3D Point Cloud Axle Detection via Wheel Shadows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting vehicle axles in traffic are not robust against adverse weather conditions and suffer from accuracy issues, and traditional induction loops are prone to wear and tear, requiring traffic disruptions for installation and maintenance.
Innovation Solution
A detection device using a profile detection sensor to generate a 3D point cloud, which evaluates deep shadows as indicators of vehicle axles, providing a contactless and robust method for axle detection even in poor weather, by identifying areas with missing depth information behind the wheels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If visual methods (camera images, 3D camera, laser scanner) are used to detect vehicle axles, then the detection can be performed contactlessly, but the accuracy is insufficient especially in bad weather conditions like snow and rain
Solution Approach 1:
Instead of detecting the wheels directly (positive information), the invention detects the deep shadows cast by the wheels on the roadway (negative information). The profile detection sensor captures areas where depth information is missing or significantly reduced, which correspond to the shadows behind the wheels. This inverted approach of detecting absence rather than presence of information provides more reliable axis detection in adverse weather conditions.
2Reliability
If induction loops are embedded in the roadway to detect vehicle axles, then the axis detection is reliable, but the assembly is complex and requires traffic shutdown for installation and maintenance
Solution Approach 1:
The invention replaces the mechanical induction loop system embedded in the roadway with a non-contact optical measurement system. The profile detection sensor uses light to measure depth information and detect wheel shadows, eliminating the need for physical contact with the roadway. This substitution removes the complexity of embedding, wiring, and maintaining induction loops while achieving comparable or superior detection reliability.
3Reliability
If induction loops are used for axle detection, then reliable detection is achieved, but wear and tear occurs especially on busy roads requiring replacement
Solution Approach 1:
The contact-based induction loop system is replaced with a contactless optical measurement system. The profile detection sensor measures wheel shadows through light reflection and depth measurement without any physical contact with the roadway or vehicles. This eliminates mechanical wear and tear completely, allowing the sensor to operate indefinitely without degradation from traffic loads, while maintaining detection reliability.
4Ease of operation
If 3D methods are used for vehicle detection, then vehicle separation from background is easy, but the lower resolution makes them sensitive to snow and rain
Solution Approach 1:
The invention inverts the detection approach by focusing not on the vehicle surface itself but on the shadow regions behind the wheels where depth information is absent. This inversion allows the use of depth information to detect axles through the presence of shadows rather than relying on surface detail, thereby maintaining ease of vehicle-background separation while improving robustness against snow and rain that would otherwise obscure surface features.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables real-time, automatic, and reliable detection of vehicle axles with high accuracy and durability, avoiding road damage and maintaining high detection rates in adverse conditions.
Implementation Method 1
with a profile detection sensor, also known as a depth sensor, a 3D point cloud is generated that contains information about the surface of the vehicle and the roadway
Data Source
Figure 1~3
Figure 4~6
Figure 7~9
AI summary
A detection device (10) for detecting the vehicle axes of a vehicle on a roadway (12) is described, wherein the detection device (10) comprises at least one profile detection sensor (14a-c) for generating a 3D point cloud of the vehicle and the roadway (12) and an evaluation unit for identifying vehicle axes in the 3D point cloud. The evaluation unit is configured to locate depth shadow areas (16) of the 3D point cloud cast by wheels and to identify the located depth shadow areas (16) as vehicle axes.