Abstract: The present invention relates to a fingerprint sensor module having excellent sensing sensitivity, to a portable electronic device including same, and to a method for manufacturing same. According to one embodiment of the present invention, a fingerprint sensor module is provided which includes: a sensing unit formed on a substrate using a conductive body; a fingerprint sensor having a sensor circuit unit electrically connected to the sensing unit; a bracket accommodating the sensing unit and receiving the fingerprint sensor; and a post-processing layer positioned on a contact surface formed on the bracket so as to face the upper surface of the sensing unit. The sum of the thickness of a supporting layer between the upper surface of the sensing unit and the contact surface, and the thickness of the post-processing layer is 0.04 mm to 0.06 mm.

Abstract: A method and a system for rectifying a distorted fingerprint are provided. The method includes following steps. A feature of a distorted fingerprint is extracted, a reference distorted fingerprint whose feature is matched with the feature of the distorted fingerprint is searched for in a reference distorted fingerprint database, a dense distortion field of the reference distorted fingerprint is obtained in the reference distorted fingerprint database and the distorted fingerprint is rectified to a normal one according to the dense distortion field of the reference distorted fingerprint.

Abstract: Techniques for facial expression capture for character animation are described. In one or more implementations, facial key points are identified in a series of images. Each image, in the series of images, is normalized from the identified facial key points. Facial features are determined from each of the normalized images. Then a facial expression is classified, based on the determined facial features, for each of the normalized images. In additional implementations, a series of images are captured that include performances of one or more facial expressions. The facial expressions in each image of the series of images are classified by a facial expression classifier. Then the facial expression classifications are used by a character animator system to produce a series of animated images of an animated character that include animated facial expressions that are associated with the facial expression classification of the corresponding image in the series of images.

Abstract: A method includes identifying a named entity, retrieving images associated with the named entity, and using a face detection algorithm to perform face detection on the retrieved images to detect faces in the retrieved images. At least one representative face image from the retrieved images is identified, and the representative face image is used to identify one or more additional images representing the at least one named entity.

Abstract: Methods, systems, and computer program products for creating a personalized, automated, receptionist are provided herein. A method for providing an automated receptionist for a host user includes obtaining one or more images of a work environment of the host user; processing the one or more images to determine if a visitor is present in the work environment; and if the visitor is determined to be present in the work environment, notifying the host user that the visitor is present in the work environment, wherein at least one of the steps are performed by at least one hardware device. The method may be initiated, for example, when the host user is not present in the work environment. The automated receptionist can optionally (i) provide a greeting to the visitor; (ii) enable communication between the visitor and the host user; and/or (iii) provide one or more information items to the visitor when the host user is not present.

Abstract: Systems and methods of automatic detection of a facial feature are disclosed. Moreover, methods and systems of yaw estimation of a human head based on a geometrical model are also disclosed.

Abstract: A moving object detection method includes acquiring two depth image frames including depth information, which are obtained by continuously taking images of a moving object, the two depth image frames including a present depth image frame and at least one past depth image frame; dividing each of the two depth image frames into a plurality of blocks; calculating differences between numbers of pixels positioned in respective different depth areas in each of the plurality of blocks in the present depth image frame, and numbers of pixels positioned in respective different depth areas in each of the corresponding plurality of blocks in each of the at least one past depth image frame, which correspond to the plurality of blocks in the present depth image frame; and detecting a moving block in the present depth image frame based on the calculated difference and constituting the detected moving object.

Abstract: A method of converting user-selected printed text to a synthesized image sequence is provided. The method includes capturing a first image of printed text and generating a model information associated with the text.

Abstract: A receipt processing system includes at least one imaging device which generates an image of a paper receipt. At least one processor programmed to acquire the image of a paper receipt from the at least one imaging device, textualize the data from the acquired receipt image, define specific local partial models based on regular expressions and geometric proximity, extract information from the textualized data according to the local partial models, apply rules to the extracted information, and generate receipt data from the application of the rules.

