Abstract: A portable electronic device with camera function and an image capture method with auto exposure control. The disclosed method comprises the following steps: obtaining a pre-view frame from a camera module; dividing the pre-view frame into a plurality of blocks and calculating representative brightness values of the blocks; assigning weight values to the blocks according to the representative brightness values; obtaining a weighted brightness value by performing a weighted operation on the representative brightness values based on the weight values of the blocks; setting an auto exposure parameter of the camera module based on the weighted brightness value; and controlling the camera module for capturing images based on the auto exposure parameter.

Abstract: Machines, systems and methods for object relationship characterization are provided. The method comprises providing a plurality of images, each having a plurality of pixels; selecting a pair of images from the plurality of images, the pair of images comprises a first image and a second image; characterizing at least one pixel of the first image and the second image by a first feature vector and a second feature vector respectively; characterizing the first image by a first probability distribution over the first feature vector; characterizing the second image by a second probability distribution over the second feature vector; assigning a list of histogram bins for the first image and the second image; computing a distribution flow descriptor (DFlow) for capturing relationship between the first probability distribution and the second probability distribution.

Abstract: Determining a label from an image is disclosed, including: obtaining an image; determining a first portion of the image associated with a special mark; determining a second portion of the image associated with a label based at least in part on the first portion of the image associated with the special mark; and applying character recognition to the second portion of the image associated with the label to determine a value associated with the label.

Abstract: The present invention is a template matching processing device capable of evaluating a similarity degree which supports even a case of intensive morphological change between a design image and a photographic image. In the template matching processing device, matching processing between the design image and the photographic image is performed, a partial design image is obtained by clipping a portion having the highest correlation (step 101), and processing for deforming the photographic image in accordance with the clipped design image (steps 102 to 105) is performed, so that correlation between the deformed image obtained and the design image is taken to be set as the similarity degree.

Abstract: An image matching system is described for finding reference images that match a query image. The image matching system performs this operation in expeditious fashion by restricting the matching to a bounding region; the bounding region, in turn, may be associated with a location at which the query image may have been captured. In addition, the image matching system provides various mechanisms that expedite adding new reference images to an image index, to thereby provide a re-enforced learning mechanism of the image matching system.

Abstract: A method for processing an image to generate a signature which is characteristic of a pattern within said image. The method comprising receiving an image; overlaying a window at multiple locations on said image to define a plurality of sub-images within said image, with each sub-image each having a plurality of pixels having a luminance level; determining a luminance value for each said sub-image, wherein said luminance value is derived from said luminance levels of said plurality of pixels; and combining said luminance values for each of said sub-images to form said signature. Said combining is such that said signature is independent of the location of each sub-image. A method of creating a database of images using said method of generating signatures is also described.

Abstract: Embodiments of the subject technology provide methods and systems of image pre-processing for improving the accuracy of optical character recognition (OCR) and reducing the power consumption on a given computing device (e.g., mobile computing device). The subject technology, in some examples, classifies an image received from a camera of a mobile computing device into one or more classes: 1) normal background, 2) textured background, 3) image with text, 4) image with barcode, 5) image with QR code, and/or 6) image with clutter or “garbage.” Based on the classes associated with the image, the subject technology may forgo certain image processing operations, when the image is not associated with a particular class, in order to save resources (e.g., CPU cycles, battery power, memory usage, etc.) on the mobile computing device.

Abstract: In general, techniques are described for performing a vocabulary-based visual search using multi-resolution feature descriptors. A device may comprise one or more processors configured to perform the techniques. The processors may generate a hierarchically arranged data structure to be used when classifying objects included within a query image based on multi-resolution query feature descriptor extracted from the query image at a first scale space resolution and a second scale space resolution. The hierarchically arranged data structure may represent a first query feature descriptor of the multi-resolution feature descriptor extracted at the first scale space resolution and a second corresponding query feature descriptor of the multi-resolution feature descriptor extracted at the second scale space resolution hierarchically arranged according to the first scale space resolution and the second scale space resolution. The processors may then perform a visual search based on the generated data structure.

