Object Matching Across Overlapping Camera Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current camera systems struggle to accurately match moving objects across overlapping monitoring regions captured by multiple image capturing units, leading to inaccurate counting and tracking of objects and their routes due to differing coordinate systems.
Innovation Solution
The method involves detecting moving objects within the overlapped region, calculating transforming parameters between coordinate systems, determining a reliability level, and using this to transform coordinate points to determine if they represent the same object, thereby ensuring accurate matching and preventing repeated counting or tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple monitoring cameras are installed to cover wider monitoring regions, then the monitoring coverage is improved, but the difficulty of matching objects across overlapping regions increases
Solution Approach 1:
The patent introduces an operation processing unit as an intermediary that receives images from multiple image capturing units, performs coordinate transformations, and matches objects across different coordinate systems. This mediator handles the complexity of cross-camera object matching by centralizing the transformation and comparison operations.
Solution Approach 2:
The patent transforms the coordinate parameters from different camera coordinate systems into a unified reference coordinate system. By changing the coordinate parameters through mathematical transformations, the system enables accurate object matching across overlapping monitoring regions despite the original coordinate system differences.
2Productivity
If coordinate transformation is performed without reliability verification, then the processing speed is improved, but the accuracy of object matching deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the operation processing unit calculates transforming parameters, verifies their reliability against threshold values, and adjusts the matching process accordingly. This feedback loop ensures that only reliable transformations are used for object matching, maintaining accuracy while avoiding unnecessary reprocessing.
Solution Approach 2:
The patent performs preliminary calculations of transforming parameters and their reliability verification before the actual object matching process. By preparing the transformation parameters in advance and validating their reliability, the system ensures accurate matching without requiring complex real-time adjustments during object detection.
3Measurement precision
If the threshold for reliability level is set high, then the accuracy of determining same object is improved, but the number of false negatives increases
Solution Approach 1:
The patent employs dynamic threshold adjustment where the reliability threshold is not fixed but adapts based on the specific transformation parameters and conditions. This dynamic approach allows the system to maintain high accuracy requirements when transformations are uncertain while being more permissive when transformations are highly reliable, thus reducing false negatives.
Solution Approach 2:
The patent applies different reliability thresholds to different regions or types of transformations based on their specific characteristics. Instead of using a uniform threshold, the system tailors the reliability requirement to the local conditions of each coordinate transformation, allowing accurate matching where possible while accepting lower thresholds in challenging scenarios to avoid missing objects.
Data Source
AI summary
An object matching method is applied to a camera system with an object matching function. The object matching method includes detecting a moving object within overlapped monitoring areas of two image capturing units to generate a first coordinate point and a second coordinate point respectively upon a first coordinate system and a second coordinate system, calculating at least one transforming parameter of the first coordinate system relative to the second coordinate system, acquiring a reliability level according to a comparison result between the transforming parameter and a threshold, determining a final transform parameter by the reliability level, utilizing the final transform parameter to transform the first coordinate point into a third coordinate point upon the second coordinate system, and determining whether the first coordinate point and the second coordinate point indicate the same object by difference between the third coordinate point and the second coordinate point.