Abstract: Techniques for comparing documents may be provided. For example, a comparison between layouts of the documents may be performed. The comparison may include segmenting the documents into blocks, where an arrangement of blocks of a document represents a layout of the document. Once segmented, similarity metrics, such as distances, between blocks of one document and blocks of the other document may be computed. The similarity metrics may be used to match the blocks between the documents. Further, the similarity metrics between the matched blocks may be added to determine an overall similarity metric between the documents. This overall similarity metric may indicate how similar the documents may be.

Abstract: Detecting text using stroke width based text detection. As a part of the text detection, a representation of an image is generated that includes pixels that are associated with the stroke widths of components of the image. Connected components of the image are identified by filtering out portions of the pixels using metrics related to stroke width. Text is detected in the image based on the identified connected components.

Abstract: An image processing apparatus comprises an image acquiring unit that acquires images from the image managing server; an image analyzing unit that analyzes the acquired images to determine evaluation values thereof; a grouping unit that groups the acquired images into groups based on collateral information of the acquired images; a group selecting unit that selects groups including images having operation information as the collateral information out of the plurality of groups; an image selecting unit that selects images out of the images included in the selected groups, based on the evaluation values and the operation information; and an image arranging unit that arranges the selected images in a predetermined layout to create the synthetic image.

Abstract: Processing of a signal includes identifying a past recurring pattern in the signal with a recurrence matrix, and filtering the signal and the recurring pattern such that the recurring pattern serves as a representation of future signal behavior.

Abstract: A process for extracting iris data for biometric identification includes a thresholding method where the thresholds are selected according to a nonparametric approach that considers the grey scale and does not require classifying pixels as edge or non-edge pixels. An eye image is first acquired, where the eye image has component images including an iris image with an inner boundary and an outer boundary. The eye image has a distribution of grey levels. Component images, such as an iris image or a pupil image, from the eye image are segmented according to the distribution of grey levels. The inner boundary and outer boundary of the iris image are determined from the component images. The iris image within the inner boundary and outer boundary is processed for biometric identification. The component images may be segmented by creating an eye histogram of pixel intensities from the distribution of grey levels.

Abstract: Novel tools and techniques for generating survey data about a survey site. Aerial photography of at least part of the survey site can be analyzed using photogrammetric techniques. In some cases, an unmanned aerial system can be used to collect site imagery. The use of a UAS can reduce the fiscal and chronological cost of a survey, compared to the use of other types aerial imagery and/or conventional terrestrial surveying techniques used alone.

Abstract: A candidate output element configured to output recognition target commodities as candidates of a recognized commodity in a descending order of the similarity degrees calculated by the similarity degree calculation element, a distance measurement element configured to measure the distance from the image capturing section to a commodity photographed by the image capturing section, and a changing element configured to change the number of candidates of a recognized commodity output by the candidate output element according to the distance measured by the distance measurement element.

Abstract: Enables intelligent synchronization and transfer of generally concise event videos synchronized with motion data from motion capture sensor(s) coupled with a user or piece of equipment. Greatly saves storage and increases upload speed by uploading event videos and avoiding upload of non-pertinent portions of large videos. Provides intelligent selection of multiple videos from multiple cameras covering an event at a given time, for example selecting one with least shake. Enables near real-time alteration of camera parameters during an event determined by the motion capture sensor, and alteration of playback parameters and special effects for synchronized event videos. Creates highlight reels filtered by metrics and can sort by metric. Integrates with multiple sensors to save event data even if other sensors do not detect the event. Also enables analysis or comparison of movement associated with the same user, other user, historical user or group of users.

Abstract: A system and method detect objects in a digital image. At least positional data associated with a vehicle is received. Geographical information associated with the positional data is received. A probability of detecting a target object within a corresponding geographic area associated with the vehicle is determined based on the geographical data. The probability is compared to a given threshold. An object detection process is at least one of activated and maintained in an activated state in response to an object detection process in response to the probability being one of above and equal to the given threshold. The object detection process detects target objects within at least one image representing at least one frame of a video sequence of an external environment. The object detection process is at least one of deactivated and maintained in a deactivated state in response to the probability being below the given threshold.