Abstract: In general, techniques are described for performing a vocabulary-based visual search using multi-resolution feature descriptors. A device may comprise one or more processors configured to perform the techniques. The one or more processors may to apply a partitioning algorithm to a first subset of target feature descriptors to determine a first classifying data structure to be used when performing a visual search with respect to a query feature descriptor. The one or more processors may then apply the partitioning algorithm to a second subset of the target feature descriptors to determine a second classifying data structure to be used when performing the visual search with respect to the same query feature descriptor.

Abstract: A method may involve analyzing eye-image data to determine observed movement of a reflected image over a predetermined period, using head-movement data to determine a first eye-movement component corresponding to head movement during the predetermined period, determining a first expected movement of the reflected image corresponding to the first eye-movement component, determining a second eye-movement component based on a difference between the observed movement and the first expected movement, determining a second expected movement of the reflected image based on the combination of the first eye-movement component and the second eye-movement component, determining a difference between the observed movement and the second expected movement, if the difference is less than a threshold, setting eye-movement data for the predetermined period based on the second eye-movement component; and if the difference is greater than the threshold, adjusting the first eye-movement component and repeating the method with the

Abstract: Systems and methods are provided for detecting anatomical components in images. In accordance with one implementation, at least one anchor landmark is detected in an image. The position of the anchor landmark is used to detect at least one bundle landmark in the image. In accordance with another implementation, at least two neighboring landmarks are detected in an image, and used to detect at least one anatomical primitive in the image.

Abstract: A moving object detection device includes a window setting unit configured to set a window having a predetermined volume in a video, an orientation of spatial intensity gradient calculation unit configured to calculate, for each pixel included in the window, an orientation of spatial intensity gradient, a spatial histogram calculation unit configured to calculate a spatial histogram that is a histogram of the orientation of spatial intensity gradient within the window, an orientation of temporal intensity gradient calculation unit configured to calculate, for each pixel included in the window, an orientation of temporal intensity gradient, a temporal histogram calculation unit configured to calculate a temporal histogram that is a histogram of an orientation of temporal intensity gradient within the window, and a determination unit configured to determine whether or not the moving object is included within the window based on the spatial histogram and the temporal histogram.

Abstract: Provided are: a feature extraction method that is a method of extracting a feature that represents a characteristic of a subject, said feature being extracted from an image that has imaged the subject, said feature being extracted without relation to the shape of said subject; and methods that use the extracted feature to classify and identify the subject. The feature extraction method extracts the feature from the image of the subject, said subject having been imaged by an imaging means.

Abstract: Methods, apparatus and articles of manufacture for detecting objects in images using color histograms are disclosed. Example methods disclosed herein include determining differences between bin values of a first color histogram corresponding to an object and respective adjusted bin values of a second color histogram corresponding to a first subregion of an image. Such disclosed example methods also include determining a first metric based on the differences. Such disclosed example methods further include comparing the first metric to a threshold to determine whether the first subregion of the image corresponds to a first possible location of the object in the image.

Abstract: According to one embodiment, an image processing apparatus includes a detail extraction module, a detail addition control module, and a detail component addition module. The detail extraction module extracts a detail component from an image signal of one frame. The detail addition control module controls an addition quantity of a detail component. The detail component addition module adds a detail component controlled by the detail addition control module to the image signal.

Abstract: Methods, systems, and apparatus, including computer program products, for identifying regions of interest in an image and identifying a barcode in a degraded image are provided. A region of interest is identified by pre-processing an image, generating a binary image based on a metric calculated on the pre-processed image, and analyzing regions of the image identified using connected components and other analysis. A barcode is identified by searching a population of barcodes, degrading ideal image intensity profiles of candidate barcodes, and comparing the degraded ideal image intensity profiles to an image intensity profile of the degraded image.

