Abstract: An apparatus may include an ultrasonic sensor array and a control system. The control system may be configured to acquire first image data generated by the ultrasonic sensor array corresponding to at least one first reflected ultrasonic wave received by at least a portion of the ultrasonic sensor array from a target object during a first acquisition time window. The control system may be configured to acquire second image data generated by the ultrasonic sensor array corresponding to at least one second reflected ultrasonic wave received by at least a portion of the ultrasonic sensor array from the target object during a second acquisition time window that is longer than the first acquisition time window. The control system may further be configured to initiate an authentication process based on the first image data and the second image data.

Abstract: Techniques for detecting anomalies in multispectral image data, and more specifically for detecting presentation attacks by using multispectral image data in biometric security applications, are provided. In some embodiments, a system may receive multispectral image data and generate an estimation of a first image of a plurality of images of the multispectral image data, wherein the estimation is based on other images of the multispectral image data, but not the first image itself. The estimation may then be compared to the first image to generate an indication as to whether the multispectral image data represents a presentation attack. In some embodiments, a system may receive multispectral training image data and may extract features from the data to generate and store a network architecture for predicting relationships of multispectral images of subjects.

Abstract: A calculation device, for the selection and the description of points of interest in at least one image decomposed into a plurality of strips, includes a first module for selecting and indexing points of interest from among a set of points of interest detected in the image, the first module being configured to select, in each strip, at most a fixed number P of candidate points of interest, and associate a unique index with them, a second module for managing calculation of descriptors for the candidate points of interest, the first module being configured to transmit the candidate points of interest and their index to the second module immediately at the end of the processing of K lines of a strip, with K an integer number greater than or equal to 1 and strictly less than the number of lines per strip.

Abstract: Provided is a method of evaluating performance of a bio-sensor. The method may include obtaining an input bio-image by using the bio-sensor that is to be evaluated; obtaining a sample database including information of M reference bio-images, where M is a natural number; and calculating a quality score of the input bio-image based on the input bio-image and the information in the sample database. The quality score may provide an objective and quantitative score for evaluating a bio-image.

Abstract: An image processing method, a photographing device and a storage medium are disclosed in the invention. The image processing method includes: determining whether a physiological feature in a current collection area needs to be detected according to a photographing parameters of a lens during capturing; and directly storing an image captured from the current collection area when the physiological feature in the current collection area does not need to be detected; or detecting the physiological feature in the current collection area and processing the captured image when the physiological feature in the current collection area needs to be detected.

Abstract: A method of identifying a snow covered road includes creating a forward image of a road surface. The forward image is analyzed to detect a tire track in the forward image. When a tire track is detected in the forward image, a message indicating a snow covered road surface is signaled. When a tire track is not detected in the forward image, a rearward image, a left side image, and a right side image are created. The rearward image, the left side image, and the right side image are analyzed to detect a tire track in at least one of the rearward image, the right side image, and the left side image. A message indicating a snow covered road surface is signaled when a tire track is detected in one of the rearward image, the left side image, or the right side image.

Abstract: Techniques for detecting an active area of interest within video data are disclosed. A videoconferencing system may include a sensor configured to capture sequential frames of video image data during a videoconference, and a processing subsystem. The processing subsystem may be configured to, for a given frame, apply filter(s) to individual pixels of the given frame to identify one or more pixels of interest within the given frame, where the filter(s) include a motion detection filter. The processing subsystem may further apply a multidimensional filter to the one or more pixels of interest, the multidimensional filter specifying one or more programmable density thresholds. The processing subsystem may further designate at least one group of pixels of interest that satisfies the one or more programmable density thresholds as an active area of interest; and control an aspect of the videoconference based at least in part upon the active area of interest.

Abstract: A plurality of recognition positions each recognized by a recognizer as a position of a target object on an input image are acquired. At least one representative position is obtained by performing clustering for the plurality of recognition positions. The representative position is edited in accordance with an editing instruction from a user for the representative position. The input image and the representative position are saved as learning data to be used for learning of the recognizer.

