Feature Block Background Modeling for Video Surveillance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video surveillance systems face challenges in change detection due to camera automatic gain control (AGC) and camera jitter, leading to false target detections and performance degradation, especially in environments with unstable illumination and camera motion.
Innovation Solution
A feature-based background modeling approach that divides video frames into image blocks, determines feature blocks, and creates a feature block map to model the background, allowing for detection and compensation of AGC effects and camera jitter, thereby improving foreground and background segmentation and stabilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pixel-based background modeling is used, then the system can detect changes in the video, but camera AGC and jitter cause false target detections and reduce system performance
Solution Approach 1:
The patent divides the video frame into multiple image blocks and processes each block independently to create feature blocks. This segmentation allows the system to handle camera AGC and jitter effects more effectively by analyzing local features rather than treating the entire frame as a single unit, thereby reducing false detections while maintaining change detection accuracy.
Solution Approach 2:
The patent transforms the background modeling approach from pixel-based to feature-based by extracting features from image blocks. This parameter change in the modeling methodology enables the system to be more robust against camera AGC and jitter, as feature blocks capture essential scene characteristics that are less sensitive to these disturbances.
2Adaptability or versatility
If pixel-based background modeling is used, then the system can model the background scene, but it fails in less-friendly environments with unstable illumination and camera motion
Solution Approach 1:
By segmenting the video frame into image blocks and creating feature blocks, the system can adapt to different environmental conditions more effectively. Each block can be processed independently, allowing the system to maintain detection accuracy in challenging environments with unstable illumination and camera motion.
Solution Approach 2:
The patent introduces feature blocks as an intermediary representation between the raw video frames and the background model. These feature blocks serve as a mediator that captures essential scene information while being less sensitive to environmental variations, thereby improving both adaptability and detection precision.
3Reliability
If feature-based background modeling is implemented, then false target detections are reduced, but the computational complexity increases
Solution Approach 1:
The patent segments the video frame into image blocks, which can be processed in parallel. This segmentation approach, when combined with efficient feature extraction, can actually reduce overall computational complexity compared to processing the entire frame at once, while simultaneously improving detection reliability through localized feature analysis.
Data Source
AI summary
Video content analysis of a video may include: modeling a background of the video; detecting at least one target in a foreground of the video based on the feature blocks of the video; and tracking each target of the video. Modeling a background of the video may include: dividing each frame of the video into image blocks; determining features for each image block of each frame to obtain feature blocks for each frame; determining a feature block map for each frame based on the feature blocks of each frame; and determining a background feature block map to model the background of the vide based on at least one of the feature block maps.