Abstract: A method for applying a filter to data to improve data quality and/or reduce file size. In one example, a region of interest of an image is identified. A histogram is generated of pixel intensity values in the region of interest. The histogram is iteratively updated to focus (zoom) in on the highest peak in the histogram. A Gaussian curve is fitted to the updated histogram. A bilateral filter is applied to the images, where parameters of the bilateral filter are based on the parameters of the Gaussian curve.

Abstract: A method comparing first and second images, including: identifying and matching first and second keypoints in the first and second images; arranging a distribution of values of a calculated plurality of first distance ratios in a histogram; determining a number of correct keypoint matches, including: determining a matrix, each matrix element corresponding to a respective pair of keypoint matches with a value corresponding to a difference between a value of the histogram including a distance ratio of the respective pair of keypoint matches and an outlier probability density value weighted by a parameter; determining the parameter value such that the matrix dominant eigenvector is equal to a vector with a first value if the pair of keypoint match is correct and a second value if the pair of keypoint match is incorrect; determining the number of correct keypoint matches based on the dominant eigenvalue associated to the dominant eigenvector.

Abstract: Techniques for generating a personalized cartoon by using a few text queries are described herein. The present disclosure describes efficiently searching multiple images from a network, obtaining clipart image from the multiple images, and vectorization of the clipart image. The present disclosure also describes techniques to change a style of the cartoon such as recoloring one or more cartoon objects.

Abstract: To provide an image evaluation apparatus and camera which are capable of evaluating an image which is comprehensively good. An image evaluation apparatus comprising: a storage portion which stores a plurality of images when an imaging portion of a camera has continuously imaged in a predetermined time interval a plurality of frames of a subject in a photographing area of the camera, and an image evaluation portion which evaluates a plurality of the images stored in the storage portion, based on a characteristic amount showing a state in which the subject in the photographing area changing in chronological sequence, when the plurality of the images is imaged.

Abstract: Methods, systems, and apparatus, including computer program products, for identifying regions of interest in an image and identifying a barcode in a degraded image are provided. A region of interest is identified by pre-processing an image, generating a binary image based on a metric calculated on the pre-processed image, and analyzing regions of the image identified using connected components and other analysis. A barcode is identified by searching a population of barcodes, degrading ideal image intensity profiles of candidate barcodes, and comparing the degraded ideal image intensity profiles to an image intensity profile of the degraded image.

Abstract: Video sequence processing is described with various filtering rules applied to extract dominant features for content based video sequence identification. Active regions are determined in video frames of a video sequence. Video frames are selected in response to temporal statistical characteristics of the determined active regions. A two pass analysis is used to detect a set of initial interest points and interest regions in the selected video frames to reduce the effective area of images that are refined by complex filters that provide accurate region characterizations resistant to image distortion for identification of the video frames in the video sequence. Extracted features and descriptors are robust with respect to image scaling, aspect ratio change, rotation, camera viewpoint change, illumination and contrast change, video compression/decompression artifacts and noise.

Abstract: A method of calculating a drop delay in a stream of a flow cytometer. In one embodiment the method includes the steps of determining a widths plot by measuring the width of a stream based on an image of the stream and setting drop delay based on the widths plot. In another aspect the invention relates to a flow cytometer system for automatically calculating drop delay. In one embodiment the flow cytometer system includes means for determining a widths plot based on an image of the stream; and means for calculating drop delay based on the widths plot.

Abstract: A system and a method are disclosed that determine images with co-occurrence groups of individuals from an image collection. A value of a similarity metric is computed for each pair of images of the image collection, the value of the similarity metric being computed based on a comparison of the number of individuals in common between the images of the pair and the total number of individuals identified in both images of the pair. The collection of images is clustered based on the computed values of the similarity metric. At least one co-occurrence group is determined based on the results of the clustering, where a co-occurrence group is determined as a cluster of images that have a similar combination of individuals.

