Abstract: Methods, apparatus and articles of manufacture for detecting objects in images using color histograms are disclosed. An example method disclosed herein comprises segmenting an image into sets of subregions, different sets having at least one of different subregion sizes or different subregion offsets relative to one another, determining a first comparison metric quantifying comparison of a first subregion with the object based on differences between bin values of a second color histogram corresponding to the object and adjusted bin values of a first color histogram corresponding to the first subregion, determining a second comparison metric quantifying comparison of a second subregion with the object, and comparing the first and second comparison metrics with a threshold to identify at least one of the subregions as being associated with a possible location of the object in the image.

Abstract: Systems, methods and computer readable media for exposure quality detection are described. In some implementations, a method can include computing an overall image exposure score for an image. The method can also include determining one or more face regions in the image. The method can further include computing a face region exposure score for each face region. The method can also include combining the overall image exposure score and each face region exposure score to generate an exposure quality score for the image.

Abstract: Techniques for identifying documents sharing common underlying structures in a large collection of documents and processing the documents using the identified structures are disclosed. Images of the document collection are processed to detect occurrences of a predetermined set of image features that are common or similar among forms. The images are then indexed in an image index based on the detected image features. A graph of nodes is built. Nodes in the graph represent images and are connected to nodes representing similar document images by edges. Documents sharing common underlying structures are identified by gathering strongly inter-connected nodes in the graph. The identified documents are processed based at least in part on the resulting clusters.

Abstract: An image processing apparatus including an acquisition unit, an extraction unit, a determination unit and an addition unit is provided. The acquisition unit acquires an image of one page from images of a plurality of pages included in one file. The extraction unit extracts color information, which is information about a color of the image of the one page acquired by the acquisition unit. The determination unit determines whether the color information extracted by the extraction unit satisfies a predetermined condition for specifying a page partition. The addition unit adds additional information, by which the page partition can be searched, to at least one of the images of the plurality of pages, when the determination unit determines that the color information satisfies the predetermined condition.

Abstract: There is provided an information processing apparatus including a storage unit in which a color arrangement database recording correspondences between color arrangement information about a combination of colors and a color arrangement mood concerning an atmosphere provided by the combination of colors is stored and a color arrangement mood analysis unit that analyzes the input color arrangement information based on the color arrangement database to determine the color arrangement mood corresponding to the combination of colors represented by the input color arrangement information.

Abstract: A method for toning mapping high dynamic range (HDR) video for display in low dynamic range display comprises accessing the HDR video; producing a luminance information for the individual frames; partitioning different consecutive groups of frames into segments responsive to the luminance information; classifying the segments into static luminance segments and transient luminance segments; producing a tone mapping parameter set for the static luminance segments; producing a tone mapping parameter set for at least one of the transient luminance segments responsive to the tone mapping parameter set of an adjacent static luminance segment; and tone mapping the static luminance segments and transient luminance segments according to the respective tone mapping parameter sets.

Abstract: Processes such as image matching can be improved by utilizing a Polar Accumulated Signed Gradient (PASG), a feature descriptor vector that utilizes polar coordinates. Such an approach is efficient because it avoids the costly image rotation and resampling of conventional processes. Further, the gradients can be sampled over different regions in order to achieve effective weighting without analyzing all gradients and applying a weighting factor. Such efficiencies enable image matching to be performed on devices with limited computing capacity, such as smart phones and tablet computers.

Abstract: An exemplary embodiment of the present invention includes a method of registering images. The method includes: for each image, determining an optimum intensity threshold set from a plurality of intensity threshold sets that maximizes a variance between classes of each set, segmenting each image using the corresponding determined optimum intensity threshold set, generating mutual information from a joint histogram of at least two of the segmented images, and registering the at least two images using the mutual information. The joint histogram may be generated using a geometry shader of a graphical processing unit.

Abstract: Aspects of the present invention comprise generating and using Multi-Order Contextual co-Occurrence (MOCO) descriptors to implicitly model the high level context using detection responses from a baseline object detector. In embodiments, a 1st-order context feature is computed as a set of randomized binary comparisons on a response map of the baseline object detector. The statistics of the 1st-order binary context features are further calculated to exercise construct a higher-order co-occurrence descriptor, which, in embodiments, may be combined with other features such as the 0th-order context features and/or the 1st-order features to form the MOCO. In embodiments, combining the MOCO feature with the original image feature, the baseline object detector may be evolved to a stronger context aware detector.

