This document describes techniques, apparatuses, and systems for a grid-based road model with multiple layers. An example road-perception system generates a roadway grid representation that includes multiple cells. The road-perception system uses data from multiple information sources to generate a road model that includes multiple layers. Each layer represents a roadway attribute of each cell in the grid and includes one or more layer hypotheses. In this way, the described techniques and systems can provide an accurate and reliable road model and quantify uncertainty therein.

This document describes techniques, apparatuses, and systems for a grid-based road model with multiple layers. An example road-perception system generates a roadway grid representation that includes multiple cells. The road-perception system uses data from multiple information sources to generate a road model that includes multiple layers. Each layer represents a roadway attribute of each cell in the grid and includes one or more layer hypotheses. In this way, the described techniques and systems can provide an accurate and reliable road model and quantify uncertainty therein.

A method, a computing device, and a non-transitory machine-readable medium for classifying documents. A document collection is sorted into a plurality of categories. A classifier corresponding to a category of the plurality of categories is trained to output a probability that a document associated with the category is of a selected type (e.g., confidential). The training includes determining, by the processor, that a cardinality of a set of negative samples in a train set is not above a pipeline threshold but is at least one and training the classifier via a first pipeline and a second pipeline using a training group that includes a first portion of a group of positive samples in the train set, a second portion of a set of negative samples in the train set, and a third portion of a group of unlabeled samples in the train set

A training method for multi-output land cover classification model and a classification method are provided. The training method includes: obtaining a training data; inputting the training data into an initial model based on deep belief nets for training to obtain a multi-output land cover classification model, wherein the initial model includes N level outputs, and the N level outputs include an output set at last network layer and (N−1) level output set at any (N−1) network layers from a first network layer to a penultimate network layer of the initial model; determining a total loss according to losses of the N level outputs; performing a backpropagation based on the total loss to adjust a parameter of the initial model, N being an integer greater than or equal to 2. The gradient is not easy to disappear during backpropagation of the model, which is beneficial to improve classification accuracy.

This document describes techniques, apparatuses, and systems for a grid-based road model with multiple layers. An example road-perception system generates a roadway grid representation that includes multiple cells. The road-perception system uses data from multiple information sources to generate a road model that includes multiple layers. Each layer represents a roadway attribute of each cell in the grid and includes one or more layer hypotheses. In this way, the described techniques and systems can provide an accurate and reliable road model and quantify uncertainty therein.

This document describes techniques, apparatuses, and systems for a grid-based road model with multiple layers. An example road-perception system generates a roadway grid representation that includes multiple cells. The road-perception system uses data from multiple information sources to generate a road model that includes multiple layers. Each layer represents a roadway attribute of each cell in the grid and includes one or more layer hypotheses. In this way, the described techniques and systems can provide an accurate and reliable road model and quantify uncertainty therein.

A dynamic facial expression recognition (FER) method based on a Dempster-Shafer (DS) theory improves a feature extraction effect of an expression video through multi-feature fusion, and deeply learns an imbalanced dynamic expression feature by using the DS theory, multi-branch convolution, and an attention mechanism. Compared with other methods, the dynamic FER method scientifically and effectively reduces an impact of sample imbalance on expression recognition, fully utilizes a spatio-temporal feature to mine potential semantic information of the video expression to perform expression classification, thereby improving reliability and accuracy and meeting a demand for the expression recognition.

The present invention encompasses a physical or virtual, computational, analysis, fusion and correlation system that can automatically, systematically and independently analyze collected sensor data (upstream) aboard or streaming from aerial vehicles and/or other fixed or mobile single or multi-sensor platforms. The resultant data is fused and presented locally, remotely or at ground stations in near real time, as it is collected from local and/or remote sensors. The invention improves detection and reduces false detections compared to existing systems using portable apparatus or cloud based computation and capabilities designed to reduce the role of the human operator in the review, fusion and analysis of cross modality sensor data collected from ISR (Intelligence, Surveillance and Reconnaissance) aerial vehicles or other fixed and mobile ISR platforms. The invention replaces human sensor data analysts with hardware and software providing two significant advantages over the current manual methods.

The present invention encompasses a physical or virtual, computational, analysis, fusion and correlation system that can automatically, systematically and independently analyze collected sensor data (upstream) aboard or streaming from aerial vehicles and/or other fixed or mobile single or multi-sensor platforms. The resultant data is fused and presented locally, remotely or at ground stations in near real time, as it is collected from local and/or remote sensors. The invention improves detection and reduces false detections compared to existing systems using portable apparatus or cloud based computation and capabilities designed to reduce the role of the human operator in the review, fusion and analysis of cross modality sensor data collected from ISR (Intelligence, Surveillance and Reconnaissance) aerial vehicles or other fixed and mobile ISR platforms. The invention replaces human sensor data analysts with hardware and software providing two significant advantages over the current manual methods.

A training method for multi-output land cover classification model and a classification method are provided. The training method includes: obtaining a training data; inputting the training data into an initial model based on deep belief nets for training to obtain a multi-output land cover classification model, wherein the initial model includes N level outputs, and the N level outputs include an output set at last network layer and (N1) level output set at any (N1) network layers from a first network layer to a penultimate network layer of the initial model; determining a total loss according to losses of the N level outputs; performing a backpropagation based on the total loss to adjust a parameter of the initial model, N being an integer greater than or equal to 2. The gradient is not easy to disappear during backpropagation of the model, which is beneficial to improve classification accuracy.