Abstract: Disclosed herein are system, method, and computer program product embodiments for automatically detecting cloned vehicle identifiers. An embodiment operates by determining legitimate trajectory information and illegitimate trajectory information corresponding to vehicular travel in a geographic area, and training a decision model to detect a cloned vehicle identifier based on the legitimate trajectory information and the illegitimate trajectory information. Furthermore, the embodiment identifies vehicles potentially using a cloned vehicle identifier based on the decision model and common driving patterns within the geographic area. In some embodiments, enforcement agents are notified of the vehicles potentially using a cloned vehicle identifier.

Abstract: An image analysis apparatus includes an extraction section, an acquisition section, an exclusion section, and a recognition section. The extraction section extracts, from a target image, a text area including a character. The acquisition section acquires a predetermined target-object condition. The predetermined target-object condition defines a target object. The exclusion section searches for the target object in the text area and excludes the target object from a target of character recognition. The text area is extracted by the extraction section. The target object satisfies the predetermined target-object condition. The recognition section performs character recognition on the text area. The text area is an area from which the target object is excluded.

Abstract: In some examples, a window in an image may be identified. At least one property of the window may be identified. Based on the at least one property, whether the window is an error block may be determined.

Abstract: An image processing device includes a memory, and a processor coupled to the memory, wherein the processor is configured to execute acquiring a captured image including an original document region, detecting a color component value of a predetermined area in the captured image, detecting an edge in the predetermined area to acquire an edge amount indicating any one or both of a density of the edge and an edge intensity, and identifying, based on the color component value and the edge amount, a difference between a background of the original document region and a background of a background region obtained by removing the original document region from the captured image.

Abstract: Embodiments of the present invention disclose a method, computer program product, and system for cataloging images based on a gridded color histogram analysis. The computer accesses an image gallery specified by a user, wherein the image gallery is at least one of an image gallery stored on a user computing device, an image gallery stored on a user account at a third-party image storage, or an image gallery searched on the web. The computer receives a request to search the image gallery specified by the user. The computer performs a search of the image gallery, wherein the search is using a color based histogram algorithm based on a user input. The computer transmits a cataloged and sorted image gallery to the user computing device to be displayed.

Abstract: In various example embodiments, a system and method for determining an item that has confirmed characteristics are described herein. An image that depicts an object is received from a client device. Structured data that corresponds to characteristics of one or more items are retrieved. A set of characteristics is determined, the set of characteristics being predicted to match with the object. An interface that includes a request for confirmation of the set of characteristics is generated. The interface is displayed on the client device. Confirmation that at least one characteristic from the set of characteristics matches with the object depicted in the image is received from the client device.

Abstract: An image comparison method constituted of: receiving images; point matching the images to identify tentatively corresponding points; responsive to the identified points, defining a plurality of first image tiles within the first image; defining a plurality of second image tiles within the second image, each corresponding to a respective first image tile; adjusting the intensity of a set of pixels responsive to the cumulative relative frequency of the respective pixel intensity within the respective tile; for each tile, applying a plurality of non-linear functions; for each function, separately determining a moment of the outcome of the respective non-linear function for each axis of the respective tile; for each first image tile and corresponding second image tile, determining a distance between the subspaces spanned by moment vectors; and responsive to determined distances, determining whether respective first and second image tiles comprise observations of the same portion.

Abstract: Provided is a carbonization time point management service providing system when food stuff such as fish or meat (or fish-meat stuff) cooked, the system receives, from a cooking management service providing server, fish-meat detecting information including a cooking time from a cooking start point in time of fish-meat stuff to a carbonization time point at which the fish-meat stuff starts to be carbonized, generates a cooking completion acquisition signal in accordance with an end timing of the cooking time, and transmits a cooking completion indication signal to cooking management dedicated application content, so that the cooking completion indication signal is externally displayed, thereby enabling real-time confirmation of a cooking process of fish-meat stuff which is placed on a heating device even in the case of performing the other work at the remote place during cooking of the fish-meat stuff, to improve convenience when cooking the fish-meat stuff.

Abstract: Aspects of the subject disclosure may include, for example, a method for determining a first set of features in first images of first media content, generating a similarity score by processing the first set of features with a favorability model derived by identifying generative features and discriminative features of second media content that is favored by a viewer, and providing the similarity score to a network for predicting a response by the viewer to the first media content. Other embodiments are disclosed.