Abstract: Systems and methods for displaying digitized waveform data on a pixilated screen are provided. Certain methods include retrieving waveform data that exists as a plurality of sequential samples in a buffer. A first sample is associated with a first point of a pixilated screen. A second sample that follows the first sample is associated with a second point of the screen. A boundary of a rectangular line segment connecting the points is determined. For each pixel of the pixilated screen that has any portion inside the boundary of the rectangular line segment, a number of subpixels within the pixel that have any portion inside the boundary of the rectangular line segment is determined and that number of subpixels is correlated with an intensity value that is then associated with the pixel. Waveform data is displayed by painting pixels of the screen with the associated intensity values.

Abstract: A signal processor which performs gradation conversion processing using a histogram of an image signal is provided with a feature calculating unit which calculates a feature of the image signal; an interval setting control unit which sets intervals of a pixel value in the histogram when a histogram of the image signal is obtained, according to the calculated feature, so as to obtain the histogram of the local region of the image signal; a gradation-conversion-characteristic calculating unit which calculates a gradation conversion characteristic, using the histogram obtained by using the intervals set at the interval setting control unit; and a gradation converting unit which performs gradation conversion to the image signal using the gradation conversion characteristic.

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: Objects in an image are classified by applying an appearance classifier to the image to determine candidates of the objects and statistics associated with the candidates, wherein the appearance classifier uses a set of windows, and the candidates are in selected windows. Then, a context classifier is applied only to the selected windows of the image to determine an identity, and location of objects in the image.

Abstract: A multi-class discriminating device for judging to which class a feature represented by data falls. The device has a first unit for generating plural first hierarchical discriminating devices for discriminating one from N, and a second unit for combining score values output respectively from the plural first hierarchical discriminating devices to generate a second hierarchical feature vector and for entering the second hierarchical feature vector to generate plural second hierarchical discriminating devices for discriminating one from N. When data is entered, the plural first hierarchical discriminating devices output score values, and these score values are combined together to generate the second hierarchical feature vector. When the second hierarchical feature vector is entered, the second hierarchical discriminating device which outputs the maximum score value is selected.

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: The present invention provide an automatic, frame-by-frame, rapid, and accurate search method for video clip. It comprises the following steps. First, acquire and store a plurality of videos to a storage unit. Each video comprises a plurality of frames, respectively. Then, a processing unit extracts the characteristics of the plurality of frames and gives a plurality of characteristic values. Next, the processing unit calculates a plurality of difference values between the characteristic value of each frame and that of the previous frame and between the characteristic value of the subsequent frame and that of the current frame. Store a data set, which is produced by corresponding the plurality of difference values between all frames of each video to the time axis, to the storage unit. Finally, the processing unit compares a data set of any video with a plurality of data sets of other videos and gives a search result.

Abstract: A color-based imaging system and method for the detection and classification of insects and other arthropods are described, including devices for counting arthropods and providing taxonomic capabilities useful for pest-management. Some embodiments include an image sensor (for example, a digital color camera, scanner or a video camera) with optional illumination that communicates with a computer system. Some embodiments include a color scanner connected to a computer. Sampled arthropods are put on a scanner to be counted and identified. The computer captures images from the scanner, adjusts scanner settings, and processes the acquired images to detect and identify the arthropods. Other embodiments include a trapping device and a digital camera connected by cable or wireless communications to the computer. Some devices include a processor to do the detection and identification in the field, or the field system can send the images to a centralized host computer for detection and identification.

Abstract: An image capture system comprises an image input and processing unit. The image input obtains image information which is then passed to the processing unit. The processing unit is coupled to the image input for determining image metrics on the image information. The processing unit initiates a capture sequence when the image metrics meet a predetermined condition. The capture sequence may store one or more images, or it may indicate that one or more images have been detected. In one embodiment, the image input is a CMOS or CCD sensor.

Abstract: Three-dimensional (3D) glasses, a method for driving 3D glasses, and a system for providing a 3D image are provided. The 3D glasses driving method includes detecting a first pattern corresponding to a left eye image and a second pattern corresponding to a right eye image from a 3D display apparatus, converting the first pattern into a first histogram and the second pattern into a second histogram, comparing the converted first and second histograms with first and second optimal histograms, and adjusting an opening or a closing of a left eye glasses part and a right eye glasses part of the 3D glasses based on the comparison.

Abstract: A tumor is tracked in multiple sequences of images acquired concurrently from different viewpoints. Features are extracted in each set of current images using a window. A regression function, subject to motion constraints, is applied to the features to obtain 3D motion parameters, which are applied to the tumor as observed in the images to obtain a 3D location of the object. Then, the shape of the 3D object at the 3D location is projected onto each image to update the location of the window for the next set of images to be processed.