Abstract: A server device determines a plurality of images for a query. One or more images, of the plurality of images, are associated with one or more senses of the query. The server device maps the plurality of images into a space by representing the plurality of images with corresponding points in the space; determines one or more hyperplanes in the space based on the corresponding points in the space; calculates one or more scores for the plurality of images based on the corresponding points and the one or more hyperplanes; and ranks the one or more images based on the one or more scores.

Abstract: Embodiments of the present disclosure include systems and methods for image processing for enhancing, e.g., photos, videos, and graphics representations. Particularly, some embodiments perform a histogram-based SMQT decomposition of various attributes in an original image. The SMQT decomposition of these attributes may then be used to create a look-up-table indicating how to modify pixel attributes in the same or a subsequent image so as to enhance the same or subsequent image. The results of the SMQT decomposition for each attribute may be interpolated based, e.g., on user preferences, to achieve a more optimal enhancement.

Abstract: Systems and methods are provided for detecting sharpness among video fields. In certain implementations of the systems and methods, a plurality of video fields is received and a sharpness metric for each of the plurality of video fields is determined. The sharpness metric of a first video field is compared to the sharpness metric of a second video field among the plurality of video fields and a video field source of the first video field and the second video field is determined based on the comparison.

Abstract: In a first exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, calculating processing scale parameter information as a function of width dimension information for penumbrae depicted in the image; and generating intrinsic illumination and material reflectance images corresponding to the image using the processing scale parameter information.

Abstract: Disclosed is a software routine which determines which photographs in a corpus are similar, groups the similar photographs, and which then determines which photographs within a group meet criteria of “better” photographs.

Abstract: Image recognition methods, apparatus and articles or manufacture to support shelf auditing for consumer research are disclosed herein. An example method disclosed herein comprises comparing an input image depicting a shelf to be audited with a set of reference images depicting reference shelves displaying respective sets of reference items, identifying a first reference image from the set of reference images that has been determined to match the input image, and determining an initial audit result for the shelf depicted in the input image based on a first set of reference items associated with the first reference image.

Abstract: Technologies are generally presented for employing enhanced expectation maximization (EEM) in image retrieval and authentication. Using uniform distribution as initial condition, the EEM may converge iteratively to a global optimality. If a realization of the uniform distribution is used as the initial condition, the process may also be repeatable. In some examples, a positive perturbation scheme may be used to avoid boundary overflow, often occurring with the conventional EM algorithms. To reduce computation time and resource consumption, a histogram of one dimensional Gaussian Mixture Model (GMM) with two components and wavelet decomposition of an image may be employed.

Abstract: The disclosure describes novel technology for inferring scenes from images. In one example, the technology includes a system that can determine partition regions from one or more factors that are independent of the image data, for an image depicting a scene; receive image data including pixels forming the image; classify pixels of the image into one or more pixel types based on one or more pixel-level features; determine, for each partition region, a set of pixel characteristic data describing a portion of the image included in the partition region based on the one or more pixel types of pixels in the partition region; and classify a scene of the image based on the set of pixel characteristic data of each of the partition regions.

Abstract: A method for occupancy detection is disclosed. The method may include capturing an image, moving a sliding window over said image, determining features for intensity image and gradient image, generating a strong classifier, detecting shape of object; and determining occupancy.

Abstract: An apparatus includes at least one a processor, and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to determine that an input defining a piece of information to be located has been received, determine the location of the information in an environment of the apparatus, and report the location.

Abstract: Method for detecting disappearance of a pattern is used to detect whether a fixed-still pattern in dynamic displayed images disappears. Method includes analyzing a pattern characteristic parameter which represents the fixed-still pattern from each of images continuously displayed in a time sequence, It is checked whether the pattern characteristic parameter fast decreases from at least greater than a high level to at least less than a low level, as a first state transition. Sum of absolute difference (SAD) values for all of the pixels between a previous image and a current image is calculated. It is checked whether the sum of the SAD values fast increases from at least less than a low level to at least greater than a high level, as a second state transition. When the first state transition and the second state transition occur simultaneously, it is determined that the fixed-still pattern disappears in the display.

