Video Object Tracking via Dwell Time Signaling
Solution Overview
Problem
Existing video monitoring systems based on modern computer architectures primarily detect objects moving in image sequences and trigger alarms based on predefined threshold values for segment distance or size, failing to effectively detect atypical behavior patterns such as prolonged stationary objects, which may lead to delayed alarm triggering.
Innovation Solution
A method that tracks objects by generating a signaling based on their dwell time at a location, using a list or matrix to store image coordinates over time, detecting motion vectors, and incrementing a counter when an object remains stationary, allowing for simultaneous monitoring of multiple objects and efficient memory usage, with the option to adapt the reference image for dynamic motion detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video monitoring systems use predefined threshold values for segment distance or size to detect moving objects, then object detection capability is improved, but detection of atypical behavior patterns such as prolonged stationary objects deteriorates
Solution Approach 1:
The system dynamically switches between two monitoring modes: movement detection mode for normal operation and stationary object detection mode when an object stops moving. This dynamic adaptation allows the system to respond appropriately to different object behaviors, resolving the contradiction between detecting moving objects and detecting atypical stationary objects.
Solution Approach 2:
The system changes the detection parameter from movement-based thresholds to dwell-time-based thresholds when an object transitions from moving to stationary. By adjusting the monitoring parameter based on object state, the system can reliably detect both moving objects (using movement thresholds) and stationary objects (using dwell time thresholds), eliminating the detection blind spot.
2Speed
If the system triggers alarm immediately upon detecting object movement, then response speed is improved, but detection of atypical behavior patterns such as prolonged stationary objects is delayed
Solution Approach 1:
The system performs preliminary monitoring of object movement and automatically transitions to stationary object monitoring mode when an object stops moving. This preliminary detection of movement followed by automatic mode switching ensures that the system is already prepared to detect atypical behavior patterns without delay, resolving the contradiction between fast alarm triggering and timely detection of stationary objects.
3Device complexity
If the system monitors only movement-based parameters, then processing simplicity is maintained, but detection of stationary objects and atypical behavior patterns becomes insufficient
Solution Approach 1:
The monitoring process is segmented into distinct phases: movement detection phase and stationary object detection phase. Each phase uses appropriate parameters for its specific purpose, maintaining processing simplicity within each phase while achieving comprehensive detection accuracy across different object behaviors through phase segmentation.
Solution Approach 2:
The system dynamically selects which monitoring parameters to use based on object state. When an object is moving, movement parameters are used; when stationary, dwell time parameters are used. This dynamic parameter selection maintains processing simplicity by using only relevant parameters for the current state while achieving comprehensive detection accuracy.
Data Source
AI summary
A method is provided for monitoring at least one object in a scene, which method provides that, once a moving object stops and comes to a standstill, the dwell time at standstill is counted so as to generate a signaling as a function of the dwell time. The object is described by a list or a matrix characterizing the direction of movement and the time.