Abstract: A computer-implemented method for training a neural network system. The method includes receiving at least a first data vector at a first layer of the neural network system; applying a function to the first data vector to generate at least a second data vector, wherein the function is based on a layer parameter of the first layer that includes at least a weight matrix of the first layer; comparing at least the first data vector and the second data vector to obtain a loss value that represents a difference between the first data vector and the second data vector; updating the layer parameter based on the loss value; and adjusting the layer parameter based on a comparison of the updated layer parameter with a threshold value of the first layer.

Abstract: Disclosed is a system for processing data streams that includes a parallel processor and a netflow aggregator module to generate a storage representation for data packets. Each storage representation includes segments of information about the data packet, the segments of information including information about a communication protocol specification related to the data packet. The netflow aggregator module generates a composite index to identify a data packet association characteristic for each data packet and stores the composite index in a segment of the storage representation. The netflow aggregator module groups data packets by their composite index. The netflow aggregator module generates a session flow identifier by identifying a beginning and/or end of a transmission netflow for each data packet having the same data packet association characteristic. The netflow aggregator module aggregates and orders the data packets having the same session flow identifiers into a flow channel.

Abstract: Disclosed is a fluid leak detection system for monitoring leaks in a pressurized vessel. The system includes a temperature sensor, a pressure sensor, and a processing module. The processing module is configured to any one or combination of: compare a temperature output from the temperature sensor to a reference temperature to produce a temperature differential; or compare a pressure output from the pressure sensor to a reference pressure to produce a pressure differential. The processing module is configured to any one or combination of: monitor the pressure differential to indicate when a leak occurs in a pressurized vessel; or monitor the temperature differential to indicate when a leak occurs in a pressurized vessel. The processing module is configured to generate an alert when a leak is detected.

Abstract: An exemplary system and method for generating a network diagram. The system includes a first interface for receiving one or more first datasets of network data. A processor configured to generate a first network diagram and identify changes in a network by overlaying one or more network diagram layers or a second network diagram on the first network diagram. The one or more network diagram layers corresponding to captured packets of the first datasets or of second datasets and the second network diagram corresponding to captured packets of the second datasets. A second interface for graphically displaying the network diagram by highlighting and/or animating one or more nodes and/or node connections based on at least one of the identified changes from the overlay.

Abstract: A thermal management system includes an open circuit refrigeration circuit that has a refrigerant fluid flow path, with the refrigerant fluid flow path including a receiver configured to store a refrigerant fluid, a first control device configured to receive refrigerant from the receiver, a liquid separator, and an evaporator configured to extract heat from a heat load that contacts the evaporator, with the evaporator coupled to the first control device and the liquid separator. The system includes a pump having an inlet and an outlet, with the outlet of the pump coupled to the liquid side outlet of the liquid separator and a second control device that is coupled to an exhaust line, that is coupled to the vapor side outlet of the liquid separator through the second control device. In operation, the evaporator in the open circuit refrigeration circuit would be coupled to a heat load.

Abstract: An exemplary system for monitoring an operational area, includes a processor that receive plural data streams. Each data stream can include a different spatial characteristic of the operational area. The processor also generates a three-dimensional (3D) virtual visualization of the operational area based on observational perspectives associated with the data streams and their associated spatial characteristics and dynamically prioritizes operational sub-regions within the operational area based on the spatial characteristics. The processor generates a signal encoded with data for verifying the 3D virtual visualization of the operational area including the prioritized operational sub-regions.

Abstract: A system for generating a cyber-attack to penetrate a network. The system includes an identification module configured to identify at least one vulnerability of the network by examining at least one of a node of the network, data transmission within the network, or data received from a cyber defense mechanism; a generation module configured to generate a cyber-attack based on the at least one vulnerability of the network, and a goal to be achieved by the cyber-attack. The system includes a penetration module configured to penetrate the network with the cyber-attack and determine an effectiveness rating of the penetration; and a feedback module configured to provide a feedback to the identification module based on at least the effectiveness rating of the penetration.

Abstract: Exemplary systems and methods are directed to generating synthetic data for computer vision. A processing device generates a synthetic three-dimensional (3D) image of an object. A background image is selected, and a composite image is generated by combining the 3D image of the object and the background image. The processing device simulates: reflection or emission of at least one type of radiant energy from the surface of the object and/or the background according to a set of parameters associated with at least one of the object and the background image; and a reflectance or emittance measurement of the at least one type of radiant energy from the surface of the object by a sensor device configured for detecting the at least one type of radiant energy. The processing device generates a plurality of two-dimensional (2D) simulated images of different perspectives of the object based on simulation data.

Abstract: An exemplary system for monitoring an operational area, includes a processor that receive plural data streams. Each data stream can include a different spatial characteristic of the operational area. The processor also generates a three-dimensional (3D) virtual visualization of the operational area based on observational perspectives associated with the data streams and their associated spatial characteristics and dynamically prioritizes operational sub-regions within the operational area based on the spatial characteristics. The processor generates a signal encoded with data for verifying the 3D virtual visualization of the operational area including the prioritized operational sub-regions.

