Back to results
US 12,762,439 B2Granted
Semantic compression in a sensor system
AssigneeCisco Technology, Inc.
InventorsHugo Latapie, Ozkan Kilic, Adam James Lawrence, Gaowen Liu
Life of this patent
Estimating the term…
Abstract
In one embodiment, a device makes an inference about an event indicated by sensor data from a plurality of sources by applying a semantic reasoning engine to the sensor data. The device receives a selected semantic compression level from a user interface. The device selects a subset of the sensor data based on the inference and on the selected semantic compression level. The device exports the subset of the sensor data and the inference made by the semantic reasoning engine about the event.
Claims20
A method comprising:
- receiving, by a device, sensor data from a plurality of sensors monitoring conditions on board a ship, the sensor data comprising sonar data and at least one of video data, audio data, or network telemetry data;
- making, by a device, an inference about an event indicated by the sensor data by applying a semantic reasoning engine to the sensor data, wherein the sensor data comprises time series data, further wherein the semantic reasoning engine uses a hierarchical knowledge graph, comprising multiple layers ranging from concepts at a highest layer and raw sensor measurements from the plurality of sensors at a lowest layer, to make the inference about the event;
- receiving, at the device and from a user interface, a selected date or time range, a selected semantic compression level, and an indication of which types of the sensor data are eligible for semantic compression that narrows the sensor data by time and type;
- selecting, by the device and based on the inference, the selected semantic compression level, and the indication of which types of the sensor data are eligible for semantic compression, a subset of the sensor data associated with the event and within the selected date or time range, wherein the subset of the sensor data comprises the sonar data and at least one of the video data, audio data, or network telemetry data;
- and
- exporting, from the device, the subset of the sensor data and the inference made by the semantic reasoning engine about the event.
3 more elements
An apparatus, comprising:
- a network interface to communicate with a computer network;
- a processor coupled to the network interface and configured to execute one or more processes;
- and
- a memory configured to store a process that is executed by the processor, the process when executed configured to:
- make an inference about an event indicated by sensor data from a plurality of sources having differing sensing modalities monitoring conditions in an environment by applying a semantic reasoning engine to the sensor data, wherein the sensor data from the plurality of sources comprises time series data, further wherein the semantic reasoning engine uses a hierarchical knowledge graph, comprising multiple layers ranging from concepts at a highest layer and raw sensor measurements from a plurality of sources at a lowest layer, to make the inference about the event;
- determine that the inference is that the event is of an unknown type;
- receive, from a user interface, a selected semantic compression level and an indication of which types of the sensor data are eligible for semantic compression;
- select, based on the inference, the selected semantic compression level, and the indication of which types of the sensor data are eligible for semantic compression, a subset of sensor data surrounding the event, the subset comprising a semantically compressed subset of the sensor data surrounding the event;
- and
- provide, to a user interface, the semantically compressed subset of the sensor data for labeling of the unknown type of event.
7 more elements
A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:
- making, by the device, an inference about an event indicated by sensor data from a plurality of sources by applying a semantic reasoning engine to the sensor data, wherein the sensor data from the plurality of sources comprises time series data, further wherein the semantic reasoning engine uses a hierarchical knowledge graph, comprising multiple layers ranging from concepts at a highest layer and raw sensor measurements from a plurality of sources at a lowest layer, to make the inference about the event;
- receiving, at the device and from a user interface, a selected semantic compression level, a selected date or time range within which the sensor data was collected, and an indication of which types of the sensor data are eligible for semantic compression, that narrows the sensor data from the plurality of sources by time and type;
- selecting, by the device and based on the inference, the selected semantic compression level, the selected date or time range, and the indication of which types of the sensor data are eligible for semantic compression, a subset of the sensor data associated with the event and collected within the date or time range, wherein the subset of the sensor data comprises a representative frame or clip selected from a larger video when the inference indicates the event is associated with benign conditions and comprises a greater amount of event-associated sensor data when the inference indicates the event is associated with a hazardous condition;
- and
- exporting, from the device, the subset of the sensor data and the inference made by the semantic reasoning engine about the event.
2 more elements
Dependencies are inferred from the claim text.