Entity detection

• US 9,646,210 B2
• Filed: 12/04/2014
• Issued: 05/09/2017
• Est. Priority Date: 12/06/2013
• Status: Active Grant

• Current Assignee: Vivint Incorporated (NRG Energy, Inc.)
• Original Assignee: Vivint Incorporated (NRG Energy, Inc.)
• Inventors: Nye, James Ellis
• Primary Examiner(s): AZARIAN, SEYED H
• Application Number:
  US14/560,585
• Publication Number:
  US 20150161451A1
• Time in Patent Office: 887 Days
• Field of Search: 382100, 382103-107, 382155, 382162, 382168, 382173, 382181, 382189, 382199, 382209, 382219, 382224, 382232, 382254, 382274, 382276, 382286-291, 382305, 382312, 348148, 348143, 340 101
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

A computer-implemented method for entity detection is described. In one embodiment, an entity passing through a perimeter of a predefined area is detected via a camera. Upon detecting the entity passing through the perimeter of the predefined area, a type of the entity is classified from an image of the entity captured by the camera. Upon classifying the type of the entity, a feature of the entity is detected from the image of the entity. An identifier is assigned to the entity based on the type and the detected feature of the entity. The identifier distinguishes the entity from another entity of a same type.

Claims

1. A computer-implemented method for entity detection, the method comprising:

detecting, by a processor, an entity passing through a perimeter of a predefined area based at least in part on camera data;

upon detecting the entity passing through the perimeter of the predefined area, classifying, via the processor, a type of the entity from an image of the entity captured by a camera;

upon classifying the type of the entity, detecting, via the processor, a feature of the entity from the image of the entity;

recognizing the entity, based at least in part on the detected feature of the entity;

determining whether the recognized entity was previously assigned an identifier;

assigning, via the processor, the identifier to the entity based at least in part on the type and the detected feature of the entity, the identifier distinguishing the entity from another entity of a same type;

determining an average length of stay of the entity at the predefined area based at least in part on the determined length of stay of the entity and a previously determined length of stay determined from a previous visit by the entity to the predefined area;

identifying the type of the entity as a vehicle, the predefined area being a parking lot;

identifying a type of a second entity exiting the vehicle as a human;

tracking a location of the second entity; and

determining whether the second entity enters a second predefined area.

2. A computer-implemented method for entity detection, the method comprising:

detecting, by a processor, an entity passing through a perimeter of a predefined area based at least in part on camera data;

upon detecting the entity passing through the perimeter of the predefined area, classifying, via the processor, a type of the entity from an image of the entity captured by a camera;

upon classifying the type of the entity, detecting, via the processor, a feature of the entity from the image of the entity;

recognizing the entity, based at least in part on the detected feature of the entity;

determining whether the recognized entity was previously assigned an identifier;

assigning, via the processor, the identifier to the entity based at least in part on the type and the detected feature of the entity, the identifier distinguishing the entity from another entity of a same type;

upon determining the entity was previously assigned an identifier, determining a number of previous visits the entity made to the predefined area;

identifying the type of the entity as a vehicle, the predefined area being a parking lot;

identifying a type of a second entity exiting the vehicle as a human;

tracking a location of the second entity; and

determining whether the second entity enters a second predefined area.

3. The method of claim 2, further comprising:

determining a number of entities that are of the same type as the entity that are located in the predefined area.

4. The method of claim 2, further comprising:

determining a length of stay of the entity at the predefined area based at least in part on a detected time of arrival of the entity at the predefined area and a detected time of departure of the entity from the predefined area.

5. The method of claim 4, further comprising:

associating, with the identifier of the entity, a transaction by the entity at the predefined area.

6. The method of claim 5, further comprising:

determining an average value of transactions from the transaction and previous transactions associated with the identifier of the entity.

7. The method of claim 2, further comprising:

upon determining the number of previous visits the entity made to the predefined area, determining an average rate of visits over a predetermined time period.

8. The method of claim 2, further comprising:

generating a notification in response to detecting the entity.

9. A computing device configured for entity detection, comprising:

a processor;

memory in electronic communication with the processor;

instructions stored in the memory, the instructions being executable by the processor to;

detect, via camera data, an entity passing through a perimeter of a predefined area;

upon detecting the entity passing through the perimeter of the predefined area, classify a type of the entity from an image of the entity captured by a camera;

upon classifying the type of the entity, detect a feature of the entity from the image of the entity;

recognize the entity, based at least in part on the detected feature of the entity;

determine that the recognized entity was previously assigned an identifier;

assign the identifier to the entity based at least in part on the type and the detected feature of the entity, the identifier distinguishing the entity from another entity of a same type;

upon determining the entity was previously assigned the identifier,

determine a number of previous visits the entity made to the predefined area;

identify the type of the entity as a vehicle, the predefined area being a parking lot;

identify a type of a second entity exiting the vehicle as a human;

track a location of the second entity; and

determine whether the second entity enters a second predefined area.

10. The computing device of claim 9, wherein the instructions are executable by the processor to:

determine a number of entities that are of the same type as the entity that are located in the predefined area.

11. The computing device of claim 9, wherein the instructions are executable by the processor to:

determine a length of stay of the entity at the predefined area based at least in part on a detected time of arrival of the entity at the predefined area and a detected time of departure of the entity from the predefined area.