Abstract: Exemplary systems and methods are disclosed for removing steganography from digital data is disclosed. The method and system involve receiving a digital data. At least one processing device accesses a steganography purifier model. The at least one processing device includes at least a generator configured to scale a magnitude of individual data elements of the digital data from a first value range to a second value range. The scaled data elements are downsampled to remove steganography data embedded in the digital data and produce a purified version. The purified version is upsampled by interpolating new data elements between one or more adjacent data elements to provide an upsampled purified version. The magnitude of the data elements of the upsampled purified version are scaled from the second value range to the first value range to generate a purified output version.

Abstract: A method and system for geo-localization in sensor-deprived or sensor-limited environments includes receiving an image file for geo-localization of a location depicted by the image file in a sensor-deprived environment; applying a plurality of geo-localization modules to the image file; generating, by each of the plurality of geo-localization modules, a module output, each module output including a module geolocation for the location and a module confidence score for the module geolocation; generating an ensemble geolocation output, the ensemble geolocation output including an ensemble geolocation for the location and an ensemble confidence score for the ensemble geolocation, the ensemble geolocation output being a weighted combination of the module outputs; and displaying the ensemble geolocation output to a display.

Abstract: Exemplary systems and methods are disclosed for detecting embedded data in a digital image. The system includes a processing device that extracts one or more features from a digital image and analyzes the one or more extracted features in a plurality of steganography analyzers, each steganography analyzer configured to execute a different steganography algorithm. The processing device generates an output data value at each steganography analyzer, the output data value indicating a probability that the digital image includes steganography according to the steganography algorithm of the steganography analyzer.

Abstract: An exemplary device for classifying an image includes a receiving unit that receives image data. The device also includes a hardware processor including a neural network architecture to extract a plurality of features from the image data, filter each feature extracted from the image data, concatenate the plurality of filtered features to form an image vector, evaluate the plurality of concatenated features in first and second layers of a plurality of fully connected layers of the neural network architecture based on an amount of deviation in the features determined at each fully connected layer, and generate a data signal based on an output of the plurality of fully connected layers. A transmitting unit sends the data signal to a peripheral or remote device.

Abstract: A system and method for anomaly detection in a networked control system can include: receiving information transmitted over a network at a device; parsing the information at the device to determine whether information specified as relevant to anomaly detection is contained therein, wherein if relevant information is identified, then extracting the relevant information, including, for example protocol information, and saving the relevant information in a first dataset, and if relevant information is not identified, saving the information in a second dataset; storing the first and second datasets on a memory to train a prediction model for anomaly detection; and monitoring network traffic using the prediction model for anomaly detection.

Abstract: An exemplary mobile apparatus for assessing an operating condition of an asset. The mobile device includes at least one sensor device for acquiring data related to one or more operational or environmental characteristics of an asset during operation. The mobile device further includes a processing device encoded with a neural network architecture having one or more models trained to identify one or more operating conditions of one or more assets according to asset type. The processing device can determine an operating condition of the asset by extracting features from the acquired data and comparing attributes of the extracted features to attributes of at least one known operating condition determined through training of the models. The mobile device can also include an output interface to output a result of the determination performed by the processing device, which can include communicating the result to a remote device.

Abstract: A method and system are disclosed for acquiring and processing data, the exemplary system includes: one or more intelligent devices connected in a dynamic ad hoc network as a network of edge devices which can optionally access a cloud storage, at least one intelligent device being configured with a software installation to selectively receive and execute analytics on data received; at least one of the intelligent devices being configured to identify data to be requested from at least one other edge device for enhancing analytics performed on the at least one intelligent device; and a switchboard for maintaining a current view of resources in the network, and functions for which each resource is tasked, the resources including the at least one intelligent device and those edge devices which can communicate with the at least one intelligent device on the network.

Abstract: A computer-implemented method for classifying an input data set within a data category using multiple data representation modes. The method includes identifying at least a first data representation source mode and a second data representation source mode; classifying the at least first data representation source mode via at least a first data recognition tool and the at least second data representation source mode via at least a second data recognition tool, the classifying including allocating a confidence factor for each data representation source mode in the data category; and combining outputs of the classifying into a single output confidence score by using a weighted fusion of the allocated confidence factors.

Abstract: A method of processing a radio frequency signal includes: receiving the radio frequency signal at an antenna of a receiver device; processing, by a radio frequency front-end device, the radio frequency signal; converting, by an analog-to-digital converter, the analog signal to a digital signal; receiving, by a neural network, the digital signal; and processing, by the neural network, the digital signal to produce an output.

Abstract: An exemplary device for classifying an image includes a receiving unit that receives image data. The device also includes a hardware processor including a neural network architecture to extract a plurality of features from the image data, filter each feature extracted from the image data, concatenate the plurality of filtered features to form an image vector, evaluate the plurality of concatenated features in first and second layers of a plurality of fully connected layers of the neural network architecture based on an amount of deviation in the features determined at each fully connected layer, and generate a data signal based on an output of the plurality of fully connected layers. A transmitting unit sends the data signal to a peripheral or remote device.

Abstract: A computer-implemented method and system are disclosed for classifying an input data set within a data category using multiple data recognition tools. The method includes identifying at least a first attribute and a second attribute of the data category; classifying the at least first attribute via at least a first data recognition tool and the at least second attribute via at least a second data recognition tool, the classifying including: allocating a confidence factor for each of the at least first and second attributes that indicates a presence of each of the at least first and second attributes in the input data set; and combining outputs of the classifying into a single output confidence score by using a weighted fusion of the allocated confidence factors.