Abstract: A driving assistance device is provided with a turning state detection unit, an imaging unit, a solid object detection unit and a detection region modification unit. When the turning state detection unit detects that a host vehicle is in a turning state, the detection region modification unit alters a position of a detection region with respect to the host vehicle, or alters a shape or an area of the detection region based on the turning state of the host vehicle. For example, the detection region modification unit sets a shorter region length of the detection region as the turning radius of the host vehicle becomes smaller. Hereby, the region closest to the host vehicle is set, to a limited extent, as the detection regions.

Abstract: Providing access to digitally published data includes creating a note having at least a portion that is handwritten by a first user, converting handwriting of the note into a content access identifier that varies according to the portion that is handwritten by the first user, associating the content access identifier with the digitally published data, and making the digitally published data available to a second user by making the note available to the second user. The digitally published data may be written to a public database and/or a private database. A portion of the note may be pre-printed. A pre-printed distinguishing pattern on the note may indicate that handwritten content corresponds to a content access identifier. The pre-printed portion may be a regular dotted pattern. The note may have a known identifiable color and size.

Abstract: A parallel object detection method for heterogeneous microarchitectures. The method is designed for increasing the throughput of object detection in a computer system that is equipped with an array of cores including a shared memory, a constant memory, and functional units. Latency reduction is achieved through a multilevel parallelization method that exploits fine-grain data-level parallelism using multithreaded SIMD computations, and coarse-grain parallelism by relying on concurrent kernel execution.

Abstract: An image processing device is provided, the image processing device comprising: an image input unit configured to be input with a frame image of an imaging area imaged by a camera; an image processing unit configured to process the frame image input to the image input unit, and detect an object imaged in the frame image; and an operation frequency determination unit configured to determine a frequency of an operation clock of the image processing unit according to the number of objects detected by the image processing unit, wherein the operation frequency determination unit lowers the frequency of the operation clock of the image processing unit as the number of objects detected by the image processing unit becomes smaller.

Abstract: Embodiments herein provide computer-implemented techniques for allowing a user computing device to extract financial card information using optical character recognition (“OCR”). Extracting financial card information may be improved by applying various classifiers and other transformations to the image data. For example, applying a linear classifier to the image to determine digit locations before applying the OCR algorithm allows the user computing device to use less processing capacity to extract accurate card data. The OCR application may train a classifier to use the wear patterns of a card to improve OCR algorithm performance. The OCR application may apply a linear classifier and then a nonlinear classifier to improve the performance and the accuracy of the OCR algorithm. The OCR application uses the known digit patterns used by typical credit and debit cards to improve the accuracy of the OCR algorithm.

Abstract: Notebook comprising a plurality of pages (1) of paper bound together with a cover (2), in which a plurality of substantially vertical and/or substantially horizontal lines are printed on the pages (1) and are formed by a plurality of dots (5) aligned with each other, which have a maximum dimension, in particular diameter, comprised between 0.21 and 0.35 mm, the distance between two adjacent dots (5) of a same line being comprised between 0.43 and 0.7 mm and the color of the dots (5) being darker than the color of the page (1), in which the sum of the four quadrichrome CMYK values of the color of the page (1) is comprised between 0 and 20, in particular comprised between 10 and 20, with the K value less than 10, and the sum of the four CMYK values of the color of the dots (5) is comprised between 15 and 50, in particular between 25 and 40, with the value K less of 40. The present invention also relates to a method for digitizing notes by means of said notebook.

Abstract: An image processing device performs: acquiring image data representing an image, the image data including a plurality of sets of pixel data each having a multiple-level gradation value, each gradation value falling within a predetermined range between and inclusive of a minimum gradation value and a maximum gradation value, a predetermined number of levels of gradation value falling within the predetermined range; setting a partial range of levels of gradation value, a first number of levels of gradation value in the partial range being smaller than the predetermined number; and determining that the image is a photograph image when the image meets a first criterion that a degree of change in distribution of gradation values is smaller than a reference value, the distribution of gradation value indicating a pixel number of sets of pixel data for each level of the gradation value falling within the partial range.

