coherence map. A coherence difference between the slave coherence map and the master coherence map is computed to produce a damage proxy map. The damage proxy map is displayed with the coherence difference displayed in a visually distinguishable manner.

Abstract: An apparatus and method for calculating a cumulative histogram of an image are provided. A cumulative histogram calculation apparatus may include a cumulative value selecting unit to select cumulative data obtained by accumulating input data, based on a number of combinations of the input data, and a loading unit to load the selected cumulative value in a corresponding bin of a histogram.

Abstract: An image processing device that, even when a degradation model differs from a real degradation process of an obtained low-resolution image, generates a high-resolution image while restricting noises from occurring due to differences between the degradation model and the real degradation process.

Abstract: An image processing device for detecting a skin region representing a skin of a subject from a pickup image obtained by imaging said subject, the image processing device includes: a first irradiating section; a second irradiating section; an image pickup section; an adjusting section; and a skin detecting section.

Abstract: Subjective user-generated data can be utilized to determine visually similar items. Various item descriptors can be determined for a pair of items, which can provide an objective measure of visual aspects of those items, such as how similar those items are in color, style, material, or texture. The ways in which users interact with data for those items, either explicitly or implicitly through user behavior, can provide a level of perceived visual similarity on behalf of these users. The perceived, subjective visual similarity data from the users can be used to adjust a weighting of the various item descriptor factors for a pair of items, or otherwise adjust a visual similarity score, such that items selected as being visually similar more accurately reflect the subjective opinions of the users.

Abstract: A method detects anomalies in time series data, wherein the time series data is multivariate, by partitioning time series training data into partitions. A representation for each partition in each time window is determined to form a model of the time series training data, wherein the model includes representations of distributions of the time series training data. The representations obtained from partitions of time series test data are compared to the model to obtain anomaly scores.

Abstract: Systems and methods are provided herein relating to speed resistant audio matching. Descriptors can be generated for a received audio signal and matched with reference descriptors. A set of hits for respective reference samples can be generated based on the matching. A histogram can then be generated that correlates probe sample hit time with reference sample hit time. In one implementation, a rolling window can be used in analyzing the histogram allowing for slight variances in the timing between probe sample hits and reference sample hits. In another implementation, the histogram generated can be based on an estimated time stretch of the probe sample. In yet another implementation, a set of histograms can be generated based on a minimum speed change, a maximum speed change, and a speed step. Histograms can be evaluated to determine a most likely matching histogram.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining image search results. One of the methods includes generating a plurality of feature vectors for each image in a collection of images, wherein each feature vector is associated with an image tile of an image, wherein each feature vector corresponds to one of a plurality of predetermined visual words. All images in the collection of images that share at least a threshold number of matching visual words associated with matching image tiles are classified as near-duplicate images.

Abstract: Various embodiments provide a method for computing color descriptors of product images. For example, a number of fine color representatives can be determined to describe color variation in an image as a histogram by assigning a saturation value and a brightness value to a plurality of color hues. For each pixel of the image, the closest color among a defined fine color representative set is computed. In this example, each of the pixels is assigned a color ID corresponding to their closest matching fine color representative and at least one family color ID corresponding one or more pure color families. In this example, a histogram of the color representatives and a histogram for the color families are computed. A single color vector descriptor for the image is then determined by combining the family histogram with the color representative histogram.

Abstract: A method for detecting features in digital numeric data comprises obtaining digital numeric data comprising values corresponding to a plurality of sampling points over a domain space having at least one dimension, computing a plurality of scale-space data comprising filtering said digital numeric data using a filter bank, determining a plurality of feature regions each corresponding to a local extremum in scale and location of the scale-space data; and determining a feature region descriptor for each of said plurality of feature regions. The filter bank is a Cosine Modulated Gaussian filter bank in which the standard deviation parameter of the Gaussian equals 1 ? ? ln ? ( 2 ) 2 ? 2 b + 1 2 b - 1 multiplied by the cosine wavelength, in which b is in the range of 0.75 to 1.25, or said filter bank is an Nth-order Gaussian Derivative filter bank with N being in the range of 5 to 20.

Abstract: A method of identifying groups of related digital images in a digital image collection, comprising: analyzing each of the digital images to generate associated feature descriptors related to image content or image capture conditions; storing the feature descriptors associated with the digital images in a metadata database; automatically analyzing the metadata database to identify a plurality of frequent itemsets, wherein each of the frequent itemsets is a co-occurring feature descriptor group that occurs in at least a predefined fraction of the digital images; determining a probability of occurrence for each the identified frequent itemsets; determining a quality score for each of the identified frequent itemsets responsive to the determined probability of occurrence; ranking the frequent itemsets based at least on the determined quality scores; and identifying one or more groups of related digital images corresponding to one or more of the top ranked frequent itemsets.

Abstract: One or more systems and/or techniques are provided to identify objects comprised in a compound object without segmenting three-dimensional image data of the potential compound object. Two-dimensional projections of a potential compound object (e.g., Eigen projections) are examined to identify the presence of known objects. The projections are compared to signatures, such as morphological characteristics, of one or more known objects. If it is determined based upon the comparison that there is a high likelihood that the compound object comprises a known object, a portion of the projection is masked, and it is compared again to the signature to determine if this likelihood has increased. If it has, a sub-object of the compound object may be classified based upon characteristics of the known object (e.g., the compound object may be classified as a potential threat item if the known object is a threat item).

