Apparatuses, systems, and methods for detecting various actions of a vehicle operator

• US 9,996,757 B1
• Filed: 06/14/2016
• Issued: 06/12/2018
• Est. Priority Date: 06/14/2016
• Status: Active Grant

• Current Assignee: State Farm Mutual Automobile Insurance Company (State Farm)
• Original Assignee: State Farm Mutual Automobile Insurance Company (State Farm)
• Inventors: Chan, Aaron Scott, Sanchez, Kenneth J.
• Primary Examiner(s): Osifade, Idowu O
• Application Number:
  US15/181,628
• Time in Patent Office: 728 Days
• Field of Search: None
• CPC Class Codes: G06F 18/29 Graphical models, e.g. Bayesian networks
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G06V 20/597 Recognising the driver's state or behaviour, e.g. attention or drowsiness
• CPC Class Codes: G06V 40/197 Matching; Classification
• CPC Class Codes: G06V 40/28 Recognition of hand or arm movements, e.g. recognition of deaf sign language static hand signs G06V40/113

Abstract

Apparatuses, systems and methods are provided for detecting various actions of a vehicle operator. More particularly, apparatuses, systems and methods are provided for detecting various actions of a vehicle operator based on digital image data.

Claims

1. A device for detecting various vehicle operator actions by tracking vehicle occupant postures using image recognition, the device comprising:

at least one vehicle interior sensor for acquiring at least one image sequence of at least one vehicle occupant;

a memory for storing the current image data, vehicle occupant posture data relating to the current image data, a current image data receiving module, a previously classified interior data receiving module, a skeletal pose data generation module, and a vehicle operator action detection module; and

a processor that is configured to determine a vehicle operation mode by;

receiving previously classified image data from at least one previously classified image database in response to the processor executing a previously classified vehicle interior data receiving module, wherein the previously classified image data is representative of previously classified vehicle operator actions;

receiving current image data from the at least one vehicle interior sensor in response to the processor executing the current image data receiving module, wherein the current image data is representative of current vehicle operator action;

creating a vehicle occupant posture record for the at least one vehicle occupant comprising the vehicle interior data of the at least one vehicle occupant;

extracting three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant from the vehicle interior data, the three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant are stored in the vehicle occupant posture record as vehicle occupant posture data relating to the vehicle interior data;

recognizing and tracking postures of the at least one vehicle occupant in the image sequence as vehicle occupant posture data by processing the three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant and the previously classified interior data using the stored skeletal pose data generation model;

detecting a vehicle operator action based on a comparison of the current image data with the previously classified image data in response to the processor executing the vehicle operator action detection module that includes a probability function; and

automatically updating the vehicle occupant posture record with the vehicle posture data.

2. The device as in claim 1, wherein the at least one vehicle interior sensor is selected from:

at least one digital image sensor, at least one ultra-sonic sensor, at least one radar-sensor, at least one infrared light sensor, or at least one laser light sensor.

3. The device as in claim 1, wherein the vehicle operator action is detected using a probability function.

4. The device as in claim 1, wherein the previously classified image data is representative of a three-dimensional representation of at least one occupant within the vehicle interior.

5. The device as in claim 1, wherein the current image data is representative of a three-dimensional representation of at least one occupant within the vehicle interior.

6. The device as in claim 1, wherein the current image data includes images and/or extracted image features that are representative of a vehicle occupant using a cellular telephone, a vehicle occupant looking out a vehicle side window, a vehicle occupant adjusting a vehicle radio, a vehicle occupant adjusting a vehicle heating, ventilation and air conditioning system, two vehicle occupants talking with one-another, a vehicle occupant reading a book or magazine, a vehicle occupant putting on makeup, or a vehicle occupant looking at themselves in a mirror.

7. The device as in claim 1, wherein the previously classified image data includes images and/or extracted image features that have previously been classified as being representative of a vehicle occupant using a cellular telephone, a vehicle occupant looking out a vehicle side window, a vehicle occupant adjusting a vehicle radio, a vehicle occupant adjusting a vehicle heating, ventilation and air conditioning system, two vehicle occupants talking with one-another, a vehicle occupant reading a book or magazine, a vehicle occupant putting on makeup, or a vehicle occupant looking at themselves in a mirror.

8. A computer-implemented method for detecting various vehicle operator actions by tracking vehicle occupant postures using image recognition, the method comprising:

receiving, at a processor of a computing device, previously classified image data from at least one previously classified image database in response to the processor executing a previously classified image data receiving module, wherein the previously classified image data is representative of previously classified vehicle operator actions;

receiving, at a processor of a computing device, current image data from at least one vehicle interior sensor a current image data receiving module, in response to the processor executing a current image data receiving module, wherein the current image data is representative of current vehicle operator actions and at least one image sequence of at least one vehicle occupant;

creating a vehicle occupant posture record for the at least one vehicle occupant comprising vehicle interior data of the at least one vehicle occupant;

extracting three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant from the vehicle interior data, the three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant are stored in the vehicle occupant posture record as vehicle occupant posture data relating to the vehicle interior data;

recognizing and tracking postures of the at least one vehicle occupant in the image sequence as vehicle occupant posture data by processing the three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant and the previously classified interior data using the stored skeletal pose data generation model;

determining, using a processor of a computing device, a vehicle operator action, based on a comparison of the current image data with the previously classified image data, in response to the processor executing a vehicle operator action detection module in response to the processor executing a vehicle operator action detection module that includes a probability function; and

automatically updating the vehicle occupant posture record with the vehicle posture data.