Abstract: Methods and apparatus are provided for determining quantization parameter predictors from a plurality of neighboring quantization parameters. An apparatus includes an encoder for encoding image data for at least a portion of a picture using a quantization parameter predictor for a current quantization parameter to be applied to the image data. The quantization parameter predictor is determined using multiple quantization parameters from previously coded neighboring portions. A difference between the current quantization parameter and the quantization parameter predictor is encoded for signaling to a corresponding decoder.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: A person detection apparatus determines a weather condition such as rain and solar radiation based on a variety of information from a weather information input portion. Then, based on a determination result of the weather condition, the umbrella ratio showing the ratio of persons with umbrellas is calculated. The person detection apparatus uses a no-umbrella recognition model describing a person with no umbrella and an umbrella-hold recognition model describing a person with umbrella in order to perform pattern recognition to an input image to derive recognition scores based on the respective recognition models. Then, the umbrella ratio depending on the weather condition is used to correct the respective recognition scores based on the pattern recognition using the no-umbrella model and the recognition score based on the pattern recognition using the umbrella-hold model; the corrected recognition scores are output as a final detection result.

Abstract: A code conversion device for image information for generating an image code which is unique to the image information from the image information, the code conversion device for image information includes a processor and a memory, wherein the memory contains instructions for causing the processor to perform operations of: converting acquired raw image information into a plurality of pieces of developed image information; extracting each piece of feature information, from each of the plurality of pieces of developed image information by performing a self-organization processing using a probability scale on each of the plurality of pieces of developed image information; and quantifying a plurality of pieces of feature information and generating an image code.

Abstract: The knowledge that “the color of a vehicle body or guard rail is uniform” applies to the detection of a human, improving the detection performance of the human of which a part of the body is hidden. That is, it is determined whether or not a human candidate area specified based on the ordinary human recognition model has a specific part, e.g., an area corresponding to a lower body, exhibiting the high degree of color uniformity. When affirmed, the human candidate area is understood to be “a human of which a part of the body is hidden by a hood or a trunk” and recognized as a human, same as in the case where the body is not hidden.

Abstract: An apparatus and a method for recognizing a character based on an input image is provided. The apparatus includes an input unit configured to receive the input image and a controller configured to select, from the input image, a region of image analysis to be used for image analysis, and to analyze the selected region of image analysis to determine a type of the input image, to apply, to the input image, an image effect for distinguishing a character region and a background region in the input image if the type of the input image indicates that the input image is obtained by photographing a display screen, to binarize output of the image effect according to the determined type of the input image, and to recognize a character from the binarized output of the image effect.

Abstract: To improve feature selection accuracy during a visual search, interest points within a query image are two-way matched to features in an affine transformed image or otherwise transformed version of the query image. A user device implements a method for selecting local descriptors in the visual search. The method includes: detecting a first set of interest points for the original image; computing an affine transform matrix; computing a new image as a transformation of the original image using the affine transform matrix; detecting a second set of interest points from the and new image; performing a two-way matching between the first set of interest points and the second set of interest points; sorting matching pairs according to a specified self-matching score (SMS); assigning an infinite value to SMS of unmatched interest points from the original image; selecting the interest points based on SMS. Significant performance gains reduce false positive matches.

Abstract: An apparatus for locating a landmark in a set of image data comprises a landmark location unit that is configured, for each of a plurality of image data items, to obtain from a first two-class classifier a first classification of the image data item as foreground or background, to obtain from a second two-class classifier a second classification of the image data item as foreground or background, and to combine the first classification and the second classification to obtain a combined classification, and wherein the landmark location unit is further configured to use the combined classifications for the plurality of image data items to determine a location for the landmark.

Abstract: A system and method for searching images and identifying images with similar facial features is disclosed. In one implementation, the system includes a pre-processing module, a feature extraction module, a model creation module, a similarity identifier module and a results display module. The pre-processing module receives a facial image, determines key-points associated with a facial feature and identifies a facial area including the facial feature. The feature extraction module extracts the key-points. The model creation module creates a similarity model for determining similar facial features at least in part by comparing the facial feature from a plurality of images. The similarity identifier module applies the similarity model to the facial feature an image in relation to the facial feature in other images and determines which other image has a most similar facial feature. The results display module presents a result based at least in part on the determination.