Abstract: A method and system for tracking eye recovery. The method includes: receiving eye data from an eyewear, wherein the eye data includes images acquired by an image capturing device in the eyewear; analyzing the eye data to determine an eye recovery status, wherein the analyzing is carried out by comparing the eye data to an initial set of eye conditions; and providing a notification of the eye recovery status.

Abstract: A method of image processing. A band-averaged spectral radiance is measured using at least one optical filter upon scanning a plurality of original radiances. The measured band-averaged spectral radiance includes a measured in-band-averaged spectral radiance and a measured band-gap-averaged spectral radiance. A multispectral radiance vector is generated from the measured band-averaged spectral radiance. The multispectral radiance vector and an out-of-band correction transform matrix corresponding to the at least one optical filter are matrix-multiplied to generate a band-averaged spectral radiances image vector representing a plurality of recovered band-averaged spectral radiances. The plurality of recovered band-averaged spectral radiances is analyzed for a presence of a target.

Abstract: There is provided a communication system that includes a vehicle and a mobile device, and a method uses that system to provide vehicle owner's manual information. The method includes the steps of: receiving a digital image of a portion of a vehicle; identifying an object of the vehicle in the digital image; associating the object with at least one reference object image stored in an object library comprising a plurality of reference object images; displaying owner's manual information based on the association; and storing at least a portion of the digital image in the object library as a new reference object image for future identifications.

Abstract: A storyboard interface displaying key frames of a video may be presented to a user. Individual key frames may represent individual shots of the video. Shots may be grouped based on similarity. Key frames may be displayed in a chronological order of the corresponding shots. Key frames of grouped shots may be spatially correlated within the storyboard interface. For example, shots of a common group may be spatially correlated so that they may be easily discernable as a group even though the shots may not be temporally consecutive and/or or even temporally close to each other in the timeframe of the video itself.

Abstract: The present invention provides a human body security inspection method and system. The method comprises: retrieving in real-time scanning row or column image data of a personal to be inspected; transmitting in real-time the image data to an algorithm processing module and processing these image data by the module; automatically recognizing a suspicious matter by a suspicious matter automatic target recognition technique, after retrieving an image data of an entire scanning image of the personal; any of the following three inspection modes is selected, so as to perform a further processing on basis of the recognition result of the suspicious matter, (1) in a manner of automatic target recognition technique, (2) in a combination manner of the automatic target recognition technique and a privacy protection image; and (3) a combination manner of the automatic target recognition technique, a privacy protection image and human intervention.

Abstract: Video is often used to count and identify people entering and exiting an area, such as a store. This information may be used to determine the number and amount of time people are in the area, and other statistics and characteristics. These variables may be skewed by the associates employed or working in the area, instead of patrons or others. The present disclosure is directed to method for eliminating associates from video by facial analysis and the application of analytics.

Abstract: Disclosed are an apparatus and a method for generating a virtual lane, which correct or estimate an undetected lane depending on detection of a lane at each side and a system for controlling lane keeping of a vehicle with the apparatus. The apparatus for generating a virtual lane according to the present invention includes: a lane detection judging unit configured to judge whether both lanes of a front road are normally detected based on a front image; a lane correction unit configured to generate a virtual lane by correcting one lane based on the other lane when it is judged that the one lane of both lanes is not normally detected; and a lane estimation unit configured to generate the virtual lane by estimating both lanes based on previously detected lane information when it is judged that either lane is not normally detected.

Abstract: A system and method for detecting a seat belt violation in a vehicle. The method may include receiving an image including a region of interest where a seat belt is visible if the seat belt is buckled. One or more first features may be extracted from the image. The one or more first features may be compared to one or more second features extracted from a plurality of sample images. It may be determined whether the seat belt is buckled in the image based upon the comparison.

Abstract: A system can include a processing device and a biometric sensing device operatively connected to the processing device. The processing device can be configured to remove noise from a biometric image captured by the biometric sensing device using at least two different calibration images. One calibration image characterizes noise of the biometric sensing device at a first signal level while another calibration image characterizes the noise of the biometric sensing device at a second signal level.