9. The method as in claim 8, wherein the at least one vehicle interior sensor is selected from:

at least one digital image sensor, at least one ultra-sonic sensor, at least one radar-sensor, at least one infrared light sensor, or at least one laser light sensor.

10. The method as in claim 8, wherein the current image data is representative of a three-dimensional representation of at least one occupant within the vehicle interior.

11. The method as in claim 8, wherein the previously classified image data is representative of a three-dimensional representation of at least one occupant within the vehicle interior.

12. The method as in claim 8, wherein the vehicle operator action is detected using a probability function.

13. The method as in claim 8, wherein the current image data includes images and/or extracted image features that are representative of vehicle occupant locations/orientations, cellular telephone locations/orientations, vehicle occupant eye locations/orientations, vehicle occupant head location/orientation, vehicle occupant hand location/orientation, a vehicle occupant torso location/orientation, a seat belt location, or a vehicle seat location/orientation.

14. The method as in claim 8, wherein the previously classified image data includes images and/or extracted image features that have previously been classified as being representative of known vehicle occupant locations/orientations, known cellular telephone locations/orientations, known vehicle occupant eye locations/orientations, known vehicle occupant head location/orientation, known vehicle occupant hand location/orientation, a known vehicle occupant torso location/orientation, a known seat belt location, or a known vehicle seat location/orientation.

15. A non-transitory computer-readable medium storing computer-readable instructions that, when executed by a processor, cause the processor to detecting various vehicle operator actions by tracking vehicle occupant postures using image recognition, the non-transitory computer-readable medium comprising:

a previously classified image data receiving module that, when executed by a processor, causes the processor to receive previously classified image data from at least one previously classified image database, wherein the previously classified image data is representative of previously classified vehicle operator actions;

a current image data receiving module that, when executed by a processor, causes the processor to receive current image data from at least one vehicle interior sensor, wherein the current image data is representative of current vehicle operator actions and at least one image sequence of at least one vehicle occupant;

a vehicle occupant posture record creation module that, when executed by a processor, causes the processor to create a vehicle occupant posture record for the at least one vehicle occupant comprising the vehicle interior data of the at least one vehicle occupant;

a joint location extraction module that, when executed by a processor, causes the processor to extract three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant from the vehicle interior data, the three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant are stored in the vehicle occupant posture record as vehicle occupant posture data relating to the vehicle interior data;

a posture recognizing and tracking module that, when executed by a processor, causes the processor to recognize and track postures of the at least one vehicle occupant in the image sequence as vehicle occupant posture data by processing the three-dimensional coordinate spatial locations of at least one joint location of the at least one vehicle occupant and the previously classified interior data using the stored skeletal pose data generation model;

a vehicle operator action detection module including a probability function that, when executed by a processor, causes the processor to detect various vehicle operator actions based on a comparison of the current image data with the previously classified image data; and

a vehicle occupant posture record updating module that, when executed by a processor, causes the processor to automatically update the vehicle occupant posture record with the vehicle posture data.

16. The non-transitory computer-readable medium as in claim 15, wherein the vehicle operator action is detected using a probability function.

17. The non-transitory computer-readable medium as in claim 15, wherein the current image data is representative of a three-dimensional representation of at least one occupant within the vehicle interior.

18. The non-transitory computer-readable medium as in claim 15, wherein the current image data includes images and/or extracted image features that are representative of a vehicle occupant using a cellular telephone, a vehicle occupant looking out a vehicle side window, a vehicle occupant adjusting a vehicle radio, a vehicle occupant adjusting a vehicle heating, ventilation and air conditioning system, two vehicle occupants talking with one-another, a vehicle occupant reading a book or magazine, a vehicle occupant putting on makeup, or a vehicle occupant looking at themselves in a mirror, vehicle occupant locations/orientations, cellular telephone locations/orientations, vehicle occupant eye locations/orientations, vehicle occupant head location/orientation, vehicle occupant hand location/orientation, a vehicle occupant torso location/orientation, a seat belt location, or a vehicle seat location/orientation.

19. The non-transitory computer-readable medium as in claim 15, wherein the previously classified image data includes images and/or extracted image features that have previously been classified as being representative of a vehicle occupant using a cellular telephone, a vehicle occupant looking out a vehicle side window, a vehicle occupant adjusting a vehicle radio, a vehicle occupant adjusting a vehicle heating, ventilation and air conditioning system, two vehicle occupants talking with one-another, a vehicle occupant reading a book or magazine, a vehicle occupant putting on makeup, a vehicle occupant looking at themselves in a mirror, known vehicle occupant locations/orientations, known cellular telephone locations/orientations, known vehicle occupant eye locations/orientations, known vehicle occupant head location/orientation, known vehicle occupant hand location/orientation, a known vehicle occupant torso location/orientation, a known seat belt location, or a known vehicle seat location/orientation.

20. The non-transitory computer-readable medium as in claim 15, wherein the previously classified image data is representative of a three-dimensional representation of at least one occupant within the vehicle interior.