Abstract: A particular method includes detecting an interaction event using an event capture object of a rendered display of a graphics file. The graphics file is rendered to generate the rendered display by layering one or more foreground objects over one or more background objects. The method also includes executing code associated with the graphics file in response to detecting the interaction event. The code is executed to determine an identifier of a highlight object based on an identifier of the event capture object. The highlight object is below the event capture object in the rendered display and may be below the object to be highlighted in the rendered display. The code is also executed to change an attribute of the highlight object to modify the rendered display.

Abstract: A media processing device enables continuous high-speed processing of recording media. A media processing device has a magnetic head 21 that reads information recorded on checks 4 in the check 4 conveyance path 10, a magnetic recognition unit 20b that acquires the result of a verification based on information read by the magnetic head 21, and a conveyance control unit 20a that starts conveying the check 4 to be processed next. The conveyance control unit 20a selectively operates in an individual processing mode that starts conveying the check 4 to be read next after the results of verifying the read data are acquired by the magnetic recognition unit 20b, and a continuous processing mode that starts conveying the check 4 to be read next after the magnetic head 21 reads at least part of the information without waiting for the magnetic head 21 to acquire the complete verification result.

Abstract: A method for controlling the metering of ink in a printing press uses a control unit for comparing measured color values established on a print control strip to target color values and for controlling the metering of ink in the inking units of a printing press. The method is distinguished by the fact that the halftone of the measurement fields used for establishing the measured color values in the print control strip is independent of the halftone of the printed image. A printing substrate, a printing plate and a printing press having a device for creating halftone image data are also provided.

Abstract: Provided is a printing apparatus having a plurality of sheet feeding units, and method of controlling the printing apparatus. When a print job including paper feed source information is received, paper information set for the sheet feeding unit designated by the paper feed source information is obtained, and stored together with data of this print job. Also, when execution of the stored print job is instructed, a sheet feeding unit that accommodates a sheet having the paper information is selected, and control is performed such that a sheet is fed from the selected sheet feeding unit, and the stored print job is executed on the fed sheet.

Abstract: The present invention aims to temporarily change print configuration information easily. To achieve such an object, an information processing apparatus of the present invention comprises: a monitoring unit configured to monitor that job data composed of print configuration information data and content data, and a difference file for temporarily changing the print configuration information of the print configuration information data to different print configuration information are put in a hot folder; and a correcting unit configured to, in a case where the job data and the difference file are put in the hot folder, correct the print configuration information data included in the job data on the basis of the difference file.

Abstract: Disclosed is an image forming apparatus including a command receiving unit that receives a print command, a detector that detects a state of the operation unit, and a controller. When the detector detects that the state of the operation unit is a predetermined state subsequent to receiving the print command by the command receiving unit and prior to completion of printing based on the print command, the controller causes the printing to be terminated. When the detector detects that the state of the operation unit is not the predetermined state subsequent to receiving the print command by the command receiving unit and prior to the completion of the printing based on the print command, the controller causes the printing to be executed.

Abstract: An image processing apparatus includes a detector, a memory, and a converter. The detector detects a ruling and an angle of a screen from a binary image of each plate, the binary image being obtained as a result of screen processing. The memory stores a detection result for a plate for which a ruling and an angle have been detected by the detector. The converter converts an image obtained as a result of screen processing into a multi-level image by using, for a plate for which a detection result has been stored in the memory, a ruling and an angle in the detection result, or by using, for a plate other than the plate for which a detection result has been stored in the memory, a ruling and an angle obtained as a result of detection performed by the detector.

Abstract: A recording material processing device, includes: an accumulation unit that is accumulated with a plurality of recording materials each of which is a recording material formed with an image by an image forming unit and inspected by an inspection unit, where the image forming unit that forms the image on the recording material based on original image data; and a processing unit that removes the recording material which was determined to have a defect which occurred in the image by the inspecting unit, among the plurality of recording materials and adds a new recording material on which a same image is formed by the image forming unit, to the plurality of recording materials, the same image being based on the original image data of the image formed on the defective recording material.