Abstract: A method of image processing using a cellular simultaneous recurrent network (CSRN) is provided. In particular, the method includes a processor that sets initial parameters of the CSRN and generates a target image based on the initial parameters. Then, a training process is executed to learn an image processing task. The training process performs an image processing task on one of a plurality of sub-images of an input image and then the sub-image is stored after the training process. The training process is then repeated until the image processing task has been performed on each of the plurality of sub-images of the input image. Once the training process is complete, an image transformation is performed on the input image and an output image is displayed based on the transformed input image.

Abstract: An apparatus, method, and computer-readable storage medium for determining a rotation angle of text. The method includes computing, for each object of a plurality of objects included in text within an image, a distance to a closest neighboring object, computing an average distance of the distances to the closest neighboring objects, determining a ratio between the average distance and an average font stroke width, the average font stroke width being an average of a font stroke width of each of the plurality of objects, and determining a rotation angle of the text by comparing the ratio to a threshold value.

Abstract: Methods and apparatus for generating a binarization threshold are described. A set of input image pixels is processed. Edges including edge pixels are identified. In various embodiments histograms of pixel values and edge pixel values are generated and jointly used in determining an appropriate binarization threshold. In some embodiments a first histogram of edge pixel values is used to determine an interval of pixel values of interest to limit the set of pixel values used for determining a binarization threshold statistically. In some embodiments a first histogram corresponding to edge pixel values and a second histogram corresponding to image pixel values are generated, subjected to one or more smoothing operations, and truncated, based on information derived from the edge histogram. Characteristics of the histograms including, e.g., minima, maxima, are determined, evaluated, and used to generate a best global threshold.

Abstract: A method and system to enhance analysis of electrophoretic bands by overlaying only the pixels of interest. The overlaid pixels are superimposed as a layer above, i.e., in the foreground of, the overlaid image, i.e., in the background. A user employs the superimposed pixels for molecular weight determination and is still able to generate densitometry analysis of the remaining pixels in the overlaid image.

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

Abstract: Image segmentation utilizing the color and brightness of sections of the image to generate super-pixels is described herein. The image segmentation is implemented pixel-wise. A color triangle is utilized in the classification of the pixels. By using a hierarchical classification table, the memory usage is significantly reduced, and the image segmentation is able to be performed on-the-fly in real-time.

Abstract: The present invention relates to the field of image processing and methodologies to construct a descriptor from binary silhouette images. The method comprises the steps of; receiving a parametric equation of a closed planar curve, choosing nodes on the closed planar curve with equal intervals, generating a continuous scale space of the nodes on the curve by successively convolving the planar curve with a Gaussian filter and down-sampling it after each octave, calculating orientation vectors and orientation angle values for each sampled point at each interval of each octave, creating a orientation scale space (OSS) which is a matrix obtained from stacking each orientation angle value on top of each other, representing the outline with a descriptor including all the orientation vectors and their respective parameters position and scale.

Abstract: An image processing apparatus comprises a registration unit configured to perform a first registration between a first set of image data and a second set of data, and a second registration between the first set of image data and the second set of data. The apparatus also comprises a registrability determination unit configured to determine a difference between the first registration and the second registration, and to determine a measure of registrability between the first set of image data and the second set of data based on the difference between the first registration and the second registration.

Abstract: An apparatus for detecting an afterimage candidate region includes: a comparison unit which compares gradation data of an n-th frame with integrated gradation data of an (n?1)-th frame and generates integrated gradation data of the n-th frame, where n is a natural number; a memory which provides the integrated gradation data of the (n?1)-th frame to the comparison unit and stores the integrated gradation data of the n-th frame; and an afterimage candidate region detection unit which detects an afterimage candidate region based on the integrated gradation data of the n-th frame, where each of the integrated gradation data of the n-th frame and the integrated gradation data of the (n?1)-th frame comprises a comparison region and a gradation region.