Abstract: Images are classified as photos (e.g., natural photographs) or graphics (e.g., cartoons, synthetically generated images), such that when searched (online) with a filter, an image database returns images corresponding to the filter criteria (e.g., either photos or graphics will be returned). A set of image statistics pertaining to various visual cues (e.g., color, texture, shape) are identified in classifying the images. These image statistics, combined with pre-tagged image metadata defining an image as either a graphic or a photo, may be used to train a boosting decision tree. The trained boosting decision tree may be used to classify additional images as graphics or photos based on image statistics determined for the additional images.

Abstract: A graphics texture data encoding arrangement in which the texels in a texel block 30 to be encoded are divided into different partitions within the block. A reference partitioning pattern for a texel block to be encoded is generated by using a partitioning function 32 to partition the data values for the texels into a number of data value partitions, and then sorting the individual texels in the texel block into respective partitions 33 based on their values. A set of predefined partitioning patterns 35 that the encoding scheme supports is then compared 36 to the generated reference partitioning pattern. The predefined partitioning pattern that best matches 39 the generated reference partitioning pattern is then used 42 to encode the block of texels.

Abstract: In a color name determination device, a color receiver receives input of a color, a document retriever retrieves for documents in which images relating to the received color are disposed, a key word extractor extracts key word character strings appearing in the retrieved documents, an image retriever retrieves for images relating to the extracted key word character strings, a degree acquirer represents, with a predetermined accuracy, colors appearing in the images as a retrieval result and acquires degrees at which the colors appear in the images, a color determiner determines color indicated by the received character string on the basis of the acquired degrees, and a color naming determiner determines, if the received color and the determined color are the same or similar, the a key word character string corresponding to the determined color is a color name character string indicating the received color.

Abstract: A fingerprint authentication device includes: a fingerprint acquisition section that acquires fingerprint image data; a fingerprint image correction processing section that corrects a pixel value by using a correction coefficient for making a first pixel value of the brightest pixel in a group of pixels at which an integrated value of a number of pixels at a dark portion side in a histogram becomes a predetermined proportion with respect to an integrated value of a number of all pixels, be a brighter second pixel value; a spectral data generation section that generates a spectral data matrix including directions of ridges of a fingerprint and a frequency of the fingerprint; a registered spectral data matrix archive section that archives a registered spectral data matrix; a fingerprint verification section that verifies the spectral data matrix and the registered spectral data matrix; and an authentication results output section that outputs results of authentication.

Abstract: Disclosed are an apparatus and a method for reinspecting a prescription drug, which reinspect the defect state of a defective drug pack after the defect state of an individually packaged drug has been inspected by analyzing the image of the drug. The apparatus includes a drug pack information transceiving module to receive primary inspection data, a drug reinspection controlling module to control a procedure of reinspecting a defect state of a drug pack which has been primarily determined as a defective drug pack, an image processing module to photograph the drug pack, which has been primarily determined as the defective drug pack, and to process an image of the drug pack, and a defect analyzing module to determine a defect state of the drug by analyzing a drug contained in the image of the drug pack provided from the image processing module.

Abstract: The invention provides for a computer-implemented method for detecting one or more archive images matching a search image, each matching archive image being a derivative of the search image or being an original image the search image was derived from accessing a plurality of the archive images. For each of said archive images, a respective archive image histogram may be calculated, wherein each archive image histogram includes a plurality of combination micro-feature values. The archive image histogram may be stored to a database.

Abstract: A method and system for calculating oblong-shape rotation angles from optical image data of arbitrary size for user interface applications is described without the need of eigenvector routines and storage of the image data. The oblong shape may be of arbitrary size and location and need not be precisely elliptical. A few running sums are calculated and stored throughout each scan, and the results are obtained in closed form by simple post-scan computation. An algorithmic embodiment can execute on one or more hardware processors with limited or otherwise constrained computation power, available instruction cycles, available memory, etc. Hardware processors may CPUs found in desktops, laptops, tablets, or handheld computing devices. The resulting arrangement may be used for touch or optical user interfaces and other purposes.