Abstract: A light-blocking object processing section produces first and second light-blocking objects to be formed on the front and back surfaces of a print medium, respectively. A first image object is formed on the first light-blocking object and a second image object is formed on the second light-blocking object. If the first and second image objects should be formed in first and second areas, respectively, and the first and second areas defining an overlapping area in which the first image object overlaps the second image object, then the light-blocking object processing section performs one of a first process in which the thickness of the first and second light-blocking objects is smaller in the overlapping area than in a third area outside the overlapping area and a second process in which one of the first and second light-blocking objects is not formed.

Abstract: High speed reading of fiscal information from memory is enabled while complying with financial regulations. Including fiscal memory 10 that stores fiscal information, a first control unit 5 that controls communication with the host computer 2 and operates at a first processing speed, and a second control unit 7 that controls operation of the fiscal memory 10 and operates at a second processing speed that is lower than the first processing speed, the second control unit 7 performs a write process writing fiscal information received from the host computer 2 to fiscal memory 10, and the first control unit 5 executes a read process reading the fiscal information stored in the fiscal memory 10 without involving processing by the second control unit 7.

Abstract: An information processing apparatus which has a first and second mode in which power consumption is lower than that of the first mode is disclosed that includes an accepting unit configured to accept an operation; a display configured to transmit a notification when the apparatus switches from the second mode to the first mode; and a control unit configured to store function data representing function providing units registered; generate, in response to the notification received from the display, a first screen indicating the units and cause the display to display the first screen; store identification data identifying one of the units selected; transmit, in response to the notification received from the display, the notification to the units represented in the function data; and when a response is received from the unit identified in the identification data, cause the display to display a second screen generated by the identified unit.

Abstract: An information processing device includes a security level determining unit that determines a security level, the security level indicating a level of security related to leakage of information in an image processing device, and an instructing unit that changes an output method to an output method having a high security level and gives at least one of two instructions, in a case where the security level does not satisfy a predetermined condition, the two instructions including an instruction for outputting information notifying the output method that has been changed, and an instruction for outputting information prompting for change of the security level in the image processing device.

Abstract: The present invention describes an optical reader system for reading a tag associated with an object in a datacenter. The optical reader system includes: a fixed first optical reader including a first image capturing device and a first light source, wherein the first light source is capable of directing light towards the tag of an object, wherein the first image capturing device is capable of taking an optical image of the tag when illuminated by light from the first light source; and a processing device capable of processing optical image data from the first image capturing device to determine tag information.

Abstract: A system of authenticating an item with a security mark includes a substrate (100); wherein the security mark is printed on the substrate with invisible ultraviolet (UV) absorbing ink (102); a coating comprised of UV fluorescent varnish (104) applied over the security mark and substrate; wherein an area comprising the security mark and coating with is illuminated with UV light; wherein a reduced fluorescence image of the security mark is identified; wherein the reduced fluorescence image is compared with the security mark; and wherein the item is authenticated if the reduced fluorescence image matches the security mark.

Abstract: An IC card may include a substrate having opposing first and second surfaces, and a circuit carried by the substrate adjacent the first surface of the substrate. The substrate may include a first area defining a first sector of the substrate carrying the circuit and configured to be separated from the IC card, the first sector having a form and size based upon a first IC card format, the first area having a first line delimiting the first sector to the first IC card format. The substrate may include a second area defining a second sector around the first sector and configured to be separated from the IC card based upon a second line. The second sector may have a form and size based upon at least one of a second IC card format and a third IC card format.

Abstract: A data analysis system to analyze data. The data analysis system includes a data buffer configured to receive data to be analyzed. The data analysis system also includes a state machine lattice. The state machine lattice includes multiple data analysis elements and each data analysis element includes multiple memory cells configured to analyze at least a portion of the data and to output a result of the analysis. The data analysis system includes a buffer interface configured to receive the data from the data buffer and to provide the data to the state machine lattice.