Abstract: Methods, systems, and apparatus for training model parameters stored in shared memory to predict risk. The method may include obtaining training data that includes a plurality of training data structures that each represent attributes of an entity, wherein each training data structure represents (i) features derived from a first set of categories defined by a first model and from a second set of categories defined by a second model, and (ii) a risk-level associated with the entity. For each respective training data structure, providing the training data structure as an input to the model, receiving an output from the model based on the model's processing of the training data structure, determining an amount of error between the output of the model and the risk-level of the training data structure, and adjusting a parameter value of the model stored in a shared memory based on the determined error.

Abstract: Provided is a learning device that includes at least one processing device configured to select image data as learning data based on classification confidences from a plurality of image data. Each of the classification confidences indicates a likelihood of accuracy of classification for a respective one of the plurality of image data. The at least one processing device is configured to: store the selected image data into a storage; perform learning using the stored image data and generating a learning model; perform classification of the image data used for the learning by using the learning model and update a classification confidence for the classified image data; and update the classification confidences each associated with a respective one of the plurality of image data when a count of image data on which the classification has been performed satisfies a predetermined condition.

Abstract: Technologies are provided for automated crime scene analysis using machine learning. Firearm models, types, or even specific firearms may be automatically detected from captured audio files or continuous audio streams (e.g., recording microphones) using machine learning techniques. The detection may also be based on (or enhanced by) captured still images or video files/streams. Further information such as crime scene layout, wound types and locations, and similar information may be provided to the analysis service through manual input or automated capture (e.g., through analysis of image/video data). A number of firearms used in the commission of the crime may also be detected. Specific firearm types may be associated with specific crime types. Similar techniques may also be used to detect and classify types and quantity of explosive material.

Abstract: A system and method multifunction peripheral policy-based configuration includes a processor and a memory storing a policy data for each of a plurality of multifunction peripheral device configuration policies. A selectable list of the device policies is displayed. A user selects one or more device configuration policies from the list, and the processor tests selected device configuration policies. Policies are identified as single or additive policy types. Selected policies are added when not identified as a single policy type.

Abstract: An encoder includes a base part, a rotation part provided rotatably about a rotation axis relative to the base part, an irregular pattern placed along about the rotation axis in the rotation part, an image pickup device placed in the base part and capturing the pattern, and a determination part that determines a rotation state of the rotation part relative to the base part using an imaging result of the image pickup device.

Abstract: Durable fabric RFID labels are provided for mounting on garments, fabrics and other fabric-containing items, the mounting and durability being before, during or after manufacturing and processing of the items. These labels are robust enough to withstand processing during manufacturing, while being capable of remaining on the item during inventory handling, merchandising and consumer use, including washing and drying. The durable labels include an RFID inlay, a face sheet overlying a first face of the RFID inlay, and a functional adhesive, such as a hot-melt adhesive, overlying a second face of the RFID inlay. The face sheet can be of printable material or have indicia or be a printed face sheet. The functional adhesive can be of a moisture-resistive type. The RFID inlay can be encased within a pocket of polymeric material. A polymeric sheet reinforcement layer can be adhered to and cover all or a portion of the RFID inlay.

Abstract: A transaction card is provided. The transaction card includes a card frame having a card inlay and a card housing. The transaction card also includes a magnetic stripe disposed inside the card frame between the card inlay and the card housing.

Abstract: A credit card shaped element with a dynamic stripe which may be activated or deactivated, where the stripe is formed by a plurality of coils provided along the curve thereof, where the plurality of coils are activated in a sequential manner, where a plurality of proximity sensing coils are provided in series along the curve thereof, where the plurality of coils are formed by one coil contacted at multiple positions along its length by contact pads and where the signals fed to the coils cause an output of a reader coil to output a signals with a minimum slope or where one signal fed to one track has no peaks when the other signal is close to zero.

