Applying image analytics and machine learning to lock systems in hotels

1. A method of using image analytics and machine learning in a lock system, the method comprising:

receiving data describing access actions, the access actions performed at an access control device based at least in part on a credential, and the data comprising, for each access action, a description of the access action, a timestamp, and an image of a person presenting the credential;

analyzing the data to identify patterns of access, the patterns including a day of week;

identifying facial characteristics of the person presenting the credential;

associating the credential with the facial characteristics;

detecting, based at least in part on facial recognition and the facial characteristics, that the person is proximate to the access control device; and

performing an access action at the access control device based on the detecting and the patterns of access, wherein the facial recognition is used in place of the credential to provide authorization to perform the access action in response to a current day of the week being within a threshold of the day of week of the patterns.

2. The method of claim 1 , wherein the patterns further include a time of day and the performing is further based on a current time of day being within a threshold of the time of day in the patterns.

3. The method of claim 1 , wherein facial characteristics of a plurality of people are identified as presenting the credential and the credential is associated with the plurality of people.

4. The method of claim 1 , wherein the access action is unlock or lock.

5. The method of claim 1 , wherein the access action is enter security mode or enter office mode.

6. The method of claim 1 , wherein additional access actions are performed subsequent to the performing as long as a confidence level of the associating is above a threshold.

7. The method of claim 1 , wherein the facial characteristics are updated based on images of the person received after the associating.

8. A system configured to use image analytics and machine learning in a lock system, the system comprising:

a processor; and

a memory comprising computer-executable instructions that, when executed by the processor, cause the processor to perform operations, the operations comprising:

receiving data describing access actions, the access actions performed at an access control device based at least in part on a credential, and the data comprising, for each access action, a description of the access action, a timestamp, and an image of a person presenting the credential;

analyzing the data to identify patterns of access, the patterns including a day of week;

identifying facial characteristics of the person presenting the credential;

associating the credential with the facial characteristics;

detecting, based at least in part on facial recognition and the facial characteristics, that the person is proximate to the access control device; and

performing an access action at the access control device based on the detecting and the patterns of access, wherein the facial recognition is used in place of the credential to provide authorization to perform the access action in response to a current day of the week being within a threshold of the day of week of the patterns.

9. The system of claim 8 , wherein the patterns further include a time of day and the performing is further based on a current time of day being within a threshold of the time of day in the patterns.

10. The system of claim 8 , wherein facial characteristics of a plurality of people are identified as presenting the credential and the credential is associated with the plurality of people.

11. The system of claim 8 , wherein the access action is unlock or lock.

12. The system of claim 8 , wherein the access action is enter security mode or enter office mode.

13. The system of claim 8 , wherein additional access actions are performed subsequent to the performing as long as a confidence level of the associating is above a threshold.

14. The system of claim 8 , wherein the facial characteristics are updated based on images of the person received after the associating.

15. A method of using image analytics and machine learning in a two-step authentication lock system, the method comprising:

receiving data describing access actions, the access actions performed at an access control device based at least in part on a credential, and the data comprising, for each access action, a description of the access action, a timestamp, and an image of an identifier of a person presenting the credential;

analyzing the data to identify patterns of access, the patterns including a day of week;

identifying characteristics of the identifier of a person presenting the credential;

associating the credential with the characteristics of the identifier;

detecting, based at least in part on image recognition and the characteristics of the identifier, that the person is proximate to the access control device;

based at least in part on the detecting, validating a personal identification number (PIN) entered at a keyboard of the access control device with an expected PIN of the person; and

performing an access action at the access control device based on the expected PIN of the person being the same as the PIN entered at the keyboard of the access control device and in response to a current day of the week being within a threshold of the day of week of the patterns.

16. The method of claim 15 , wherein the characteristics include facial characteristics.

17. The method of claim 15 , wherein the characteristics include license plate characteristics.

18. The method of claim 15 , wherein the access action is unlock.