Back to results
US 12,111,865 B2Granted
Video analysis for efficient sorting of event data
AssigneeLytx, Inc.
InventorsBrett Hoye, Stephen Krotosky
Life of this patent
Estimating the term…
Abstract
A system for sorting event data includes an interface and a processor. The interface is configured to receive a small set of video frame data from a vehicle event recorder. The processor is configured to determine that the small set of video frame data indicates to fetch more vehicle event recorder data; determine a categorization based at least in part on the more vehicle event recorder data; and in response to the categorization, route the more vehicle event recorder data to a reviewing queue.
Claims18
A system for sorting event data, comprising:
- an interface configured to:
- receive a small set of video frame data from a vehicle event recorder;
- and
- a processor configured to:
- determine that the small set of video frame data indicates to fetch more vehicle event recorder data, comprises to:
- determine that the small set of video frame data indicates that an anomalous event has occurred using a model, wherein the model implements a machine learning technique, wherein the machine learning technique includes machine vision, and wherein the determining that the small set of video frame data indicates that the anomalous event has occurred comprises to:
- determine, using the model, that the small set of video frame data indicates that a trash bin is overflowing;
- and
- in response a determination that the small set of video frame data indicates that the anomalous event has occurred, fetch the more vehicle event recorder data, comprising to:
- determine that a score associated with the model falls between a first threshold and a second threshold, wherein in the event the score falls below the first threshold, determine that the anomalous event has not occurred and omit fetching the more vehicle event recorder data, wherein in the event the score falls above the second threshold, determine that the anomalous event has occurred and omit fetching the more vehicle event recorder data;
- and
- in the event that the score falls between the first threshold and the second threshold, fetch the more vehicle event recorder data;
- determine a categorization based at least in part on the more vehicle event recorder data, wherein the categorization includes human review;
- and
- in response to the categorization, route the more vehicle event recorder data to a reviewing queue to determine whether the trash bin is overflowing.
12 more elements
A method for sorting event data, comprising:
- receiving a small set of video frame data from a vehicle event recorder;
- determining, using a processor, that the small set of video frame data indicates to fetch more vehicle event recorder data, comprises:
- determining that the small set of video frame data indicates that an anomalous event has occurred using a model, wherein the model implements a machine learning technique, wherein the machine learning technique includes machine vision, and wherein the determining that the small set of video frame data indicates that the anomalous event has occurred comprises:
- determining, using the model, that the small set of video frame data indicates that a trash bin is overflowing;
- and
- in response a determination that the small set of video frame data indicates that the anomalous event has occurred, fetching the more vehicle event recorder data, comprising:
- determining that a score associated with the model falls between a first threshold and a second threshold, wherein in the event the score falls below the first threshold, determine that the anomalous event has not occurred and omit fetching the more vehicle event recorder data, wherein in the event the score falls above the second threshold, determine that the anomalous event has occurred and omit fetching the more vehicle event recorder data;
- and
- in the event that the score falls between the first threshold and the second threshold, fetching the more vehicle event recorder data;
- determining a categorization based at least in part on the more vehicle event recorder data, wherein the categorization includes human review;
- and
- in response to the categorization, routing the more vehicle event recorder data to a reviewing queue to determine whether the trash bin is overflowing.
9 more elements
A computer program product for sorting event data, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
- receiving a small set of video frame data from a vehicle event recorder;
- determining that the small set of video frame data indicates to fetch more vehicle event recorder data, comprises:
- determining that the small set of video frame data indicates that an anomalous event has occurred using a model, wherein the model implements a machine learning technique, wherein the machine learning technique includes a deep machine vision, and wherein the determining that the small set of video frame data indicates that the anomalous event has occurred comprises:
- determining, using the model, that the small set of video frame data indicates that a trash bin is overflowing;
- and
- in response a determination that the small set of video frame data indicates that the anomalous event has occurred, fetching the more vehicle event recorder data, comprising:
- determining that a score associated with the model falls between a first threshold and a second threshold, wherein in the event the score falls below the first threshold, determine that the anomalous event has not occurred and omit fetching the more vehicle event recorder data, wherein in the event the score falls above the second threshold, determine that the anomalous event has occurred and omit fetching the more vehicle event recorder data;
- and
- in the event that the score falls between the first threshold and the second threshold, fetching the more vehicle event recorder data;
- determining a categorization based at least in part on the more vehicle event recorder data, wherein the categorization includes human review;
- and
- in response to the categorization, routing the more vehicle event recorder data to a reviewing queue to determine whether the trash bin is overflowing.
9 more elements
Dependencies are inferred from the claim text.