Monitoring inmate movement with facial recognition
The disclosure relates to systems and methods for automating offender documentation with facial or other recognition technique. A method can include generating, by cameras situated about a correctional facility, pixel data including first pixels corresponding to a face of a first offender in the correctional facility and an environment around the first offender corresponding to a location within the field of view of the camera, and providing, to a server, the pixel data from the cameras through a network connection, and receiving, from the server, an alert that, based on facial recognition and location detection analysis performed on the pixel data by the server, indicates a monitoring rule associated with the first offender is violated, an offender identification associated with the first offender, a location of the first offender in the correctional facility, and an indication of the monitoring rule violated.
1. A system comprising:
cameras situated about a correctional facility to generate pixel data including first pixels corresponding to a face of a first offender in the correctional facility and an environment around the first offender corresponding to a location within the field of view of the camera;
a device through which a user defines progressive validation of a scheduled or non-scheduled movement of the first offender in the correctional facility, the progressive validation including data indicating an origin of the first offender, a destination of the first offender, a path from the origin to the destination the first offender is allowed to take, and a camera of the cameras between the origin and the destination and along the path that is to provide pixel data; and
a server configured to receive the pixel data from the cameras through a network connection, transmit the received data to a recognition service through the network, and receive an alert from the alert service that, based on facial recognition and location detection analysis performed on the pixel data by the recognition service, indicates the progressive validation associated with the first offender is violated by the first offender not being detected by the camera, an offender identification associated with the first offender, a location of the first offender in the correctional facility, and an indication of the progressive validation definition violated.
2. The system of claim 1 , wherein the pixel data includes second pixels corresponding to a face of a second offender and a monitoring rule includes data identifying the first offender and the second offender and that the first offender and the second offender are not permitted to associate with each other.
3. The system of claim 1 , wherein the progressive validation includes data identifying a location that the first offender is not allowed to occupy, the location identified by an object in the pixel data or a location associated with the camera of the cameras that produced the pixel data.
4. The system of claim 1 , wherein the recognition service determines a direction of the offender based on the pixel data and the progressive validation indicates a direction along the path that the first offender is allowed to travel.
5. The system of claim 4 , wherein the pixel data indicates a time and the progressive validation indicates a range of times in which the first offender is to be in the field of the view of respective cameras of the origin, destination, and the camera between the origin and the destination.
6. The system of claim 1 , wherein the progressive validation further includes data defining that the first offender is required to be in good standing to be allowed on the path without a correctional facility officer.
7. The system of claim 1 , wherein the progressive validation further includes data defining that a correctional facility officer and the first offender are to be within a field of view of the camera at the origin simultaneously before the first offender is allowed to move towards the destination.
8. The system of claim 1 , wherein the recognition service is to determine that the likelihood an entity is associated with the face is less than a threshold and the alert indicates that the entity is to be further verified by an officer.
9. The system of claim 5 , wherein the camera between the origin and the destination is configured to only provide pixel data within a window of time after the offender has left the origin.
10. A method for inmate monitoring comprising:
generating, by cameras situated about a correctional facility, pixel data including first pixels corresponding to a face of a first offender in the correctional facility and an environment around the first offender corresponding to a location within the field of view of the camera; and
providing, to a server, the pixel data from the cameras through a network connection;
providing data of a user-defined progressive validation of a scheduled or non-scheduled movement of the first offender in the correctional facility, the progressive validation including data indicating an origin of the first offender, a destination of the first offender, a path from the origin to the destination the first offender is allowed to take, and a camera of the cameras between the origin and the destination and along the path that is to provide pixel data; and
receiving, from the server, an alert that, based on facial recognition and location detection analysis performed on the pixel data by the server, indicates the progressive validation associated with the first offender is violated, an offender identification associated with the first offender, a location of the first, offender in the correctional facility, and an indication of the progressive validation violated.
11. The method of claim 10 , wherein the pixel data includes second pixels corresponding to a face of a second offender and a monitoring rule includes data identifying the first offender and the second offender and that the first offender and the second offender are not permitted to associate with each other.
12. The method of claim 10 , wherein the progressive validation includes data identifying a location that the first offender is not allowed to occupy, the location identified by an object in the pixel data or a location associated with the camera of the cameras that produced the pixel data.
13. The method of claim 10 , wherein the server determines a direction of the offender based on the pixel data and the progressive validation indicates a direction along the path that the first offender is allowed to travel.
14. The method of claim 13 , wherein the pixel data indicates a time and the progressive validation indicates a range of times in which the first offender is expected to be in the field of the view of respective cameras at the origin, destination, and the camera between the origin the destination.
15. The method of claim 10 , wherein the pixel data includes third pixels corresponding to a face of an officer and wherein a monitoring rule indicates a gender of the first offender is different from a gender of the officer.
16. A non-transitory machine-readable medium including instructions that, when executed by a machine, cause the machine to perform operations comprising:
generating pixel data including first pixels corresponding to a face of a first offender in the correctional facility and an environment around the first offender corresponding to a location within the field of view of the camera;
providing, to a server, the pixel data from the cameras through a network connection;
providing data of a user-defined progressive validation of a scheduled or non-scheduled movement of the first offender in the correctional facility, the progressive validation including data indicating an origin of the first offender, a destination of the first offender, a path from the origin to the destination the first offender is allowed to take, and a camera of the cameras between the origin and the destination and along the path that is to provide pixel data; and
receiving, from the server, an alert that, based on facial recognition and location detection analysis performed on the pixel data by the server, indicates the progressive validation associated with the first offender is violated, an offender identification associated with the first offender, a location of the first offender in the correctional facility, and an indication of the progressive validation violated.
17. The non-transitory machine-readable medium of claim 16 , wherein a monitoring rule includes data identifying a location that the first offender is not allowed to occupy, the location identified by an object in the pixel data or a location associated with the camera of the cameras that produced the pixel data.
18. The non-transitory machine-readable medium of claim 16 , wherein the server determines a direction of the offender based on the pixel data and the progressive validation indicates a direction along the path that the first offender is allowed to travel.
19. The non-transitory machine-readable medium of claim 18 , wherein the pixel data indicates a time and the progressive validation indicates a range of times in which the first offender is expected to be in the field of the view of respective cameras.
20. The non-transitory machine-readable medium of claim 16 , wherein the pixel data includes third pixels corresponding to a face of an officer and wherein a monitoring rule indicates a gender of the first offender is different from a gender of the officer.