Abandoned Object Detection via Person-Object Association Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Security systems face challenges in effectively identifying and tracking abandoned objects in crowded areas, as existing technologies struggle to differentiate between harmless left-behind items and potential threats, such as bombs, without raising false alarms.
Innovation Solution
A method utilizing video analytics to detect and track objects and persons through video streams, establishing object-person associations, and determining abandonment by tracking their separation over time, with dynamic thresholding and attribute recognition to minimize false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video analytics are used to detect and track all objects in crowded areas, then the ability to identify abandoned objects improves, but the number of false alarms increases due to inability to differentiate between harmless items and potential threats
Solution Approach 1:
The system segments the object detection process into multiple analytical stages: initial object detection, person-object association analysis, abandonment behavior detection, and threat classification. This segmentation allows the system to analyze different aspects separately and combine them to reduce false alarms while maintaining identification accuracy.
Solution Approach 2:
The system performs preliminary object-person association analysis before making abandonment determinations. By pre-establishing relationships between persons and objects and monitoring their interactions over time, the system can distinguish between legitimately left items and potential threats before generating alarms.
2Device complexity
If traditional security monitoring methods are used in crowded places, then the system complexity remains low, but the ability to detect and track abandoned objects effectively deteriorates
Solution Approach 1:
The video analytics system automatically performs object detection, tracking, association analysis, and abandonment determination without requiring manual security personnel intervention for each object. The system self-manages the complex analysis tasks, maintaining low operational complexity while achieving high detection effectiveness.
Solution Approach 2:
The system replaces manual security monitoring with automated video analytics technology. Computer vision algorithms and machine learning models substitute for human analysts, enabling the system to process and analyze video data at scales and speeds that manual methods cannot achieve.
3Measurement precision
If video analytics track every object and person separately, then the accuracy of abandonment detection improves, but the computational resources and system complexity increase
Solution Approach 1:
The system merges object tracking and person tracking into a unified object-person association framework. By combining these tracking streams and analyzing their relationships, the system achieves accurate abandonment detection while reducing the complexity of maintaining separate tracking systems for each object and person.
Data Source
AI summary
A method for detecting an abandoned object in a video stream captured by a video camera includes receiving a plurality of video frames of the video stream. Video analytics are performed on one or more of the plurality of video frames to detect one or more objects and one or more persons in the video stream. An object-person association between one of the detected objects and one of the detected persons is identified, resulting in an object/person pair. With the object-person pair identified, the object of the object/person pair and the person of the object/person pair are each tracked through subsequent video frames. Based on the tracking of the object of the object/person pair and the tracking of the person of the object/person pair, a determination is made as to when the object of the object/person pair becomes abandoned by the person of the object/person pair.