Abstract: Apparatus, methods, and articles of manufacture for obtaining examples that break a visual expression classifier at user devices such as tablets, smartphones, personal computers, and cameras. The examples are sent from the user devices to a server. The server may use the examples to update the classifier, and then distribute the updated classifier code and/or updated classifier parameters to the user devices. The users of the devices may be incentivized to provide the examples that break the classifier, for example, by monetary reward, access to updated versions of the classifier, public ranking or recognition of the user, a self-rewarding game. The examples may be evaluated using a pipeline of untrained crowdsourcing providers and trained experts. The examples may contain user images and/or depersonalized information extracted from the user images.

Abstract: An image processing device includes an input reception section that receives a learning image and a correct answer label, a processing section that performs a process that generates classifier data and a processing target image, and a storage section. The processing section generates the processing target image that is the entirety or part of the learning image, calculates a feature quantity of the processing target image, generates the classifier data based on training data that is a set of the feature quantity and the correct answer label assigned to the learning image that corresponds to the feature quantity, generates an image group based on the learning image or the processing target image, classifies each image of the image group using the classifier data to calculate a classification score of each image, and regenerates the processing target image based on the classification score and the image group.

Abstract: A tincture adjustment value used to adjust a monochrome signal to a tincture desired by a user is set, and a tincture conversion table and chromaticity line table are generated based on that tincture adjustment value and the profile of an image output apparatus. Using the generated tables, a lightness signal L* corresponding to an input monochrome signal is converted into a distance signal l on a chromaticity line, and the distance signal l is converted into a chromaticity signal (a*, b*). The lightness signal L* and chromaticity signal (a*, b*) are converted into a color signal of the image output apparatus.

Abstract: An image forming apparatus for performing image formation using a recording material includes an acquisition unit configured to acquire a remaining amount of the recording material, a prediction unit configured to predict a remaining amount of the recording material based on data used for the image formation, a setting unit configured to set the predicated remaining amount of the recording material as a remaining amount of the recording material to be displayed, and a control unit configured to perform control so that when the predicted remaining amount of the recording material reaches a remaining amount of the recording material to be acquired next by the acquisition unit, the remaining amount of the recording material to be displayed is maintained at the remaining amount of the recording material to be acquired next by the acquisition unit until the acquisition unit acquires the next remaining amount of the recording material.

Abstract: A method of selectively addressing one or more transponders in a set (17) of transponders (21, 22, 23, 24, 25) is described. The method comprise transmitting a first set of one or more commands (25; 251, 252) selected to be not executable by a first subset (24; 242) of transponders or to be executable in a shorter time by the first subset of transponders than by a second subset of transponders (23; 231, 232) in the set, and transmitting a second set of one or more commands (26; 253) such that the second set of commands are received by the transponders while the second subset of transponders are still executing the first set of commands.

Abstract: An RFID transponder in one embodiment comprises a radio frequency (RF) transceiver, processing logic coupled to the RF transceiver, a switch coupled to the processing logic, a tunneling device coupled to the switch and a differential sensing circuit having a first input coupled to the tunneling device and a second input coupled to a predetermined reference voltage. In one embodiment, the tunneling device can discharge to a voltage below the predetermined reference voltage.

Abstract: A Non-transferable Radio Frequency Identification (RFID) assembly for attachment to an article comprises a RFID module; and a antenna module coupled with the RFID module, the antenna module comprising a conductive layer, a substrate, and an adhesive modification layer between the conductive layer and the substrate, the adhesive modification layer configured such that when the assembly is attached to the article and attempt to remove the assembly will cause the substrate to release and leave the conductive layer intact.

Abstract: An article of manufacture includes a substrate, a first patterned layer deposited on the substrate in a first region, and a second patterned layer deposited on the substrate in a second region. The first patterned layer encodes first information in first symbols that are detectable by an optical sensor. The second patterned layer encodes second information in second symbols that are detectable by a non-optical sensor. The second information is distinct from the first information and the second region at least partially overlaps the first region.