Abstract: Systems and related methods for detecting a scene in video content are provided. In one example, a scene detection program identifies a plurality of shots in the video content and selects a target shot. The program builds a forward window including the target shot and having shots that are temporally ahead of the target shot, and a rearward window having shots that are temporally behind the target shot. For each of the shots in the forward window the program determines a dissimilarity between a selected shot in forward window and each of the other shots in the rearward window. If one or more of the dissimilarities is less than a scene boundary threshold, the program determines that the scene does not begin at the target shot. If none of the dissimilarities is less than the scene boundary threshold, the program determines that the scene begins at the target shot.

Abstract: An image manipulation process is controlled by a spatial periodicity measure formed for an image or an image block by measuring the sparseness of the two-dimensional spatial frequency spectrum of the image on a scale of zero to unity in which a two-dimensional spatial frequency spectrum having all equal values has a sparseness of zero and in which a two-dimensional spatial frequency spectrum having only one non-zero value has a sparseness of unity. Sparseness may be measured by allocating values of the spectrum to frequency bins and counting the number of bins that contain non-zero values; comparing values of the spectrum with a threshold and counting the number of values that exceed the threshold; or forming a function of the mean-square value and the mean value of the spectrum values.

Abstract: Systems and methods are provided for creating contour images that represent the contour of objects reflected in images, calculating contour histogram descriptors of the contour images, and classifying images based in part on the histogram descriptors of the contour images. For example, a contour image of an image is created. A radial-polar grid having a plurality of radial-polar bins is then positioned on the contour image. A contour histogram descriptor is created to include a number of bins that correspond to the radial-polar bins of the radial-polar grid, where the contents of the bins of the contour histogram descriptor represent the number of pixels of the contour image that are located in the corresponding radial-polar bins of the radial-polar grid. Images are classified at least based in part on comparisons between contour histogram descriptors of the images and contour histogram descriptors of training images.

Abstract: A method and system for vehicular clear path detection using adaptive machine learning techniques including additional classifiers. Digital camera images are segmented into patches, from which characteristic features are extracted representing attributes such as color and texture. The patch features are analyzed by a Support Vector Machine (SVM) or other machine learning classifier, which has been previously trained to recognize clear path image regions. For image regions or patches which result in a low confidence value, an additional classifier can be used, where the additional classifier is adaptively trained using real world test samples which were previously classified with high confidence as clear path roadway. Output from the original, offline trained classifier and the additional, adaptively-updated classifier are then used to make a joint decision about clear path existence in subsequent image patches.

Abstract: An apparatus and method are provided for performing detail enhancement. The apparatus includes an edge detector that detects an edge pixel from contents based on a luminance component of the contents, a histogram analyzer that generates at least one histogram with respect to the edge pixel and determines a gain variable value of the edge pixel based on the at least one histogram, and a gain regulator that determines a gain corresponding to the gain variable value of the edge pixel.

Abstract: A system for image processing includes determining edge pixels of a model image using an edge based technique and determining an angular orientation for each the edge pixels of the model image. The system determines a histogram of gradient orientations based upon the angular orientations of the model image. The system determining edge pixels of an input image using an edge based technique and determining an angular orientation for each of the edge pixels of the input image. The system determines a histogram of gradient orientations based upon the angular orientations of the input image. The system compares the histogram of gradient orientations of the model image with the histogram of gradient orientations of the input image to determine candidate locations of an object within the input image.

Abstract: An image generation apparatus includes: an extracting section to extract characteristic information of a hair region in a face image; an image specifying section to specify a hairstyle image on the basis of the characteristic information extracted by the extracting section; and a first generating section to generate a portrait image of a face in the face image by using the hairstyle image specified by the image specifying section.

Abstract: Devices and techniques are described for automatically calibrating a projector system. The projector system undergoing calibration is coupled to a computing device and an automated positioning platform coupled to a target structure. The computing device acquires images using a calibrated camera of one or more patterns projected by the projector onto the target structure when the target structure is in a plurality of repeatable poses. From these acquired images, intrinsic projector parameters may be determined. Once determined, the parameters may be used to correct images produced by the projector or acquired by the camera which include a projected pattern or portion thereof.

