Monocular Camera Road Marking Identification via Intensity Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing visual camera systems struggle to accurately distinguish road markings and objects, such as traffic cones and road dotted lines, due to their similarity in images captured by monocular cameras.

Innovation Solution

A method and device for identifying road markings using a monocular camera, which involves acquiring an original image, determining image intensity distributions in specific regions, and comparing these intensities to determine if an object is a road marking, with additional steps including image preprocessing, edge point grouping, and linear regression compensation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a monocular camera is used to capture road images, then the device complexity is reduced and cost is lowered, but the ability to accurately distinguish road markings from objects deteriorates

Engineering Contradiction:
Improvecamera system complexityVSAvoidroad marking identification accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent divides the image processing task into multiple stages: preprocessing to obtain differential images, edge detection to identify edge points, grouping edge points by continuity, and classification based on geometric features. This segmentation allows the monocular camera system to accurately distinguish road markings from objects through systematic analysis of edge point distributions and geometric characteristics.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If a stereo camera with two cameras is used to distinguish road markings and objects, then the identification accuracy is improved, but the device complexity and cost increase

Engineering Contradiction:
Improveroad marking identification accuracyVSAvoidcamera system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent creates a virtual stereo effect by detecting edge points in the monocular camera image and grouping them according to continuity. It then estimates road marking positions and orientations by analyzing the spatial relationships between edge point groups, effectively copying the functional capability of a stereo camera system using only a single camera and image processing algorithms.

Inventive Principle:
Principle #26Copying

3Measurement precision

If multiple images at different distances are captured to distinguish road markings and objects, then the identification accuracy is improved, but the loss of time increases

Engineering Contradiction:
Improveroad marking identification accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary edge detection and edge point grouping on the captured image to identify potential road markings before final classification. By pre-processing the image to extract edge features and organize them into continuous groups, the system prepares the data structure needed for rapid road marking identification, reducing the time required for subsequent analysis and enabling real-time processing.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12300003B2Method for identifying road markings and monocular camera
Publication Date: 2025.05.13 GREAT WALL MOTOR CO LTD
  • US12300003B2 patent drawing
  • US12300003B2 patent drawing
  • US12300003B2 patent drawing

AI summary

A method and a device for identifying road markings, and a monocular camera is provided. The method includes: acquiring an original image captured by a monocular camera on a vehicle; determining a middle region where an object is located and left and right regions on both sides of the middle region from the original image; acquiring an image intensity distribution along a vertical direction of the original image in the left region, the right region and the middle region; and determining whether the object is a road marking by comparing an image intensity of the middle region relative to image intensities of the left region or the right region. The present application can achieve the object of accurately identifying road markings.