Abstract: A wireless tag apparatus and related medication compliance monitoring techniques are disclosed. In some embodiments, the apparatus may include a motion detector configured to detect at least one of a movement of and an impact to the apparatus. The apparatus also may include a sensing circuit configured to scan, in response to the detection by the motion detector, a sensor array associated with a medication dispenser external to the apparatus to determine whether contents of at least one medication compartment of the medication dispenser have been expelled. The apparatus further may include a transmitter configured to transmit, in response to the detection by the motion detector, a signal including data pertaining to whether the contents of the at least one medication compartment have been expelled, wherein the signal is a radio frequency signal. In some embodiments, the apparatus may be configured for use, for example, in monitoring medication compliance.

Abstract: An RFID reader, tuning method and computer readable medium. A carrier signal is transmitted. A plurality of feedback signals produced in the RFID reader are measured when transmitting the carrier signal with a variable matching network set at a plurality of impedance values that are different from one another. It is determined which impedance value of the plurality of impedance values produced a feedback signal having a highest level of the plurality of feedback signals. The variable matching network is set to the determined impedance value.

Abstract: Disclosed embodiments generally relate to a transaction card with a fabric inlay. The transaction card may include a housing component having a first housing surface opposite a second housing surface and an inlay component having a first inlay surface opposite a second inlay surface. The inlay and housing may be joined along the second inlay surface and the first housing surface. In addition, the first inlay surface may include a fabric material and a backer layer configured to support the fabric material of the first inlay surface.

Abstract: Systems and methods for monitoring an attachment, such as a breaker head, for a mining machine. The system includes a breaker head, a wireless identification tag, a sensor, and an electronic processor. The wireless identification tag is coupled to the breaker head and includes a tag memory storing an identifier of the breaker head and an antenna. The sensor is configured to generate measurement data associated with the breaker head. The electronic processor is configured to receive measurement data from the sensor, generate a notification based on the measurement data, and transmit the notification to the wireless identification tag. Additionally, the wireless identification tag is configured to store the notification. The notification can then be later retrieved from the wireless identification tag by various devices, such as a further mining machine or a reader device of maintenance personnel.

Abstract: An RFID device includes a substrate and an antenna. The antenna includes a first antenna section on a first side of the substrate and a second antenna section on the first side of the substrate and separated from the first antenna section by a non-conductive part of the substrate. The first antenna section includes a first plurality of vias that extend through the substrate from the first side of the substrate to a second side of the substrate. The antenna includes a third antenna section on at least the second side of the substrate to electrically connect conductive portions of the first plurality of vias to the second antenna section. The RFID device includes a circuit on the substrate and to process signals of the antenna.

Abstract: Approaches to leveraging knowledge base of groups in mining organizational data. A communication service initiates operation(s) to leverage knowledge base of groups upon detecting a question supplied by a requestor. Contextual information associated with the requestor is determined in relation to the question. Next, a knowledge graph is queried with the question and the contextual information. An answer associated with the question is identified within the knowledge graph. The answer includes a source. Furthermore, the answer and the source is provided to the requestor. Upon receiving to feedback associated with the answer from the requestor, the knowledge graph is modified based on the feedback.

Abstract: Techniques for mapping actions and objects to tasks may include identifying a task to be performed by a virtual assistant for an action and/or object. The task may be identified based on a task map of the virtual assistant. In some examples, the task may be identified based on contextual information of a user, such as a conversation history, content output history, user preferences, and so on. The techniques may also include customizing a task map for a particular context, such as a particular user, industry, platform, device type, and so on. The customization may include assigning an action, object, and/or variable value to a particular task.

Abstract: SLAM systems are provided that utilize an artificial neural network to both map environments and locate positions within the environments. In some example embodiments, a sensor arrangement is used to map an environment. The sensor arrangement acquires sensor data from the various sensors and associates the sensor data, or data derived from the sensor data, with spatial regions in the environment. The sensor data may include image data and inertial measurement data that effectively describes the visual appearance of a spatial region at a particular location and orientation. This diverse sensor data may be fused into camera poses. The map of the environment includes camera poses organized by spatial region within the environment. Further, in these examples, an artificial neural network is adapted to the features of the environment by a transfer learning process using image data associated with camera poses.