Abstract: A method and system implementing the method for color pixel counting in images by determining color pixels that, upon marking, appear as visibly color or gray to the human eye by applying a 2-dimensional filter and a reverse color lookup process. An area color computation is performed on a window around each pixel of interest to determine average gray level values for output colors (C, M, Y). Gray line values for each output color are determined and the values for at least two output colors are compared to the average gray level values. If average gray level value for both colors is within a tolerance for the gray line, the pixel is determined as visibly gray. Otherwise, if at least one count is not within the tolerance, the pixel is determined as visibly color. The count of visibly color pixels can be used to determine the billable color pixel count.

Abstract: Color management which converts source device colors into counterpart colors in a destination device color space. Device independent source and destination device gamut boundaries are obtained. A color image is generated in the source device color space. The color image is transformed into an image in a device independent color space. A gamut boundary for the device independent image is generated, based on the content of the device independent image. The source device gamut boundary is shrunk in a hue symmetric manner, such that color hues do not change, until it touches the image gamut boundary. Colors of the device independent image are mapped onto a gamut of the destination device by invoking a compression-type gamut mapping algorithm that uses the modified source device gamut boundary and the destination device gamut boundary to perform gamut mapping. The gamut mapped colors are converted into colors in destination device dependent color space.

Abstract: An apparatus for analyzing a video capture screen includes: a video frame extracting unit extracting at least one frame from a video having a plurality of frames; an extracted frame digitizing unit digitizing features of each of the at least one frame extracted by the video frame extracting unit; an image digitizing unit digitizing features of at least one collected search target image; an image comparing and searching unit comparing the search target image with the at least one frame extracted from the plurality of frames by digitized values of the collected search target image and the at least one frame; and a search result processing unit mapping related information of the collected search target image to a frame coinciding with the search target image and storing the related information in a database, when the extracted at least one frame coinciding with the search target image is present in a comparison result.

Abstract: A congestion estimating apparatus includes an area dividing unit that divides a moving image into partial areas. A movement information determining unit determines whether there is movement, and a person information determining unit determines whether there is a person, in each partial area. A staying determining unit determines a state for each partial area. The staying determining unit determines the state as a movement area in which there is a movement of person when there is movement and there is a person; and determines the state as a noise area when there is movement and there is no person; and determines the state as a staying area in which there is a person who is staying when there is no movement and there is a person; and determines the state as a background area in which there is no person when there is no movement and there no person.

Abstract: An image determining method, for determining which pixels in an image are specific image pixels of a specific image, comprising: (a) determining which pixels in the image have brightness values larger than a threshold value; (b) determining the pixels having brightness values larger than the threshold value as the specific image pixels; and determining pixels in a predetermined range of at least one the specific image pixel as the specific image pixels as well.

Abstract: Methods and apparatus for detection and identification of duplicate or near-duplicate videos using a perceptual video signature are disclosed. The disclosed apparatus and methods (i) extract perceptual video features, (ii) identify unique and distinguishing perceptual features to generate a perceptual video signature, (iii) compute a perceptual video similarity measure based on the video edit distance, and (iv) search and detect duplicate and near-duplicate videos. A complete framework to detect unauthorized copying of videos on the Internet using the disclosed perceptual video signature is disclosed.

Abstract: A pedestrian detector includes an image-taking mechanism and an image analyzer. The image-taking mechanism is provided on a vehicle and used to take an image in front of the vehicle. The image analyzer is connected to the image-taking mechanism and used to receive the image from the image-taking mechanism and analyze the image to determine whether the image conforms to a pedestrian to determine whether there is a pedestrian in front of the vehicle.

Abstract: A method of clustering and reducing hyperspectral image data having a plurality of spatial pixels, and a plurality of spectral dimensions associated with each spatial pixel, includes computing an initial basis vector associated with the hyperspectral image data, unmixing the initial basis vector with the hyperspectral image data to generate an initial set of coefficients and an associated set of residual vectors, generating a set of clusters based on the initial set of coefficients, and iteratively computing one or more additional basis vectors and updating the set of clusters. The iterative computing includes calculating a subsequent basis vector based on a residual vector associated with a prior unmixing, unmixing the subsequent basis vector with a prior set of residual vectors to generate additional coefficients associated with each pixel, and iteratively computing cluster centers and content including an additional dimension associated with the subsequent basis vector.

Abstract: Systems and methods may automatically evaluate printed line weights in an image composition. An image composition may be received and the number of horizontal and vertical rows of pixels may be determined. The number of positive pixels in each of the horizontal and vertical rows may then be determined. An actual size of a pixel may be calculated. The actual size of a pixel may be multiplied by the number of positive pixels in each of the horizontal rows and each of the vertical rows to determine a positive line weight of each of the horizontal rows and each of the vertical rows. A predetermined minimum permissible positive line weight for the image composition may be identified and compared to the positive line weight of each of the horizontal rows and each of the vertical rows.

