Cross-camera obstacle tracking method, system and medium

A cross-camera obstacle tracking system, comprising at least two cameras, and an electronic device, wherein the electronic device is configured to,

• obtain obstacle tracking results of images captured by at least two cameras, wherein each obstacle tracking result comprises results after identifying and tracking at least one obstacle in an image captured by each camera, and each obstacle is labeled with a local identifier for each camera;
• in response to a triggering condition of a main camera in the cameras to fusion of cross-camera obstacles being satisfied, establish a mapping relation between local identifiers of the obstacles in the cameras and global identifiers according to the obstacle tracking result of each of the cameras;
• determine similarities of obstacles according to the obstacle tracking result of each of the cameras, and perform fusion of the global identifiers on same obstacles according to the similarities;
• and
• obtain final obstacle tracking results labeled with the global identifiers;
• wherein the triggering condition of the main camera in the at least two cameras to fusion of cross-camera obstacles is satisfied, when detecting that an obstacle tracking result of the main camera is obtained after identifying and tracking obstacles in a current image of the main camera is completed;
• wherein a priority level of the main camera in driving control is higher than priority levels of other cameras in driving control;
• wherein the electronic device is configured to,
• for the obstacle tracking results corresponding to different global identifiers, perform similarity calculation on position information and/or contour volume information in pairs according to the obstacle tracking result of each of the cameras, to generate a similarity matrix;
• perform an obstacle matching according to the similarity matrix to obtain a matching result, and determine the global identifiers belonging to the same obstacle according to the matching result;
• and
• combine and fuse the global identifiers belonging to the same obstacle.

A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer execute a cross-camera obstacle tracking method, the method comprises:

• obtaining obstacle tracking results of images captured by at least two cameras, wherein each obstacle tracking result comprises results after identifying and tracking at least one obstacle in an image captured by each camera, and each obstacle is labeled with a local identifier for each camera;
• in response to a triggering condition of a main camera in the cameras to fusion of cross-camera obstacles being satisfied, establishing a mapping relation between local identifiers of the obstacles in the cameras and global identifiers according to the obstacle tracking result of each of the cameras;
• determining similarities of obstacles according to the obstacle tracking result of each of the cameras, and performing fusion of the global identifiers on same obstacles according to the similarities;
• and
• obtaining final obstacle tracking results labeled with the global identifiers;
• wherein the triggering condition of the main camera in the at least two cameras to fusion of cross-camera obstacles is satisfied, when detecting that an obstacle tracking result of the main camera is obtained after identifying and tracking obstacles in a current image of the main camera is completed;
• wherein a priority level of the main camera in driving control is higher than priority levels of other cameras in driving control;
• wherein determining the similarities of the obstacles according to the obstacle tracking result of each of the cameras, and performing the fusion of the global identifiers on the same obstacle according to the similarities, comprises:
• for the obstacle tracking results corresponding to different global identifiers, performing similarity calculation on position information and/or contour volume information in pairs according to the obstacle tracking result of each of the cameras, to generate a similarity matrix;
• performing an obstacle matching according to the similarity matrix to obtain a matching result, and determining the global identifiers belonging to the same obstacle according to the matching result;
• and
• combining and fusing the global identifiers belonging to the same obstacle.