Abstract: The present disclosure is directed to systems and methods of bit-serial, in-memory, execution of at least an nth layer of a multi-layer neural network in a first on-chip processor memory circuitry portion contemporaneous with prefetching and storing layer weights associated with the (n+1)st layer of the multi-layer neural network in a second on-chip processor memory circuitry portion. The storage of layer weights in on-chip processor memory circuitry beneficially decreases the time required to transfer the layer weights upon execution of the (n+1)st layer of the multi-layer neural network by the first on-chip processor memory circuitry portion. In addition, the on-chip processor memory circuitry may include a third on-chip processor memory circuitry portion used to store intermediate and/or final input/output values associated with one or more layers included in the multi-layer neural network.

Abstract: Neuromorphic processing apparatus is provided. The present invention may include a spiking neural network comprising a set of input spiking neurons each connected to each of a set of output spiking neurons via a respective synapse for storing a synaptic weight which is adjusted for that synapse in dependence on network operation in a learning mode of the apparatus, and each synapse is operable to provide a post-synaptic signal, dependent on its synaptic weight, to its respective output neuron. The present invention may further include a pre-processor unit adapted to process input data, defining a pattern of data points, to produce a first set of input spike signals which encode values representing respective data points, and a second set of input spike signals which encode values complementary to respective said values representing data points, and to supply the input spike signals to respective predetermined input neurons of the network.

Abstract: A memory device may include a memory cell circuit; a memory interface circuit configured to receive a read command and a write command from a host and to control the memory cell circuit according to the read command and the write command; and a neural network processor configured to receive a neural network processing command from the host, to perform a neural network processing operation according to the neural network processing command, and to control the memory cell circuit to read or write data while performing the neural network processing operation.

Abstract: Processors and methods for neural network processing are provided. A method in a processor including a matrix vector unit is provided. The method includes receiving vector data and actuation vector data corresponding to at least one layer of a neural network model for processing using the matrix vector unit, where each of digital values corresponding to the vector data and the actuation vector data is represented in a sign magnitude format. The method further includes converting each of the digital values corresponding to at least one of the vector data or the actuation vector data to corresponding analog values and multiplying the vector data and the actuation vector data in an analog domain and providing corresponding multiplication results in a digital domain.

Abstract: The present disclosure is directed to systems and methods of implementing an analog neural network using a pipelined SRAM architecture (“PISA”) circuitry disposed in on-chip processor memory circuitry. The on-chip processor memory circuitry may include processor last level cache (LLC) circuitry. One or more physical parameters, such as a stored charge or voltage, may be used to permit the generation of an in-memory analog output using a SRAM array. The generation of an in-memory analog output using only word-line and bit-line capabilities beneficially increases the computational density of the PISA circuit without increasing power requirements.

Abstract: An example a method of optimizing a neural network having a plurality of layers includes: obtaining an architecture constraint for circuitry of an inference platform that implements the neural network; training the neural network on a training platform to generate network parameters and feature maps for the plurality of layers; and constraining the network parameters, the feature maps, or both based on the architecture constraint.

Abstract: A method of operating a health tracking system is disclosed. The method comprises: receiving a first data record comprising at least a first descriptive string regarding a consumable item, the first descriptive string having at least one word thereof incorrectly spelled; generating a vector using the first descriptive string using a machine learning model; identifying a second descriptive string which corresponds to the consumable item and which has a correct spelling of the at least one incorrectly spelled word by applying the machine learning model to the generated vector; calculating a confidence factor regarding the identified second descriptive string using the machine learning model; and when it is determined that the confidence factor exceeds a predetermined threshold, (i) modifying the first data record by replacing the first descriptive string with the second descriptive string, and (ii) storing the modified first data record in the database.

Abstract: Example implementations described herein are directed to systems and methods for estimating the remaining useful life of a component or equipment through the application of models for deriving functions that can express the remaining useful life over time. In an aspect, the failure acceleration time point is determined for a given type of component, and a function is derived based on the application of models on the failure acceleration time point.

