Method for person re-identification in closed place, system, and terminal device
The present disclosure provides a method, system, and terminal device for re-identifying a person in a closed place, relating to the field of video processing technologies. The method comprises: dividing a closed place into a plurality of regions comprising an entrance region, any of the regions comprising at least one camera for photographing; confirming a person first entering based on a trajectory tracking of the person in the entrance area, and assigning a unique identity number to the person, acquiring and recording feature information thereof and binding the identity number with the feature information; and when re-identification of a person in the closed place is performed, matching person's feature information captured by the camera of the region where the person is, with recorded feature information selecting the identity number bound with recorded feature information that best matches the feature information of the person as the identity number of the person.
1. A method for re-identifying a human being in a closed place, comprising:
dividing the closed place into multiple areas comprising an entrance area, and having at least one camera for shooting in any area;
confirming a human being first entering based on a trajectory tracking of the human being in the entrance area, assigning an identity number to the human being, acquiring and recording the feature information of the human being, and binding the identity number with the feature information of the human being; and
when re-identification of a human being in the closed place is performed, matching the feature information of the human being captured by the camera in an area where the human being is with the recorded feature information, and selecting the identity number bound with the recorded feature information that best matches the feature information of the human being as the identity number of the human being.
2. The method according to claim 1 , further comprising: when the human being leaves, deleting the feature information of the human being and marking the human being as having left.
3. The method according to claim 1 , wherein having at least one camera for shooting in any area comprises:
having two cameras with different shooting angles for shooting in any area.
4. The method according to claim 1 , wherein confirming a human being first entering based on a trajectory tracking of the human being in the entrance area comprises:
judging whether the human being is the human being first entering according to a trajectory of the human being displayed on continuous shot screens with multiple frames of the entrance area.
5. The method according to claim 1 , wherein acquiring feature information of the human being is performed by a convolutional neural network technology.
6. The method according to claim 1 , wherein the feature information of the human being comprises: stature, dress, and/or appearance.
7. The method according to claim 1 , wherein matching the feature information of the human being captured by the camera in an area where the human being is with the recorded feature information comprises: performing matching on all human beings in the area.
8. The method according to claim 1 , wherein the feature information and the identity number are stored in a human being feature database.
9. A system for re-identifying a human being in a closed place, comprising:
a processor; and
a memory storing instructions for the processor to control the following operations:
dividing the closed place into multiple areas comprising an entrance area, and having at least one camera for shooting in any area;
confirming a human being first entering based on a trajectory tracking of the human being in the entrance area, assigning an identity number to the human being, acquiring and recording the feature information of the human being, and binding the identity number with the feature information of the human being; and
when re-identification of a human being in the closed place is performed, matching the feature information of the human being captured by the camera in an area where the human being is with the recorded feature information, and selecting the identity number bound with the recorded feature information that best matches the feature information of the human being as the identity number of the human being.
10. The system according to claim 9 , further comprising a feature database, configured to store the feature information and the identity number.
11. The system according to claim 9 , wherein the processor is further configured to delete the feature information of the human being and marking the human being as having left, when the human being leaves.
12. The system according to claim 9 , wherein having at least one camera for shooting in any area comprises:
having two cameras with different shooting angles for shooting in any area.
13. The system according to claim 9 , wherein confirming a human being first entering based on a trajectory tracking of the human being in the entrance area comprises: judging whether the human being is the human being first entering according to a trajectory of the human being displayed on continuous shot screens with multiple frames of the entrance area.
14. The system according to claim 9 , wherein acquiring feature information of the human being is performed by a convolutional neural network technology.
15. The system according to claim 9 , wherein the feature information of the human being comprises: stature, dress, and/or appearance.
16. The system according to claim 9 , wherein matching the feature information of the human being captured by the camera in an area where the human being is with the recorded feature information comprises: performing matching on all human beings in the area.
17. The system according to claim 9 , wherein the feature information and the identity number are stored in a human being feature database.
18. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, causes to perform steps of the following method:
dividing the closed place into multiple areas comprising an entrance area, and having at least one camera for shooting in any area;
confirming a human being first entering based on a trajectory tracking of the human being in the entrance area, assigning a unique identity number an identity number to the human being, acquiring and recording the feature information of the human being, and binding the identity number with the feature information of the human being; and
when re-identification of a human being in the closed place is performed, matching the feature information of the human being captured by the camera in an area where the human being is with the recorded feature information, and selecting the identity number bound with the recorded feature information that best matches the feature information of the human being as the identity number of the human being.