Camera-Based Vehicle Detection Using Solid Angle Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle detection systems using coils are prone to incorrect detection of non-metallic objects and are costly to install and maintain, as they require burial in the ground, and can mistakenly identify humans as vehicles due to the lack of distinction between vehicles and humans.
Innovation Solution
A vehicle detection device that uses a combination of first and second signals to determine if a subject object is a vehicle, based on distance and solid angle measurements, without relying on coils, allowing for accurate differentiation between vehicles and other objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If coils are used for vehicle detection, then detection reliability is improved, but installation complexity and cost increase due to ground burial requirements
Solution Approach 1:
The patent replaces the mechanical coil-based detection system with a camera-based optical detection system. The camera captures images of vehicles, and the system processes these images to detect vehicle presence and movement, eliminating the need for ground burial and complex coil installation while maintaining detection reliability.
Solution Approach 2:
The patent introduces a camera as an intermediary device between the vehicle and the detection system. Instead of directly detecting metallic objects with coils, the camera captures visual information about vehicles, which is then processed to determine vehicle presence and movement, simplifying installation while maintaining detection accuracy.
2Reliability
If coils are used for vehicle detection, then vehicle detection capability is improved, but maintenance cost and effort increase
Solution Approach 1:
The patent replaces the mechanical coil system with an electronic camera-based system. Cameras are more durable, require no ground burial, and can be easily replaced or calibrated if issues arise, significantly reducing maintenance effort compared to coil systems that require excavation and precise positioning.
3Measurement precision
If coil-based detection is used, then metallic object detection is improved, but false detection of non-vehicle objects increases
Solution Approach 1:
The patent uses dynamic image processing to distinguish vehicles from other objects. By analyzing sequential images and detecting movement patterns characteristic of vehicles (wheel rotation, body movement), the system can differentiate vehicles from stationary metallic objects or non-vehicle moving objects, reducing false detections while maintaining detection precision.
Solution Approach 2:
The patent transitions from one-dimensional coil detection to two-dimensional image analysis. The camera captures spatial and temporal information across multiple dimensions, allowing the system to analyze vehicle shape, size, movement patterns, and position changes over time, thereby distinguishing vehicles from other objects more reliably than coil-based metallic detection.
4Length of stationary object
If microwave-based detection is used, then detection range is improved, but inability to distinguish vehicles from humans causes incorrect detection
Solution Approach 1:
The patent analyzes dynamic movement patterns in sequential images to distinguish vehicles from humans. Vehicles exhibit characteristic movement patterns such as wheel rotation, suspension movement, and larger displacement distances, which differ from human walking patterns. This dynamic analysis maintains long detection range while achieving accurate object classification.
Solution Approach 2:
The patent adds temporal and spatial dimension analysis to the detection process. By processing sequences of images and analyzing movement trajectories, speed, acceleration, and object dimensions over time, the system can classify objects as vehicles or humans based on their movement characteristics, maintaining long detection range while achieving high classification accuracy.
Data Source
AI summary
A vehicle detection device includes: a first signal obtaining section which obtains a first signal corresponding to a distance to a subject object; a second signal obtaining section which obtains a second signal corresponding to a solid angle formed by the subject object; and a vehicle determining section which determines, based on the first and second signals, that the subject object is a vehicle, in a case where an actual size of the subject object is equal to or greater than a predetermined size.