Video processing system using target property map
A system for detecting behavior of a target may include: a target detection engine, adapted to detect at least one target from one or more objects from a video surveillance system recording a scene; a path builder, adapted to create at least one mature path model from analysis of the behavior of a plurality of targets in the scene, wherein the at least one mature path model includes a model of expected target behavior with respect to the at least one path model; and a target behavior analyzer, adapted to analyze and identify target behavior with respect to the at least one mature path model. The system may further include an alert generator, adapted to generate an alert based on the identified behavior.
1. A video processing system comprising:
memory storing data and instructions; and
one or more computer processors configured to receive an input video sequence including a plurality of first targets and a second target, and further configured to access the memory and execute the instructions, causing the one or more computer processors to:
derive target property data from analysis of the plurality of first targets of the input video sequence;
calculate a plurality of statistical models, wherein each of the plurality of statistical models is based on the target property data derived from the input video sequence
build a size map and an entry/exit map using a target property map training algorithm; and
using the size map, the entry/exit map and a statistical model of the plurality of statistical models that corresponds to a location associated with the input video sequence of the second target, determining whether behavior of the second target is consistent with respect to a classification of the second target and expected behavior of the second target in relation to a path in the location.
2. The video processing system of claim 1 , wherein when the one or more computer processors are configured to derive the target property data from the input video sequence, the one or more computer processors are further configured to:
derive, from the input video sequence, the target property data based on one or more properties of the plurality of first targets.
3. The video processing system of claim 1 , wherein when the one or more computer processors are configured to calculate the plurality of statistical models, the one or more computer processors are further configured to:
calculate the plurality of statistical models based on a same target property of the plurality of first targets at each location associated with the input video sequence.
4. The video processing system of claim 1 , wherein when the one or more computer processors are configured to calculate the plurality of statistical models, the one or more computer processors are further configured to:
calculate the plurality of statistical models based on a function of one or more target properties of the plurality of first targets at each location.
5. The video processing system of claim 1 , wherein when the one or more computer processors are configured to build a target property map, the one or more computer processors are further configured to:
map each of the plurality of statistical models to a corresponding single one of the locations.
6. The video processing system of claim 5 , wherein the one or more computer processors are further configured to:
initialize an array corresponding to a size of the target property map.
7. The video processing system of claim 1 , wherein the one or more computer processors are further configured to:
initialize an array corresponding to a size of a frame of the input video sequence.
8. The video processing system of claim 1 , wherein the one or more computer processors are further configured to:
identify regions of interest based on a target property map.
9. The video processing system of claim 8 , wherein the one or more computer processors are further configured to:
determine trajectory information based on the regions of interest.
10. The video processing system of claim 1 , wherein the one or more computer processors are further configured to:
identify regions of interest based on a target property map.
11. The video processing system of claim 10 , wherein the one or more computer processors are further configured to:
determine trajectory information based on the regions of interest.
12. The video processing system of claim 1 , wherein the one or more computer processors are further configured to:
receive new target property data corresponding to a new target instance of the second target; and
compare at least a portion of the new target property data with the target property map to determine a degree of conformance.
13. The video processing system of claim 12 , wherein the one or more computer processors are further configured to:
determine whether the new target instance is consistent with behavior predicted by a target property map based on the degree of conformance to detect abnormal behavior.