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

Abstract: A system provides product annotation in a video to one or more users. The system receives a video from a user, where the video includes multiple video frames. The system extracts multiple key frames from the video and generates a visual representation of the key frame. The system compares the visual representation of the key frame with a plurality of product visual signatures, where each visual signature identifies a product. Based on the comparison of the visual representation of the key frame and a product visual signature, the system determines whether the key frame contains the product identified by the visual signature of the product. To generate the plurality of product visual signatures, the system collects multiple training images comprising multiple of expert product images obtained from an expert product repository, each of which is associated with multiple product images obtained from multiple web resources.

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: A method and system for calculating oblong-shape rotation angles from image data of arbitrary size using running sums 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, real-time image recognition, real-time machine vision, and other purposes.

Abstract: Techniques are disclosed involving contrast adjustment for images. For example, an input image is classified based on its pixel value characteristics, as expressed in an input brightness histogram. From such a classification, a target histogram distribution for a corresponding output image (i.e., a contrast-adjusted transformation of the input image) may be specified. With the target histogram of the output image specified, a transformation function may be derived that maps input image values to output image values. Moreover, transitions of such transformation functions may be smoothed. Such smoothing may provide advantages, such as a reduction in flickering associated with video data.

Abstract: Systems and methods receive, from a building, an inventory identifying a plurality of electronic devices associated with a building automation system. The systems and methods compare the inventory to an additional inventory associated with an additional building automation system of an additional building to determine that the inventory is the same as or similar to the additional inventory, wherein the additional inventory identifies a plurality of additional electronic devices. The systems and methods further identify an automation scene that coordinates operation of at least a portion of the plurality of additional electronic devices and provide the automation scene to the building for implementation in the building automation system.

Abstract: A method and system for generating a feature descriptor for robust facial expression recognition pre-processes a facial image using a Gaussian filter to smooth the facial image. Then, gradient based images at M scales and N orientations are generated from the pre-processed facial image. Further, a portion of an image corresponding to each action unit is selected from each of the gradient based images. Thereafter, appearance of at least one facial event in the selected portion of the image is captured. Also, a geometry of the at least one facial event in the selected portion of image is determined to obtain a feature descriptor for each action unit for robust facial expression recognition.

Abstract: An image processing apparatus identifies tissues in respective parts of a tissue image. A tissue image subdivider subdivides a tissue image for identification into local regions. A detector detects texture feature values of the local regions. A determining unit compares the detected texture feature value of a local region to a learned feature value for identification associated with a predetermined tissue, and on the basis of the comparison result, determines whether or not the local region belongs to the predetermined tissue.

Abstract: A system described herein includes a receiver component that receives a first image and a symmetry signature generator component that generates a first global symmetry signature for the image, wherein the global symmetry signature is representative of symmetry existent in the first image. The system also includes a comparer component that compares the first global symmetry signature with a second global symmetry signature that corresponds to a second image, wherein the second global symmetry signature is representative of symmetry existent in the second image. The system additionally includes an output component that outputs an indication of similarity between the first image and the second image based at least in part upon the comparison undertaken by the comparer component.

Abstract: Disclosed are methods and apparatus for automatic optoelectronic detection and inspection of objects, based on capturing digital images of a two-dimensional field of view in which an object to be detected or inspected may be located, analyzing the images, and making and reporting decisions on the status of the object. Decisions are based on evidence obtained from a plurality of images for which the object is located in the field of view, generally corresponding to a plurality of viewing perspectives. Evidence that an object is located in the field of view is used for detection, and evidence that the object satisfies appropriate inspection criteria is used for inspection. Methods and apparatus are disclosed for capturing and analyzing images at high speed so that multiple viewing perspectives can be obtained for objects in continuous motion.

Abstract: A method of detecting a predefined set of characteristic points of a face from an image of the face includes a step of making the shape and/or the texture of a hierarchy of statistical models of face parts converge over real data supplied by the image of the face.

Abstract: A system and method to classify forms. An image representing a form of an unknown document type is received. The image includes line-art. Further, a plurality of template models corresponding to a plurality of different document types is received. The plurality of different document types is intended to include the correct document type of the unknown document. A subset of the plurality of template models are selected as candidate template models. The candidate template models include line-art junctions best matching line-art junctions of the received image. One of the candidate template models is selected as a best candidate template model. The best candidate template model includes horizontal and vertical lines best matching horizontal and vertical lines of the received image, respectively, aligned to the best candidate template model.

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.