Dash cam with artificial intelligence safety event detection
A vehicle dash cam may be configured to execute one or more neural networks (and/or other artificial intelligence), such as based on input from one or more of the cameras and/or other sensors associated with the dash cam, to intelligently detect safety events in real-time. Detection of a safety event may trigger an in-cab alert to make the driver aware of the safety risk. The dash cam may include logic for determining which asset data to transmit to a backend server in response to detection of a safety event, as well as which asset data to transmit to the backend server in response to analysis of sensor data that did not trigger a safety event. The asset data transmitted to the backend server may be further analyzed to determine if further alerts should be provided to the driver and/or to a safety manager.
1 . A computerized method, performed by a computing system having one or more hardware computer processors and one or more non-transitory computer readable storage device storing software instructions executable by the computing system to perform the computerized method comprising:
accessing first images of a first area in front of a vehicle acquired by an outward-facing camera associated with the vehicle;
accessing second images of a second area inside the vehicle acquired by an inward-facing camera associated with the vehicle;
evaluating at least some of the accessed first images with reference to first safety event criteria associated with a first safety event type;
evaluating at least some of the accessed second images with reference to second safety event criteria associated with a second safety event type;
triggering a first safety event in response to determining that first features of the at least some of the accessed first images match the first safety event criteria associated with the first safety event type;
triggering a second safety event in response to determining that second features of the at least some of the accessed second images match the second safety event criteria associated with the second safety event type;
accessing an asset data structure including associations between safety event types and respective asset data types, wherein a first asset data type associated with the first safety event type is different than a second asset data type associated with the second safety event type;
in response to triggering the first safety event, transmitting first asset data of the first asset data type to a remote event analysis system; and
in response to triggering the second safety event, transmitting second asset data of the second asset data type to the remote event analysis system.
2 . The computerized method of claim 1 , wherein the first safety event indicates one or more of: a forward collision warning or a tailgating warning.
3 . The computerized method of claim 1 , wherein the first features include one or more of: identification of an object in front of the vehicle, identification of other objects in the first images, metadata associated with identified objects, accelerometer data, global positioning system (GPS) data, ECU data, and vehicle speed data.
4 . The computerized method of claim 1 , wherein the first features include a distance to a lead vehicle or object in front of the vehicle.
5 . The computerized method of claim 1 , wherein the first asset data type includes one or more of accelerometer data, global positioning system (GPS) data, ECU data, or vehicle speed data.
6 . The computerized method of claim 1 , wherein the second safety event indicates one or more of: a distracted driver warning, or a policy violation.
7 . The computerized method of claim 1 , wherein the second features include one or more of: an eye pose of a driver, a head pose of the driver, or identification of an object.
8 . The computerized method of claim 1 , wherein the second features include a head pose of a driver.
9 . The computerized method of claim 8 , wherein the head pose of the driver indicates how far away from straight ahead the driver's head is positioned.
10 . The computerized method of claim 9 , wherein a default head pose is customized for the driver.
11 . The computerized method of claim 1 , further comprising:
providing an in-cab alert in response to triggering the first safety event.
12 . The computerized method of claim 1 , wherein the first safety event criteria and the second safety event criteria are user-configurable.
13 . The computerized method of claim 1 , further comprising:
storing the first images and the second images in a local data store associated with the vehicle.
14 . The computerized method of claim 13 , further comprising:
deleting the first images and the second images from the local data store after a predetermined time period.
15 . The computerized method of claim 1 , wherein the first asset data type includes video data from the outward-facing camera.
16 . The computerized method of claim 1 , wherein the second asset data type includes video data from the inward-facing camera.
17 . The computerized method of claim 1 , wherein the computing system is mounted inside the vehicle.
18 . A non-transitory computer-readable medium storing software instructions that are executable by a hardware computer processor of a vehicle device to perform operations comprising:
accessing first images of a first area in front of a vehicle acquired by an outward-facing camera of the vehicle device;
accessing second images of a second area inside the vehicle acquired by an inward-facing camera of the vehicle device;
evaluating at least some of the accessed first images with reference to first safety event criteria associated with a first safety event type;
evaluating at least some of the accessed second images with reference to second safety event criteria associated with a second safety event type;
triggering a first safety event in response to determining that the at least some of the accessed first images match the first safety event criteria;
triggering a second safety event in response to determining that the at least some of the accessed second images match the second safety event criteria;
in response to triggering the first safety event, transmitting first asset data of a first asset data type to a remote event analysis system; and
in response to triggering the second safety event, transmitting second asset data of a second asset data type to the remote event analysis system, wherein the first asset data type is different than the second asset data type.
19 . The non-transitory computer-readable medium of claim 18 , wherein the first safety event type is one of: a forward collision warning or a tailgating warning, and the second safety event type is one of: a distracted driver or a policy violation.
20 . The non-transitory computer-readable medium of claim 18 , wherein the first asset data type is video data from the outward-facing camera and the second asset data type is video data from the inward-facing camera.