Optimal camera selection in array of monitoring cameras
Technologies are generally described for automatically optimizing an efficiency of camera placement, numbers, and resolution in multi-camera monitoring and surveillance applications. In some examples, a fraction of a total area may be monitored at a higher resolution than the rest. Employing techniques such as combinatorial state Viterbi technique or combinatorial state trellis technique, a minimum number of cameras that provide the coverage at the needed resolution may be selected. Similarly, a number of points may be covered with at least a predefined number of cameras. For example, a subject of interest may be tracked in a public area, where specific camera(s) may be used to image the subject's face at a higher resolution than the background.
1. A method performed under a control of a computing device for optimal camera selection in an array of cameras, for monitoring and surveillance applications, the method comprising:
determining a plurality of resolutions associated with a plurality of cameras defined for intervals along a linear axis;
receiving information associated with points on the intervals and desired resolutions for the points;
forming a combinatorial state trellis, wherein each level of the combinatorial state trellis represents a point according to a linear order of the points and possible combinations of camera resolutions that cover the point are listed as states on a corresponding level;
forming transitions from one level to a next level in the combinatorial state trellis, while maintaining a resolution consistency for one or more common cameras, of the plurality of cameras, that correspond to a start state and an end state of each transition,
wherein maintaining the resolution consistency for the one or more common cameras comprises controlling one or more cameras to cover a plurality of points, among the points on the intervals, with same resolution;
evaluating optimal paths through the levels in the combinatorial state trellis while obeying resolution constraints in each path that is traversed in the combinatorial state trellis until a survival path is determined; and
determining, among the optimal paths, an optimal configuration path that involves a lowest cost, as compared to cost of other optimal paths, to cover the points on the intervals as the survival path.
2. The method according to claim 1 , further comprising:
enabling a transition from a non-zero resolution to a zero resolution in response to a determination that a maximum possible resolution in the next level is less than a resolution at a start of the transition.
3. The method according to claim 1 , further comprising:
enabling a transition from a zero resolution to a non-zero resolution in response to a determination that a maximum possible resolution in a previous level is less than a non-zero resolution of a desired state.
4. The method according to claim 1 , further comprising:
monitoring transitions from a non-zero resolution to a zero resolution for each path in the combinatorial state trellis;
setting a variable to a particular value when a transition from the non-zero resolution to the zero resolution occurs: and
selecting a current branch as a branch having a lowest cost that enters a current state.
5. The method according to claim 1 , further comprising:
defining combinations of directions for multidirectional cameras as states on each branch of the combinatorial state trellis.
6. The method according to claim 1 , further comprising:
performing an exhaustive search by:
examining all eligible combinations,
computing a cost of each combination, and
selecting a combination with the lowest cost.
7. The method according to claim 1 , further comprising:
evaluating the points and possible combinations of camera assignment to the points;
upon determining one or more possible camera and resolution combinations for each point, examining the resolution consistency;
omitting possible camera and resolution combinations that violate the resolution consistency; and
selecting a camera and resolution combination among remaining camera and resolution combinations.
8. The method according to claim 1 , further comprising:
setting a resolution for a camera of the one or more common cameras to zero in response to a determination that the camera is to be turned off or set to a resolution that fails to cover a point to which the camera is assigned.
9. The method according to claim 1 , further comprising:
defining a total cost of each transition as a linear combination of a number of cameras and resolution costs.
10. A computing device for optimal camera selection in an array of cameras, for monitoring and surveillance applications, the computing device comprising:
a memory configured to store instructions; and
a processor configured to:
determine a plurality of resolutions associated with a plurality of cameras defined for intervals along a linear axis;
receive information associated with points on the intervals and desired resolutions for the points;
form a combinatorial state trellis, wherein each level of the combinatorial state trellis represents a point according to a linear order of the points and possible combinations of camera resolutions that cover the point are listed as states on a corresponding level;
form transitions from one level to a next level in the combinatorial state trellis, while a resolution consistency for one or more common cameras of the plurality of cameras that correspond to a start state and, an end, state of each transition is maintained,
wherein to maintain the resolution consistency for the one or more common cameras, the processor is configured to control one or more cameras to cover a plurality of points, among the points on the intervals, with same resolution; and
evaluate optimal paths through the levels in the combinatorial state trellis while obeying resolution constraints in each path that is traversed in the combinatorial state trellis until a survival path is determined.
11. The computing device according to claim 10 , wherein the processor is further configured to:
enable a transition from a non-zero resolution to a zero resolution in response to a determination that a maximum possible resolution in the next level is less than a resolution at a start of the transition.
12. The computing device according to claim 10 , wherein the processor is further configured to:
enable a transition from a zero resolution to a non-zero resolution in response to a determination that a maximum possible resolution in a previous level is less than a non-zero resolution of a desired state.
13. The computing device according to claim 10 , wherein the processor is further configured to:
monitor transitions from a non-zero resolution to a zero resolution for each path in the combinatorial state trellis;
set a variable to a particular value when a transition from the non-zero resolution to the zero resolution occurs; and
select a current branch as a branch with a lowest cost that enters a current state.
14. The computing device according to claim 10 , wherein the processor is further configured to:
perform an exhaustive search by:
examination of all eligible combinations,
computation of a cost of each combination, and
selection of a combination with a lowest cost.
15. The computing device according to claim 10 , wherein the processor is further configured to:
evaluate the points and possible combinations of camera assignment to the points;
upon determination of one or more possible camera and resolution combinations for each point, examine the resolution consistency;
omit possible camera and resolution combinations that violate the resolution consistency; and
select a camera and resolution combination among remaining camera and resolution combinations.
16. The computing device according to claim 10 , wherein the processor is further configured to:
set a resolution for a camera to zero in response to a determination that the camera is to be turned off or set to a resolution that fails to cover a point to which the camera is assigned.
17. The computing device according to claim 10 , wherein the processor is further configured to:
define a total cost of each transition as a linear combination of a number of cameras and resolution costs.
18. A non-transitory computer readable medium with instructions stored thereon that, when executed, cause one or more processors to perform operations comprising:
determining a plurality of resolutions associated with a plurality of cameras defined for intervals along a linear axis;
receiving information associated with points on the intervals and desired resolutions for the points;
forming a combinatorial state trellis, wherein each level of the combinatorial state trellis represents a point according to a linear older of the points and possible combinations of camera resolutions that cover the point are listed as states on a corresponding level;
forming transitions from one level to a next level in the combinatorial state trellis, while maintaining a resolution consistency for one or more common cameras of the plurality of cameras that correspond to a start state and an end state of each transition,
wherein maintaining the resolution consistency for the one or more common cameras comprises controlling one or more cameras to cover a plurality of points, among the points on the intervals, with same resolution; and
evaluating optimal paths through the levels in the combinatorial state trellis while obeying resolution constraints in each path that is traversed in the combinatorial state trellis until a survival path is determined.