Abstract: Described herein are methods and systems for detecting electrical discharges that precede electrical fires in electrical wiring. One or more sensor devices coupled to a circuit detect one or more signal waveforms generated by electrical activity on the circuit. The sensor devices identify one or more transient signals within the one or more signal waveforms, and generate one or more transient characteristics based upon the identified transient signals. A server communicably coupled to the sensor devices receives the one or more transient characteristics. The server analyzes the one or more transient characteristics to identify one or more electrical discharge indications. The server generates one or more alert signals when one or more electrical discharge indications are identified.

Abstract: System, methods, and devices for measuring the performance of a component of a communications network recently installed by a technician using proactive network maintenance parameters, and further assessing the installation performed by the technician.

Abstract: Described herein are methods and systems for detecting electrical discharges that precede electrical fires in electrical wiring. One or more sensor devices coupled to a circuit detect one or more signal waveforms generated by electrical activity on the circuit. The sensor devices identify one or more transient signals within the one or more signal waveforms, and generate one or more transient characteristics based upon the identified transient signals. A server communicably coupled to the sensor devices receives the one or more transient characteristics. The server analyzes the one or more transient characteristics to identify one or more electrical discharge indications. The server generates one or more alert signals when one or more electrical discharge indications are identified.

Abstract: System, methods, and devices for measuring the performance of a component of a communications network recently installed by a technician using proactive network maintenance parameters, and further assessing the installation performed by the technician.

Abstract: Disclosed is an approach of on-device partial recognition that includes performing partial recognition on an image of a document captured by a mobile device to detect and/or recognize a specific area (e.g., barcodes, non-relevant text, etc.) and filling the recognized area with a solid color. Because the solid color area has a maximum compression ratio, this approach can lead to image size reduction and increased network throughput for client-server based data recognition where further processing such as advanced data extraction is performed at the server side. The approach can be enforced with neural network algorithms to exclude non-relevant information (e.g., logos, phrases, words, etc.).

Abstract: An auto-rotation module having a single-layer neural network on a user device can convert a document image to a monochrome image having black and white pixels and segment the monochrome image into bounding boxes, each bounding box defining a connected segment of black pixels in the monochrome image. The auto-rotation module can determine textual snippets from the bounding boxes and prepare them into input images for the single-layer neural network. The single-layer neural network is trained to process each input image, recognize a correct orientation, and output a set of results for each input image. Each result indicates a probability associated with a particular orientation. The auto-rotation module can examine the results, determine what degree of rotation is needed to achieve a correct orientation of the document image, and automatically rotate the document image by the degree of rotation needed to achieve the correct orientation of the document image.

Abstract: Described herein are methods and systems for detection and notification of electrical power outages and power quality. A sensor coupled to a circuit transmits a keepalive packet to a server. The sensor detects an input signal generated by electrical activity. The sensor generates an output signal based upon the input signal. The sensor monitors the output signal. During a clock cycle, the sensor determines whether a rising edge occurred and transmits a fault packet to the server when the rising edge occurred prior to a predetermined clock value or when no rising edge occurred. The server receives the fault packet from the sensor and listens for keepalive packets. The server transmits a power outage notification when no keepalive packets are received for at least a defined time period after the fault packet is received. The server transmits a power restoration notification when one or more keepalive packets are subsequently received.

Abstract: An auto-rotation module having a single-layer neural network on a user device can convert a document image to a monochrome image having black and white pixels and segment the monochrome image into bounding boxes, each bounding box defining a connected segment of black pixels in the monochrome image. The auto-rotation module can determine textual snippets from the bounding boxes and prepare them into input images for the single-layer neural network. The single-layer neural network is trained to process each input image, recognize a correct orientation, and output a set of results for each input image. Each result indicates a probability associated with a particular orientation. The auto-rotation module can examine the results, determine what degree of rotation is needed to achieve a correct orientation of the document image, and automatically rotate the document image by the degree of rotation needed to achieve the correct orientation of the document image.

Abstract: Disclosed is an approach of on-device partial recognition that includes performing partial recognition on an image of a document captured by a mobile device to detect and/or recognize a specific area (e.g., barcodes, non-relevant text, etc.) and filling the recognized area with a solid color. Because the solid color area has a maximum compression ratio, this approach can lead to image size reduction and increased network throughput for client-server based data recognition where further processing such as advanced data extraction is performed at the server side. The approach can be enforced with neural network algorithms to exclude non-relevant information (e.g., logos, phrases, words, etc.).