Abstract: An apparatus and method for authenticating an item wherein a code is assigned to the item and includes a plurality of modules arranged in an ideal image. A barcode is encoded within the modules and a picture of the code is printed onto the item. A picture of the printed code is recorded with a low resolution camera at high speed during the production process from which the barcode content is decoded. Irregularities in the picture are analyzed and a numerical value describing an individual imperfection profile is calculated. The numerical value of the imperfection profile depends on at least one unique feature inherently related to the original printing device. The individual imperfection profile is printed onto the item and/or a database record is generated. To authenticate a suspect item, the imperfection profile of the suspect item is compared with the imperfection profile of the original item.

Abstract: Described is a system for action selection based on a combination of neuromodulatory and prefrontal cortex models. The system inputs group attack probability estimates for multiple groups in a prefrontal cortex (PFC) input area of a model instance. The system encodes a dispersion of the group attack probability estimates in an anterior cingulated cortex (ACC) conflict input area of the model instance, resulting in activation of the ACC conflict input area. The activation is propagated to an action area and a neuromodulatory area of the model instance. An action strategy is selected in the action area. The action strategy is implemented, and a reward and a cost is generated for the implemented action strategy. An assessment of possible action strategies is updated based on the generated reward and cost. Each model instance modulates its subsequent action strategy selection based on the updated assessment of the possible action strategies.

Abstract: A filter assembly is provided for collecting samples in a fluid environment, the assembly utilizing a laminar convective flow to cause the flow of a preservative/fixative and a filtrate through a filter. A substantially concave dead space around the filter is flushed free of the preservative and the differential in density of the preservative and filtrate causes the more dense fluid to collect in the reservoir and preserve the collected specimen.

Abstract: Apparatus and methods for plasticity in spiking neuron networks. In various implementations, the efficacy of one or more connections of the network may be adjusted based on a plasticity rule during network operation. The rule may comprise a connection depression portion and/or a potentiation portion. Statistical parameters of the adjusted efficacy of a population of connections may be determined. The statistical parameter(s) may be utilized to adapt the plasticity rule during network operation in order to obtain efficacy characterized by target statistics. Based on the statistical parameter exceeding a target value, the depression magnitude of the plasticity rule may be reduced. Based on a statistical parameter being below the target value, the depression magnitude of the plasticity rule may be increased. The use of adaptive modification of the plasticity rule may improve network convergence while alleviating a need for manual tuning of efficacy during network operation.

Abstract: Electronic communications can be normalized using neural networks. For example, an electronic representation of a noncanonical communication can be received. A normalized version of the noncanonical communication can be determined using a normalizer including a neural network. The neural network can receive a single vector at an input layer of the neural network and transform an output of a hidden layer of the neural network into multiple values that sum to a total value of one. Each value of the multiple values can be a number between zero and one and represent a probability of a particular character being in a particular position in the normalized version of the noncanonical communication. The neural network can determine the normalized version of the noncanonical communication based on the multiple values. Whether the normalized version should be output can be determined based on a result from a flagger including another neural network.

Abstract: Deep learning is used to identify specific, potential risks to an enterprise (of which litigation is the prime example) while such risks are still internal electronic communications. The system involves mining and using existing classifications of data (e.g., from a litigation database) to train one or more deep learning algorithms, and then examining the internal electronic communications with the trained algorithm, to generate a scored output that will enable enterprise personnel to be alerted to risks and take action in time to prevent the risks from resulting in harm to the enterprise or others.

Abstract: Systems and techniques are provided for a ranking approach to train deep neural nets for multilabel image annotation. Label scores may be received for labels determined by a neural network for training examples. Each label may be a positive label or a negative label for the training example. An error of the neural network may be determined based on a comparison, for each of the training examples, of the label scores for positive labels and negative labels for the training example and a semantic distance between each positive label and each negative label for the training example. Updated weights may be determined for the neural network based on a gradient of the determined error of the neural network. The updated weights may be applied to the neural network to train the neural network.

