Roadway Course Determination Using Grayscale Object Tracking

Solution Overview

Problem

Existing methods for determining the course of a roadway in front of a vehicle are inefficient, particularly due to dependency on color information and susceptibility to distortion from artificial lighting, which increases processing effort and accuracy issues.

Innovation Solution

A method and device that determine the roadway course using image data from a camera, without relying on color evaluation, by identifying and tracking objects like delineators and streetlights, and assigning them to a structure based on regular distances, utilizing a vehicle coordinate system and polynomial equations to describe the road course, and verifying the course with road models and construction regulations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If color information is used to determine the course of the roadway, then object classification accuracy is improved, but processing effort increases and reliability decreases due to distortion from artificial lighting

Engineering Contradiction:

Improveobject classification accuracyVSAvoidprocessing effort

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and removes the dependency on color information from the roadway determination process. By using only grayscale intensity values from the image data instead of full color information, the system eliminates the problems of artificial lighting distortion while maintaining sufficient accuracy for detecting roadway features like lane markings and curbs.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses simple grayscale intensity thresholds that can be quickly adjusted or replaced based on lighting conditions, rather than relying on complex color analysis. This allows the system to adapt to different environments without requiring sophisticated color processing algorithms, reducing both computational complexity and improving reliability.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

2Measurement precision

If color information is used to determine the course of the roadway, then object classification accuracy is improved, but processing effort increases

Engineering Contradiction:

Improveobject classification accuracyVSAvoidprocessing efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent extracts and removes the dependency on color information from the roadway determination process. By using only grayscale intensity values from the image data instead of full color information, the system eliminates the problems of artificial lighting distortion while maintaining sufficient accuracy for detecting roadway features like lane markings and curbs.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses simple grayscale intensity thresholds that can be quickly adjusted or replaced based on lighting conditions, rather than relying on complex color analysis. This allows the system to adapt to different environments without requiring sophisticated color processing algorithms, reducing both computational complexity and improving reliability.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

3Adaptability or versatility

If delineators are detected to determine the course of the road, then roadway course determination is achieved in scenarios with poor or missing lane markings, but object detection complexity increases

Engineering Contradiction:

Improveroadway determination capability in various scenariosVSAvoidobject detection complexity

Core Design Contradiction:

Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent creates a universal object detection system that can identify multiple types of roadway features (lane markings, curbs, delineators, streetlights) using the same grayscale-based image processing approach. This multi-functional capability allows the system to adapt to different roadway scenarios without requiring separate detection algorithms for each feature type.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent changes the detection parameters based on the expected feature type and lighting conditions by adjusting grayscale intensity thresholds and spatial filtering parameters. This allows the system to optimize detection sensitivity for different objects (e.g., reflective delineators vs. painted lane markings) without fundamentally changing the detection approach, thereby reducing overall system complexity.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentEP2116958B1Method and device for calculating the drive path in the area in front of a vehicle

Publication Date: 2016.02.10 HELLA GMBH & CO KGAA

Figure 1
Figure 2
Figure 3

AI summary

The invention relates to a method and a device for determining the road alignment (10, 32, 34) in the area in front of a vehicle (46). Starting from provided image data of at least one image of the area in front of the vehicle (46), at least two objects are identified in the at least one image described by the image data. It is checked whether the identified objects (12 to 30, 38, 44) can be assigned to the same object type. Furthermore, it is checked whether, for one of the two object types (12 to 30, 38, 44), the object property is preset to state that objects (12 to 30, 38, 44) of this object type are arranged in a structure corresponding to the road alignment (10, 32, 34). Finally, at least one road alignment (10, 32, 34) is determined depending on the positions of the identified objects (12 to 30, 38, 44) in the image.