Abstract: Systems, methods, and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: Compression set downhole clutch (100) can include an upper body (201), a lower body (202), and a hydraulic member (205). The upper body (201) can include at least one engagement surface (204). The lower body (202) can include at least one engagement surface (203). The at least one engagement surface (204) of the upper body (201) and the at least one engagement surface (203) of the lower body (202) can each be configured to couple with one another and rotate in unison in an engaged configuration (102). The hydraulic member (205) can be configured to be responsive to a predetermined pressure; the hydraulic member (205) can also be configured to disengage the engagement surface (204) of the upper body (201) from the engagement surface (203) of the lower body (202).

Abstract: The present invention teaches a punch and die measurement device which includes a linear array LED micrometer arranged so that the light beams and shadows pass vertically from transmitters to receivers, and further teaches a 1:1 ratio pantograph for transferring measurements from inside of a necker die to outside the die where it may then be measured by the linear array LED micrometers. The invention may include a precision alignment block, captive adjustment screws to adjust tilt plates angle of the linear array measurement carriage and lever arms to reduce motion of the adjustment screws as applied to the tilt plates.

Abstract: The invention relates to a positive drive spiral conveyor which includes, a drive tower rotatable about a vertical axis and a plurality of drive members extending in length from a bottom to a top of the drive tower. The drive members are spaced radially around the drive tower, with each drive member having a projecting driving ridge extending in length along at least a section of each drive member with a projecting shaft positioned on each drive member proximate an end of the driving ridge thereby defining an engagement zone. The projecting shaft defines a guiding surface around the shaft, such that in use a positive drive protrusion of a conveyor belt engages the guiding surface of the projecting shaft and is guided towards a leading side of the driving ridge.

Abstract: Systems, methods and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: Methods, systems, and computer program products are described herein for generating proxy reflectivity data associated with lightning activity. A computing device receives data associated with lightning activity for one or more geographical areas. The computing device determines the logarithmic scale of the combined in-cloud and cloud-to-ground flash rate (dBR) for one or more grids of lightning activity based on the received data. The computing device generates proxy reflectivity data based on the dBR of the grids, a rate of change of the dBR, and the geographical area of the lightning activity.

Abstract: System, methods, and devices for measuring the performance of a component of a communications network recently installed by a technician using proactive network maintenance parameters, and further assessing the installation performed by the technician.

Abstract: Disclosed is an approach of on-device partial recognition that includes performing partial recognition on an image of a document captured by a mobile device to detect and/or recognize a specific area (e.g., barcodes, non-relevant text, etc.) and filling the recognized area with a solid color. Because the solid color area has a maximum compression ratio, this approach can lead to image size reduction and increased network throughput for client-server based data recognition where further processing such as advanced data extraction is performed at the server side. The approach can be enforced with neural network algorithms to exclude non-relevant information (e.g., logos, phrases, words, etc.).

Abstract: Systems and methods for image modification to increase contrast between text and non-text pixels within the image. In one embodiment, an original document image is scaled to a predetermined size for processing by a convolutional neural network. The convolutional neural network identifies a probability that each pixel in the scaled is text and generates a heat map of these probabilities. The heat map is then scaled back to the size of the original document image, and the probabilities in the heat map are used to adjust the intensities of the text and non-text pixels. For positive text, intensities of text pixels are reduced and intensities of non-text pixels are increased in order to increase the contrast of the text against the background of the image. Optical character recognition may then be performed on the contrast-adjusted image.

Abstract: Systems and methods for image modification to increase contrast between text and non-text pixels within the image. In one embodiment, an original document image is scaled to a predetermined size for processing by a convolutional neural network. The convolutional neural network identifies a probability that each pixel in the scaled is text and generates a heat map of these probabilities. The heat map is then scaled back to the size of the original document image, and the probabilities in the heat map are used to adjust the intensities of the text and non-text pixels. For positive text, intensities of text pixels are reduced and intensities of non-text pixels are increased in order to increase the contrast of the text against the background of the image. Optical character recognition may then be performed on the contrast-adjusted image.

Abstract: Disclosed is an approach of on-device partial recognition that includes performing partial recognition on an image of a document captured by a mobile device to detect and/or recognize a specific area (e.g., barcodes, non-relevant text, etc.) and filling the recognized area with a solid color. Because the solid color area has a maximum compression ratio, this approach can lead to image size reduction and increased network throughput for client-server based data recognition where further processing such as advanced data extraction is performed at the server side. The approach can be enforced with neural network algorithms to exclude non-relevant information (e.g., logos, phrases, words, etc.).