Abstract: In an embodiment, a system for detection of trains at railroad crossings is provided. The system comprises a field-deployed detection and reporting device comprising a microphone, a communication module, a microprocessor, and an application. When executed on the microprocessor, the application receives data describing sounds captured by the microphone and identifies frequencies of a first received sound. The application also transmits, based on the identified frequencies and a formula, a first message via the communication module. The first received sound is generated by a warning bell sounded at the railroad crossing. The first message is received by a backend server that issues a first broadcast based on receipt of the first message. The application further determines that the first received sound discontinues. Based on the determination, the application sends a second message via the module to the backend server which issues a second broadcast that the crossing is reopened.

Abstract: A method and system for extracting text from a video stream is disclosed. The method may include determining at least one visual text region in each image of a plurality of images. The at least one visual text region may include a plurality of text characters and determining the at least one visual text region is based on analysis of one of lines and curves associated with each of the plurality of text characters, blob associated with the plurality of text characters along an axis, distribution of the plurality of text characters along a major axis, and a common attribute associated with the plurality of text characters. The method may further include segmenting the plurality of text characters from a background, and inpainting each of the plurality of text characters with a predefined color.

Abstract: A system for detection of trains at railroad crossings is provided comprising a field-deployed detection and reporting device. The device comprises a microphone, a communication module, a microprocessor, an application executing on the microprocessor. The application receives periodically generated samples of ambient sound captured by the microphone and performs Fast Fourier Transformation on data associated with the generated samples. The application also creates a spectrogram based on the output of the performed Fast Fourier Transformation, analyzes components of the spectrogram using at least a proprietary algorithm, and transmits, based on the analysis, a first message via the communication module. The device is situated proximate a railroad crossing. The ambient sound is generated by a warning bell sounded at the railroad crossing. The warning bell is sounded to indicate presence of a railroad train.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: In an embodiment, a system for detection of trains at railroad crossings is provided. The system comprises a field-deployed detection and reporting device comprising a microphone, a communication module, a microprocessor, and an application. When executed on the microprocessor, the application receives data describing sounds captured by the microphone and identifies frequencies of a first received sound. The application also transmits, based on the identified frequencies and a formula, a first message via the communication module. The first received sound is generated by a warning bell sounded at the railroad crossing. The first message is received by a backend server that issues a first broadcast based on receipt of the first message. The application further determines that the first received sound discontinues. Based on the determination, the application sends a second message via the module to the backend server which issues a second broadcast that the crossing is reopened.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

Abstract: A system, computer-readable medium, and a method including receiving flight data engine measurements for at least one engine of the specific aircraft asset; receiving flight data aerodynamics measurements for the specific aircraft asset; combining a physics based parametric aerodynamic performance model tuned for the specific aircraft asset using the flight data aerodynamics measurements and a physics based engine model tuned for the specific aircraft asset using the flight data engine measurements; calculating, based on the combined tuned aerodynamic performance model and the tuned engine model, a performance model for the specific aircraft asset as a whole; and storing a record of the calculated performance model for the specific aircraft asset for a future deployment.

Abstract: A system, computer-readable medium, and a method including receiving flight data engine measurements for at least one engine of the specific aircraft asset; receiving flight data aerodynamics measurements for the specific aircraft asset; combining a physics based parametric aerodynamic performance model tuned for the specific aircraft asset using the flight data aerodynamics measurements and a physics based engine model tuned for the specific aircraft asset using the flight data engine measurements; calculating, based on the combined tuned aerodynamic performance model and the tuned engine model, a performance model for the specific aircraft asset as a whole; and storing a record of the calculated performance model for the specific aircraft asset for a future deployment.