Grid-Based Motion Deviation Detection in Video Surveillance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing algorithms for detecting motion in video surveillance are computationally heavy, leading to high hardware costs and inability to distinguish between normal and abnormal motion, particularly in distinguishing direction deviation.
Innovation Solution
A grid-based model is used to detect motion deviation in surveillance videos, which requires less storage and computational resources, allowing for efficient detection and analysis of motion patterns using metadata, reducing hardware costs and network capacity requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image processing algorithms are used for motion detection, then motion detection accuracy is improved, but hardware costs and computational resources increase significantly
Solution Approach 1:
The video frame is divided into a grid of blocks, and motion detection is performed independently on each block. This segmentation allows the system to focus computational resources on detecting motion in specific regions rather than processing the entire frame with heavy algorithms, reducing hardware costs while maintaining detection accuracy.
Solution Approach 2:
The patent extracts only the essential motion information from video frames by comparing pixel intensity values at the same positions in consecutive frames. This extraction approach retrieves motion data without requiring complex image processing algorithms, significantly reducing computational resources and hardware costs while preserving motion detection capability.
2Reliability
If conventional image processing algorithms are used for motion detection, then motion detection capability is improved, but processing time and computational load increase
Solution Approach 1:
The patent extracts only the necessary motion information by comparing pixel intensities at corresponding positions in consecutive frames. This selective extraction avoids the computational overhead of conventional algorithms while maintaining motion detection reliability, enabling faster processing speeds.
Solution Approach 2:
The patent changes the detection parameter from complex image feature analysis to simple pixel intensity comparison. This parameter change simplifies the computational process significantly, reducing processing time and computational load while maintaining the ability to detect motion reliably through the generation of deviation maps that highlight motion regions.
3Measurement precision
If complete video recordings are transferred for analysis, then detection accuracy is improved, but network capacity requirements increase
Solution Approach 1:
The patent extracts only the essential motion deviation information from video frames and stores it as metadata. This extracted metadata contains the critical detection data without requiring transfer of complete video recordings, significantly reducing network capacity requirements while preserving detection accuracy through the use of deviation maps that capture motion patterns.
Solution Approach 2:
Instead of transferring and analyzing complete video recordings, the patent creates a simplified copy in the form of deviation maps that represent motion information. These deviation maps serve as compact representations that maintain detection accuracy while minimizing data transfer volume, allowing analysis to be performed on the copied metadata rather than the original video data.
Data Source
AI summary
Detection of motion deviation in a video sequence is provided. Change grids each comprise elements generated by storing in each element of the change grid an indication of whether there is change between corresponding elements of at least two images. A current direction grid is generated from a pair of change grids by searching for movement of a corresponding segment identified in each change grid, the movement occurring between the locations of the segment in each of the pair of change grids and, storing in elements of the current direction grid a vector corresponding to the movement of the segment. A vector stored in an element of the current direction grid is compared with a reference vector. It is determined whether there is motion deviation in the video sequence in accordance with the comparison.