Video Surveillance Motion Analysis via Virtual 3D Coordinates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Surveillance systems with multiple cameras face challenges in comparing motion information due to differing installation conditions, such as camera height and angle, leading to inconsistent imaging and difficulty in analyzing motion data across cameras.
Innovation Solution
A video surveillance system that extracts motion feature quantities from video images, converts them using a virtual coordinate system with unique axes, and calculates virtual viewpoints to unify the data for comparison across cameras with different installation conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion information is converted into coordinate positions of a specific camera for comparison, then coordinate comparison can be performed, but comparison is only possible when fields of view overlap
Solution Approach 1:
The patent transforms the coordinate system from two-dimensional camera-specific coordinates to three-dimensional real-world space coordinates. By introducing the Z-axis (depth distance from camera), motion information is expressed in a unified 3D coordinate system that represents actual physical space rather than camera-specific viewing coordinates. This dimensional transformation enables comparison of motion information across cameras with different installation positions and angles, as all cameras observe the same 3D space.
Solution Approach 2:
The patent introduces real-world space coordinates as an intermediary reference system between multiple cameras. Instead of directly comparing camera coordinates or converting to a single camera's coordinate system, the system uses the unified 3D real-world coordinate system as a mediator. Motion information from different cameras is transformed into this intermediary space, enabling accurate comparison and analysis of motions across the surveillance area regardless of camera installation differences.
2Area of stationary object
If multiple cameras with different installation conditions are used to expand surveillance coverage, then surveillance area is increased, but motion information becomes inconsistent and difficult to compare
Solution Approach 1:
The patent resolves the inconsistency issue by changing the coordinate representation from 2D camera-specific coordinates to 3D real-world coordinates. This dimensional change incorporates depth information (distance from camera along the optical axis) as a third dimension, allowing motion information from cameras at different positions and angles to be expressed in a common spatial reference frame. As a result, motions can be accurately compared and tracked across the expanded surveillance area covered by multiple cameras.
Data Source
AI summary
A video surveillance system, including motion information from a plurality of surveillance cameras even when the installation conditions of the cameras are different, is equipped with an extraction unit which, upon receiving video images from the plurality of cameras, extracts motion feature quantities from a plurality of frames which constitute the video images, a storage unit for storing extraction results from the extraction unit, a conversion unit for converting the extracted motion feature quantities, and an analysis unit for analyzing the converted motion feature quantities, wherein the conversion unit retains virtual coordinate axes which are different from the coordinate axes of the plurality of cameras and converts the feature quantities by calculating virtual viewpoints on the virtual coordinate axes.