Abstract: Evaluating tissue characteristics including identification of injured tissue or alteration of the ratios of native tissue components such as shifting the amounts of normal myocytes and fibrotic tissue in the heart, identifying increases in the amount of extracellular components or fluid (like edema or extracellular matrix proteins), or detecting infiltration of tumor cells or mediators of inflammation into the tissue of interest in a patient, such as a human being, is provided by obtaining a first image of tissue including a region of interest from a first acquisition, and obtaining a second image of the tissue including the region of interest during a second, subsequent acquisition. The subsequent acquisition may be obtained after a period of time to determine if injury has occurred during that period of time. Such a comparison may include comparison of mean, average characteristics, histogram shape, such as skew and kurtosis, or distribution of intensities within the histogram.

Abstract: An image processing method of picking up an image of a substrate and converting pixel values of the picked-up substrate image makes the pixel values of the picked-up substrate image into a histogram, and creates a tone curve T composed of a periodic function of a predetermined amplitude and a predetermined period based on a distribution of the pixel values in the histogram. The pixel values of the picked-up substrate image are converted using the tone curve T to obtain a substrate image with a high contrast.

Abstract: A system comprising an image display; a digital camera positioned to capture images of persons viewing the image display; and a memory system storing instructions configured to cause a data processing system to implement a method for presenting digital images having a high interest level to a particular person selected from a set of candidate digital images. The method includes using the digital camera to capture an image including a particular person positioned to view the image display. The candidate digital images are analyzed to designate one or more image elements, and familiarity levels of the designated image elements to the particular person are determined. For each candidate digital image, an associated interest level to the particular person is determined responsive to the determined familiarity levels. One or more of the candidate digital images are selected based on the determined interest levels and are presented to the particular person.

Abstract: A method and system characterizes an image of an object. A plurality of interest points are detected within a first image and a local image feature descriptor is built for at least some of the interest points, including mapping information about the interest points according to at least circular distribution information.

Abstract: A feature amount calculation apparatus calculates a feature amount of a target object from image data, and is provided with: a feature value calculator that calculates an edge direction and edge magnitude as input image data pixel-unit feature values; a feature amount calculator that has an edge direction group calculator that calculates a group of edge directions, and a correlation value calculator that takes all pixels or predetermined pixels among a plurality of pixels used in feature value calculation as pixels subject to correlation value calculation and calculates an edge magnitude correlation value between the pixels subject to correlation value calculation for each feature value; and a histogram creator that counts feature amounts in a histogram for each correlation value, and creates a histogram as a feature vector.

Abstract: There are provided an image processing apparatus, a character recognition method and a computer-readable, non-transitory medium that can perform character recognition at high speed while retaining character recognition accuracy. The image processing apparatus includes a histogram generator for generating a histogram based on a pixel value of each pixel in an input image, a component judging unit for judging whether the input image contains a character component and whether the input image contains the character component and a non-character component, a binarization unit for producing a binary image based on edge strength of each pixel, when the input image contains both the character component and the non-character component, and for producing a binary image based on a luminance value of each pixel, when the input image contains the character component but does not contain the non-character component, and a character recognition unit for performing character recognition on the binary image.

Abstract: Distributional information for a set of ? vectors is determined using a sparse basis selection approach to representing an input image or video. In some examples, this distributional information is used for a classification task.

Abstract: The invention provides a method, system, and program product for detecting an object in a digital image. In one embodiment, the invention includes: deriving an initial object indication mask based on pixel-wise differences between a first digital image and a second digital image, at least one of which includes the object; performing an edge finding operation on both the first and second digital images, wherein the edge finding operation includes marking added edges; generating a plurality of straight linear runs of pixels across an image containing the object, wherein each of the plurality of straight linear runs starts and ends on an added edge and is contained within the initial object indication mask; and forming a final object indication mask by retaining only pixels that are part of at least one of the plurality of straight linear runs.