Abstract: Discriminative pretraining technique embodiments are presented that pretrain the hidden layers of a Deep Neural Network (DNN). In general, a one-hidden-layer neural network is trained first using labels discriminatively with error back-propagation (BP). Then, after discarding an output layer in the previous one-hidden-layer neural network, another randomly initialized hidden layer is added on top of the previously trained hidden layer along with a new output layer that represents the targets for classification or recognition. The resulting multiple-hidden-layer DNN is then discriminatively trained using the same strategy, and so on until the desired number of hidden layers is reached. This produces a pretrained DNN. The discriminative pretraining technique embodiments have the advantage of bringing the DNN layer weights close to a good local optimum, while still leaving them in a range with a high gradient so that they can be fine-tuned effectively.

Abstract: A method for the computer-aided learning of a recurrent neural network for modeling a dynamic system which is characterized at respective times by an observable vector with one or more observables as entries is provided. The neural network includes both a causal network with a flow of information that is directed forwards in time and a retro-causal network with a flow of information which is directed backwards in time. The states of the dynamic system are characterized by first state vectors in the causal network and by second state vectors in the retro-causal network, wherein the state vectors each contain observables for the dynamic system and also hidden states of the dynamic system. Both networks are linked to one another by a combination of the observables from the relevant first and second state vectors and are learned on the basis of training date including known observables vectors.

Abstract: Systems and methods are disclosed for using machine learning (e.g., neural networks and/or combinatorial learning) to solve the non-linear problem of predicting the provisioning of a server farm (e.g., cloud resources). The machine learning may be performed using commercially available products, such as the SNNS product from The University of Stuttgard of Germany. The system, which includes a neural network for machine learning, is provided with an identification of inputs and outputs to track, and the system provides correlations between those. Rather than static rules, the machine learning provides dynamic provisioning recommendations with corresponding confidence scores. Based on the data collected/measured by the neural network, the provisioning recommendations will change as well as the confidence scores.

Abstract: A method, article of manufacture, and apparatus for identifying defects and resolutions is discussed. Computer systems may be grouped into clusters, and their health may be determined based on these groupings. System optimizations or defect solutions may be distributed to systems in unhealthy clusters, bringing them into a healthy cluster.

Abstract: A web extension and content are linked contextually. A user interaction with the content launches a process to match a regular expression rule from the web extension to a portion of the content. An entity matching the regular expression rule is recorded and transmitted back to the web extension. Matched entity is integrated into the rule. The integrated rule is displayed in a web extension pane. The integrated rule and the entity are highlighted within the web extension pane and the content pane, respectively. The highlighted entity and the highlighted rule illustrate a contextual link between the web extension and the content.

Abstract: Sensor data having values received from several sensors of a mobile device and response data associated with the sensor data may be used in the determination or training of a predictive model. Received sensor data may be input into the predictive model, and the output of the predictive model may be used in the selection and serving of content items to the mobile device. Data to effect presentation of the selected content item may be outputted to the mobile device to effect presentation. In some instances, the predictive model may be updated using the received plurality of values. The updated predictive model may be used in the selection of a subsequent content item for the mobile device. In other implementations, historical sensor data may be used with the set of received sensor data as input for the predictive model.

Abstract: The present invention is related to a decision-support system and method for providing with a score an object represented by a target item from a multidimensional space, said score being representative of the probability that the object satisfies a given binary property. The method and/or decision support-system may be performed in a computer environment comprising processing means connected to memory means.

Abstract: Techniques for customizing knowledge representation systems including identifying, based on a plurality of concepts in a knowledge representation (KR), a group of one or more concepts relevant to user context information, and providing the identified group of one more concepts to a user. The KR may include a combination of modules. The modules may include a kernel and a customized module customized for the user. The kernel may accessible via a second KR.

Abstract: A device detecting at least one scenario from predetermined scenarios wherein a physical system observed by at least one sensor is likely to be found, including: at least one sensor providing a physical system observation data sequence; a storage storing at least one statistical model associating possible observation data values with the predetermined scenarios; a computer programmed to select at least one of the scenarios based on the observation data sequence and at least one statistical model. The storage includes plural statistical models broken down into plural ordered levels between an input level, wherein each statistical model associates possible values of at least a portion of the observation data with intermediate states for the statistical model, and an output level, wherein each statistical model associating possible values of at least a portion of the intermediate states of a lower level with at least a portion of the predetermined scenarios.