Lane Marking Color Detection Using Dynamic Focus Area

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for determining the color of lane markings in autonomous driving are not reliable enough for highly automated driving, especially when the lane marking color changes, as they often rely solely on edge detection which can be inaccurate.

Innovation Solution

A method that determines the color of lane markings by defining a focus area completely inside a bounding box based on sensor data, with the size of the focus area adjusted according to the size of the bounding box, ensuring accurate color detection with Automotive Safety Integrity Level (ASIL) integrity, using machine learning algorithms for plausibility checks and color distribution analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If edge detection methods are used to determine lane marking color, then the detection process is simple, but the reliability is insufficient for highly automated driving

Engineering Contradiction:
Improvedetection process complexityVSAvoidlane marking color detection reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent divides the lane marking detection process into distinct segments: first determining a bounding box of the lane marking, then defining a focus area completely inside the bounding box, and finally determining color only within this focused region. This segmentation isolates the color measurement to a controlled area, improving reliability by preventing contamination from adjacent regions while maintaining manageable process complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by creating a focus area with specific properties (completely inside the bounding box, size dependent on bounding box dimensions) that differs from the surrounding regions. By restricting color determination to this locally optimized area, the system ensures high reliability for color detection without requiring complex global processing, thus resolving the contradiction between simplicity and reliability.

Inventive Principle:
Principle #3Local quality

2Area of stationary object

If color determination is performed without a defined focus area, then the detection covers more region, but measurement may occur outside the lane marking reducing accuracy

Engineering Contradiction:
Improvedetection areaVSAvoidlane marking color measurement precision
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent performs preliminary actions by first determining the bounding box of the lane marking and then defining the focus area completely inside this bounding box before performing color determination. This preliminary positioning ensures that subsequent color measurements are guaranteed to occur within the lane marking boundaries, eliminating the risk of measuring outside the marking while maintaining efficient detection coverage.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The focus area acts as an intermediary between the bounding box and the color determination process. It is defined completely inside the bounding box with size dependent on the bounding box dimensions, serving as a controlled intermediate region that guarantees measurements occur within lane marking boundaries while still providing sufficient area for accurate color analysis.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of operation

If a fixed size area is used for color determination, then the processing is straightforward, but it cannot adapt to varying lane marking sizes reducing reliability

Engineering Contradiction:
Improveprocessing simplicityVSAvoiddetection reliability across varying conditions
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent implements dynamics by making the focus area size dependent on the bounding box size rather than using a fixed area. The focus area is defined completely inside the bounding box with dimensions that adapt to the detected lane marking size. This dynamic adjustment maintains processing simplicity through automated scaling while significantly improving reliability across varying lane marking conditions, including different distances and perspectives.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4177849A1Method and device for determining a color of a lane marking
Publication Date: 2023.05.10 BAYERISCHE MOTOREN WERKE AG
  • EP4177849A1 patent drawingFigure 1
  • EP4177849A1 patent drawingFigure 2
  • EP4177849A1 patent drawingFigure 3

AI summary

Provided is a method for determining a color of a lane marking for an automated vehicle. The method comprises determining a bounding box of the lane marking based on sensor data of a sensor system installed at the automated vehicle, determining a focus area being located completely inside the bounding box, and determining (S4) the color of the lane marking by determining a color inside the focus area, wherein a size of the focus area depends on a size of the bounding box.