Road Marker Boundary Detection With Integer DWT Filters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing methods for detecting road markers are computationally expensive and inefficient, particularly in real-time systems, due to the use of orthogonal wavelets with complex filter coefficients, and struggle with varying lighting and noise conditions.

Innovation Solution

A method using discrete wavelet transform (DWT) with symmetrical, integer-valued filters based on Pascal's triangle coefficients to decompose images, focusing on low-frequency components to enhance contrast and detect road markers efficiently, and optionally using a trained statistical classifier for verification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If orthogonal wavelets with complex filter coefficients are used for image processing, then measurement precision of road markers is improved, but device complexity and computational cost increase significantly

Engineering Contradiction:
Improveroad marker detection accuracyVSAvoidfilter coefficient complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces complex orthogonal wavelet filters with simple integer-valued symmetrical filters that can be implemented using basic arithmetic operations. These simplified filters act as disposable, computationally inexpensive alternatives to the complex filters, achieving adequate detection precision without the computational burden of orthogonal wavelets.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent changes the parameters of the filter coefficients from complex orthogonal wavelet values to simple integer values based on symmetrical sequences. This parameter transformation maintains the essential filtering functionality while dramatically reducing computational complexity and enabling real-time processing.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If complex image processing algorithms are used to detect road markers, then measurement precision is improved, but processing time increases

Engineering Contradiction:
Improveroad marker boundary detection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent employs computationally inexpensive integer-based filtering operations that can be executed rapidly, replacing time-consuming complex algorithms. The simplified filters enable real-time processing by performing adequate road marker detection with minimal computational resources.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent extracts and processes only the essential low-frequency components of the image using simple averaging filters, discarding unnecessary high-frequency details. This selective extraction of critical information reduces processing time while maintaining adequate detection accuracy for road marker boundaries.

Inventive Principle:
Principle #2Taking out (Extraction)

3Productivity

If high computational power is used for real-time image processing, then productivity is improved, but device complexity increases

Engineering Contradiction:
Improvereal-time processing capabilityVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent uses simple integer arithmetic operations and basic filtering that require minimal computational power, replacing complex algorithms that would demand high-performance hardware. This enables real-time processing on standard vehicle processors without requiring specialized high-computational-power systems.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent replaces complex mechanical/computational systems with simplified arithmetic operations. The integer-based filters substitute for complex floating-point computations, enabling real-time processing with standard processors rather than requiring specialized high-performance computing hardware.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4148668B1Method and apparatus for processing an image of a road having a road marker to identify a region of the image which represents the road marker
Publication Date: 2025.10.29 APTIV SERVICES POLAND SA
  • EP4148668B1 patent drawingFigure 1
  • EP4148668B1 patent drawingFigure 2
  • EP4148668B1 patent drawingFigure 3

AI summary

A method of processing an image of a road having a road marker acquired by a vehicle-mounted camera to generate boundary data indicating a boundary of the road marker region of the image which represents the road marker, comprising: generating (S410) an LL sub-band image of an Mth level of an (M+1)-level discrete wavelet transform, DWT, decomposition of the image by iteratively low-pass filtering and down-sampling the image M times; generating (S420) a sub-band image of an (M+1)th level of the (M+1) level DWT decomposition by high-pass filtering the LL sub-band image and down-sampling a result of the high-pass filtering; and determining (S430) a boundary of a region of pixels of the sub-band image of the (M+1)th level, the region being surrounded by pixels having pixel values substantially different to the pixel values of the pixels in the region, the determined boundary indicating the boundary of the road marker region.