Abstract: A method for pruning a neural network includes initializing a plurality of threshold values respectively corresponding to a plurality of layers included in the neural network; selecting one of the plurality of layers; adjusting the threshold value of the selected layer; and adjusting a plurality of weights respectively corresponding to a plurality of synapses included in the neural network.

Abstract: There is provided an information processing apparatus which is capable of more efficiently designing a neural network which is fit for processing by a plurality of hardware devices, and an information processing method. The information processing apparatus includes: an acquiring unit configured to acquire restrictions relating to a plurality of hardware devices; and a determining unit configured to perform determination as to whether or not a neural network satisfies the restrictions.

Abstract: An expert system processes communication data to extract entities and topics. The expert system generates relationship graphs and relationship scores between the entities and topics. The system can identify entities that are expert in a given topic. The expert system uses a knowledge engine to provide different services and applications.

Abstract: Machine logic that uses weather data and play style of players to predict outcomes of sporting contests. These predicted outcomes can be used for various purposes such as sporting contest scheduling, or controlling of play conditions. Also, machine logic to agglomerate data from various players together to increase data purity and/or learn from observed data (for example, learn from outcomes of training matches).

Abstract: Methods and systems for providing transportation service are described. The method may include receiving, from a passenger terminal, a transportation service request indicating a first type of service vehicle. The method may further include determining route information associated with the transportation service request, and determining an estimated waiting time for the transportation service request to be fulfilled based on the first type of service vehicle and the route information. The method may also include comparing the estimated waiting time to a predetermined time period, and when the estimated waiting time is equal to or longer than the predetermined time period, determining a second type of service vehicle. The second type of service vehicle may have an estimated waiting time shorter than the predetermined time period. The method may also include transmitting, to the passenger terminal, a message recommending the second type of service vehicle.

Abstract: Recommendations for new experiments are generated via a pipeline that includes a predictive model and a preference procedure. In one example, a definition of a development task includes experiment parameters that may be varied, the outcomes of interest and the desired goals or specifications. Existing experimental data is used by machine learning algorithms to train a predictive model. The software system generates candidate experiments and uses the trained predictive model to predict the outcomes of the candidate experiments based on their parameters. A merit function (referred to as a preference function) is calculated for the candidate experiments. The preference function is a function of the experiment parameters and/or the predicted outcomes. It may also be a function of features that are derived from these quantities. The candidate experiments are ranked based on the preference function.

Abstract: Systems and methods are provided for embedded data inference. The systems and methods may process camera and other sensor data in by leveraging processing and storage capacity of one or more devices nearby or in the cloud to augment or update the sensor processing of an embedded device. The joint processing may be used in stationary cameras or in vehicular systems such as cars and drones, and may improve crop assessments, navigation, and safety.

Abstract: Disclosed herein are system, method, and computer-readable device embodiments for automatically correcting erroneous license plate numbers generated by automatic license plate recognition. An embodiment operates by selecting a set of adjacent license plate recognition (LPR) stations comprising a first LPR station, a second LPR station, and a third LPR station, accessing an error pattern probability matrix for at least the third recognized license plate number accessed in relation to at least one of the first recognized license plate number or the second recognized license plate number differing from the third recognized license plate number, and determining a corrected license plate number based on the error pattern probability matrix. In some embodiments, LPR correction may be accurately realized even when an erroneous license plate numbers are missing characters, such as by visual obstruction from an LPR camera or sensor, or has no characters in common with the real license plate number.

Abstract: The present disclosure provides a medical data analysis method and device. The method comprises locating a semantic subspace of the subject in a medical data set by taking the physical parameter as a feature; and analyzing the probability P1 of the subject being in the semantic subspace that the subject belongs to by judging the semantic consistency of the semantic subspace where the physical parameter of the subject exists. In addition, it is also possible to analyze the probability P2 of the subject being in the node that the subject belongs to based on the evidence transference score of the physical parameter of the subject on the medical knowledge graph. P=?×P1+(1??)×P2 The probability P of the subject being in the semantic subspace or node that the subject belongs to can be determined by P=?×P1+(1+?)×P2, wherein ? is a reconciling parameter, 0<?<1. Through these solutions, the analysis accuracy and efficiency can be improved and the cost can be decreased.