PTZ Video Background Modeling with Segmentation Reset
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional background modeling for cameras with movement capabilities, such as PTZ cameras, struggles to accurately distinguish between camera movements and scene changes, leading to incorrect categorization of static scene parts as foreground, and is limited by processing power constraints.
Innovation Solution
A hybrid approach combining conventional background model updating during stable conditions and a quick reset using image segmentation and/or object detection algorithms upon camera stop, allowing for efficient and accurate background modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If conventional background modelling based on motion analysis is used for PTZ cameras, then processing power consumption is reduced, but background-foreground differentiation accuracy deteriorates due to inability to distinguish camera movement from scene changes
Solution Approach 1:
The system dynamically switches between two background modelling approaches: conventional motion-based modelling during camera movement to save power, and AI-based segmentation when the camera is stationary to ensure accuracy. This dynamic adaptation resolves the contradiction by optimizing power consumption while maintaining differentiation accuracy through conditional method selection.
Solution Approach 2:
The system changes the operational parameters of background modelling based on camera movement detection. When movement is detected, it uses lightweight motion analysis; when stationary, it employs computationally intensive AI-based segmentation. This parameter change strategy allows the system to achieve high accuracy during stationary periods while conserving power during movement, effectively resolving the contradiction.
2Measurement precision
If AI-based background modelling is continuously applied, then background-foreground differentiation accuracy is improved, but processing power consumption increases
Solution Approach 1:
Instead of continuous AI-based modelling, the system applies AI-based segmentation periodically only when the camera is stationary. During movement periods, it uses conventional motion-based modelling. This periodic application of the computationally intensive method reduces overall power consumption while maintaining accuracy during the periods when AI-based modelling is active.
Solution Approach 2:
The system dynamically adjusts the level of processing based on camera movement state. It transitions from high-power AI-based modelling during stationary periods to low-power motion-based modelling during movement. This dynamic adjustment resolves the contradiction by applying the accurate but power-intensive method only when necessary, thereby maintaining accuracy while reducing overall power consumption.
3Productivity
If conventional background modelling is used during camera movement, then processing speed is maintained, but background model validity deteriorates leading to incorrect foreground detection
Solution Approach 1:
The system extracts and separates the camera movement component from the scene analysis during movement periods. It uses motion-based modelling that specifically tracks camera movement patterns, allowing it to maintain processing speed while accounting for camera motion. When stationary, it performs a complete reset using AI-based segmentation to ensure background model validity, thus resolving the contradiction between processing speed and model validity.
Solution Approach 2:
The system dynamically adapts the background modelling approach based on camera movement detection. During movement, it uses motion-based modelling that maintains processing speed while being aware of camera motion. When the camera stops, it transitions to AI-based segmentation to reset and validate the background model. This dynamic switching resolves the contradiction by maintaining speed during movement and ensuring validity during stationary periods.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
The present disclosure relates to a method of background modelling for a video stream acquired by a camera having movement capabilities. The method comprises: acquiring a video stream; repeatedly updating a background model for the video stream by analyzing changes in the sequence of image frames and categorizing image areas in the image frames which do not change over time as background; detecting camera movement; performing a reset of the background model by: applying an image segmentation and/or object detection algorithm to identify at least foreground objects; after having performed the reset of the background model, returning to repeatedly updating the background model for the video stream by analyzing changes in the sequence of image frames and categorizing image areas in the image frames which do not change over time as background. The disclosure further relates to an image processing system.