Abstract: In one embodiment, a committed information rate (CIR) prediction is received from a machine learning model that corresponds to a predicted average traffic rate supported by a network connection. A traffic shaping strategy is adjusted based on the CIR prediction. A rate at which data is communicated over the network connection may be based on the traffic shaping policy. The effects of the adjusted traffic shaping strategy are also monitored. Feedback is further provided to the machine learning model based on the monitored effects of the adjusted traffic shaping strategy.

Abstract: A spatial and temporal memory system (STMS) processes input data to detect whether spatial patterns and/or temporal sequences of spatial patterns exist within the data, and to make predictions about future data. The data processed by the STMS may be retrieved from, for example, one or more database fields and is encoded into a distributed representation format using a coding scheme. The performance of the STMS in predicting future data is evaluated for the coding scheme used to process the data as performance data. The selection and prioritization of STMS experiments to perform may be based on the performance data for an experiment. The best fields, encodings, and time aggregations for generating predictions can be determined by an automated search and evaluation of multiple STMS systems.

Abstract: A system and method for managing online map information determines map records that represent more than one map feature by identifying map feature records where at least one attribute of the map feature record has more than value wherein each of the values has a probability of being correct that exceeds a threshold.

Abstract: Disclosed are various embodiments for an item preparation application. Business logic and machine learning may be applied to attributes of an item to generate a recommendation for item preparation prior to storage in a physical storage facility. The recommendation may indicate whether preparation is to be applied, or what type of preparation is required. A machine learning knowledge base may be updated based on feedback or overrides of recommendations.

Abstract: A method for target identification includes receiving, at one or more processors, a number of measurements of a target, each measurement from the number of measurements being observed at a predetermined time (zk), a number of target types (T), each one of the number of measurements, each one the number of target type and each one of one or more hidden states, each hidden state (xk) being characterized at the predetermined time, being correlated to one another, providing, using the one or more processors, a first conditional probability distribution, a conditional probability of a target type given a number of measurements, defined inductively, and obtaining an estimate of the target type from the first conditional probability. Systems that implement the method are also disclosed.

Abstract: Mechanisms for recommending content items based on topics are provided. In some implementations, a method for recommending content items is provided that includes: determining a plurality of accessed content items associated with a user, wherein each of the plurality of content items is associated with a plurality of topics; determining the plurality of topics associated with each of the plurality of accessed content items; generating a model of user interests based on the plurality of topics, wherein the model implements a machine learning technique to determine a plurality of weights for assigning to each of the plurality of topics; applying the model to determine, for a plurality of content items, a probability that the user would watch a content item of the plurality of content items; ranking the plurality of content items based on the determined probabilities; and selecting a subset of the plurality of content items to recommend to the user based on the ranked content items.

Abstract: In an example embodiment, for each of a plurality of possible presented actions corresponding to a particular flow of actions in a computer network, and for each of one or more possible performed actions for each possible presented action, a likelihood that a user will perform the possible performed action is determined. Then, a first presented action is identified by determining a presented action having a maximum determined likelihood, based on the determined likelihood, wherein the identifying a first presented action includes utilizing a machine learning model having one or more user covariates and one or more performed action covariates, and interactions between the one or more user covariates and the one or more performed action covariates, the user covariates including information specific to the user, the one or more performed action covariates including information specific to one or more of the possible performed actions.

Abstract: Techniques for visual representation of chart scaling are described. In implementations, a chart is employed to display data associated with a resource. Examples of such a resource include a computer hardware resource, a network resource, a financial resource, a commodity (e.g., a manufactured commodity), and so on. In implementations, a chart is associated with a scale that corresponds to a set of data values represented in the chart. Further, a chart can be rescaled to display data values in a variety of different scales, e.g., that correspond to different sets of data values. In at least some embodiments, a scale mark is provided which provides a visual indication of a scale represented in a chart. When the chart is rescaled, the scale mark is repositioned to indicate a change in scale. Thus, a scale mark can be dynamically repositioned to visually indicate different chart scales.