Abstract: A method and a system for detecting and monitoring of at least one object in a multimedia content are provided. The method includes extracting at least one contour from the multimedia content using non-textural information in a segmented region within the multimedia content, computing a histogram for the extracted at least one contour in the segmented region of the selected multimedia content to represent an object shape based on the computed histogram, and determining an orientation change of the represented object shape within the segmented region of the multimedia content.

Abstract: A method and system for processing a sequence of images using fingerprints. A current image in the sequence of images is segmented into a plurality of segments. A plurality of segment fingerprints is created for the plurality of segments. Segments in the plurality of segments are fused together using the plurality of segment fingerprints and a set of prior segment fingerprints created for a previous image in the sequence of images to form a set of master segments.

Abstract: The Embodiments provides a method for detecting leakage stage associated with a multimedia. The method includes storing histograms associated with various stages of the multimedia. Further, the method includes receiving candidate histograms associated with various stages of a candidate multimedia, matching the stored histograms with the candidate histograms, and detecting a leakage stage associated with the multimedia in response to a match.

Abstract: Image similarity operations are performed in which a seed image is analyzed, and a set of semantic classifications are determined from analyzing the seed image. The set of semantic classifications can include multiple positive semantic classifications. A distance measure is determined that is specific to the set of semantic classifications. The seed image is compared to a collection of images using the distance measure. A set of similar images is determined from comparing the seed image to the collection of images.

Abstract: An image forming apparatus includes a specific graphic form storage section, an image determination section, and determination color conversion processing section. The specific graphic form storage section is configured to store a visual image of a specific graphic form and attribute information specific to the specific graphic form as a reference image and reference attribute information, respectively. The image determination section is configured to determine at least one of (i) whether or not an image matching with the reference image is included in target data and (ii) whether or not an image having the reference attribute information is included in the target data, and detect the image identified through the determination as an image corresponding to the specific graphic form. The color conversion processing section is configured to perform specific color conversion processing for the image corresponding to the specific graphic form.

Abstract: A hardware coprocessor architecture calculates the Difference-of-Gaussian (DoG) pyramid of an input image and extracts from this the interest points to be used in several image detection algorithms. Advantages of the architecture include the possibility to process the image by stripes, namely by blocks having one dimension coincident with the input image width, in the absence of an input frame buffer and the possibility to avoid RAM memory. The coprocessor is suitable to be tightly coupled with raw image sources like sensors.

Abstract: A method is disclosed for analyzing video to detect far-view scenes in sports video to determine when certain image processing algorithms should be applied. The method comprises analyzing and classifying the fields of view of images from a video signal, creating and classifying the fields of view of sets of sequential images, and selectively applying image processing algorithms to sets of sequential images representing a particular type of field of view.

Abstract: An apparatus includes a first acquisition unit configured to acquire main object information specifying a main object in generation of a layout image, a second acquisition unit configured to acquire object correlation information specifying an object having a correlation with the main object, an extraction unit configured to extract at least one image including the main object and at least one image including the object having the correlation with the main object from a plurality of images based on the acquired main object information and the acquired object correlation information acquired, and a generation unit configured to generate, using a layout template, a layout image in which the at least one image extracted by the extraction unit and including the main object and the at least one image extracted by the extraction unit and including the object having the correlation with the main object are laid out therein.

Abstract: A method and apparatus for detecting and recognizing an object using a vector histogram based on a local binary pattern are disclosed. The apparatus of detecting and recognizing an object using a local binary pattern includes: a feature map creator configured to extract an object area in which a moving object exists from an input image, to create a local binary pattern by designating a local area in the object area, and to create a vector component map including information about magnitude vector components and direction vector components using the local binary pattern; a feature map configuring unit configured to divide the object area into a plurality of blocks and to create a feature vector map through a histogram using the vector component map in a unit of the block; and an object detector configured to detect and classify the moving object based on the feature vector map.

Abstract: A method for comparing a first image with a second image. The method identifies first keypoints in the first image and second keypoints in the second image and associates each first keypoint with a corresponding second keypoint to form a corresponding keypoint match. For each pair of first keypoints, the method further calculates the distance therebetween for obtaining a corresponding first length. Similarly, for each pair of second keypoints, the method calculates the distance therebetween for obtaining a corresponding second length. The method further calculates a plurality of distance ratios; each distance ratio is based on a length ratio between a selected one between a first length and a second length and a corresponding selected one between a second length and a first length, respectively.

Abstract: A method of cropping an image based on a content of the image according to one example embodiment includes identifying image features in the content of the image, determining horizontal histogram frequencies and vertical histogram frequencies of the identified image features, determining borders of a document bounding box based on the determined horizontal histogram frequencies and vertical histogram frequencies, and cropping the image according to the document bounding box.

Abstract: A method for processing data includes receiving a depth map of a scene containing a human hand, the depth map consisting of a matrix of pixels having respective pixel depth values. The method continues by extracting from the depth map respective descriptors based on the depth values in a plurality of patches distributed in respective positions over the human hand, and matching the extracted descriptors to previously-stored descriptors in a database. A pose of the human hand is estimated based on stored information associated with the matched descriptors.

Abstract: Parallel processing of an image using an array of addressable registers. Image features are extracted from the image. The image features are storable as data. According to respective values of a sorting key derived from a parameter of the data, the image features are sorted into N buckets. Using an array of M addressable registers, where M is less than N, the data are summed within the buckets to perform a histogram of the image features, according to values of a histogram key derived from said a parameter of the data.