Towbar Detection Using Reference Image Comparison and License Plate Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Automatic vehicle detection systems face precision issues in differentiating between vehicles towing trailers and those following closely, leading to erroneous detections and increased operational costs due to false alarms, and existing methods for detecting drawbars are complex and expensive.

Innovation Solution

A method using a single control camera to capture images of a control zone, comparing them to a reference image, and recognizing license plates to determine the presence of a drawbar by identifying non-matching plates, implemented with a varifocal sensor and periodic image capture, reducing computational intensity and equipment costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a single control camera is used to detect drawbars, then equipment cost and complexity are reduced, but detection precision and reliability deteriorate due to difficulty in differentiating between towed vehicles and closely following vehicles

Engineering Contradiction:
Improveequipment complexityVSAvoiddetection precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The detection process is segmented into multiple independent techniques: (1) comparing control zone images to reference images to detect drawbar presence, (2) recognizing license plates on both vehicles to verify towing relationship, and (3) using learning techniques for drawbar recognition. This segmentation allows the system to achieve high detection precision through multiple verification steps while maintaining relatively simple equipment by using a single camera.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by capturing reference images of the control zone when no vehicles are present, and pre-identifying license plates before making the final drawbar detection determination. This preliminary preparation enables more accurate comparison and reduces false detections during actual vehicle passage.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If automated detection systems are used without human intervention, then operating costs are reduced, but detection precision deteriorates leading to erroneous detections and false alarms

Engineering Contradiction:
Improveoperating efficiencyVSAvoiddetection reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system implements feedback mechanisms by comparing detected drawbars against multiple criteria: image comparison results, license plate matching verification, and learning technique outcomes. This multi-layered feedback validation ensures high detection reliability while maintaining automated operation, reducing false alarms without requiring human intervention.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system applies partial action by selectively implementing detection techniques based on confidence levels. When image comparison and license plate verification provide sufficient evidence, the system makes automated determinations without requiring all possible verification steps, thus maintaining both automation and reliability.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If multiple detection techniques are implemented including image comparison and license plate recognition, then detection reliability is improved, but computational intensity and processing time increase

Engineering Contradiction:
Improvedetection reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system implements detection techniques in a staged manner, performing image comparison first as a quick filter, then applying license plate recognition only when needed for verification. Learning techniques are applied selectively based on initial detection results. This partial application of techniques reduces processing time while maintaining high reliability through targeted use of computationally intensive methods.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3394795B1Method for detecting a towbar and associated computer program product
Publication Date: 2022.07.13 REVENUE COLLECTION SYST FRANCE SAS
  • EP3394795B1 patent drawingFigure 1~3
  • EP3394795B1 patent drawingFigure 2

AI summary

The invention relates to a method for detecting a towbar (105) connecting two motor vehicles (20, 25) moving between a first zone (55) and a second zone (60), the method comprising steps of: acquiring a single reference image of the control area (65) by means of the control camera (45), no vehicles (20, 25) being present in the control area (65) during the acquisition step; determining the potential presence of a towbar (105) by implementing at least one technique selected from: a first technique comprising the steps of comparing each working image to the reference image, and a second technique comprising steps of identifying a registration plate at the front of the first vehicle (20) and at the rear of the second vehicle (25).