Lane Line Detection Using Optical Flow Search Regions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing lane line detection technologies face significant computation burdens, which can lead to slow detection speeds, compromising the ability to rapidly identify lane lines and prevent accidents.

Innovation Solution

A method utilizing a dynamic vision sensor to generate an optical flow image, determining initial and subsequent search regions based on the optical flow, and iteratively calculating centers of gravity to detect lane lines without edge detection, thereby reducing computational load and improving detection speed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If edge detection and curve fitting methods are used for lane line detection, then detection accuracy is improved, but computation burden increases and detection speed decreases

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetection speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the image processing task by dividing it into multiple search regions rather than processing the entire image. Each search region is processed independently to detect lane lines, which reduces the overall computational burden while maintaining detection accuracy through focused local analysis.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the necessary features (optical flow information and center of gravity) from the image data, ignoring unnecessary details. This selective extraction approach reduces computation by focusing only on relevant characteristics needed for lane line detection.

Inventive Principle:
Principle #2Taking out (Extraction)

2Loss of information

If traditional edge detection methods are applied to extract lane line information, then comprehensive lane line features are obtained, but computation resources are heavily consumed

Engineering Contradiction:
Improvefeature completenessVSAvoidcomputation resources
Core Design Contradiction:
Loss of informationVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential optical flow information and center of gravity characteristics from the image, discarding unnecessary edge detection computations. This selective feature extraction maintains sufficient lane line detection capability while significantly reducing computation resource consumption.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses optical flow information as a simplified representation or 'copy' of the full image data, which contains the necessary motion and lane line information without requiring processing of the complete original image, thus reducing computational requirements.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS11941891B2Method for detecting lane line, vehicle and computing device
Publication Date: 2024.03.26 OMNIVISION SENSOR SOLUTION (SHANGHAI) CO LTD
  • US11941891B2 patent drawing
  • US11941891B2 patent drawing
  • US11941891B2 patent drawing

AI summary

The present disclosure provides a method for detecting a lane line, a vehicle and a computing device. The method includes: generating an optical flow image in accordance with a series of event data from a dynamic vision sensor coupled to a vehicle; determining an initial search region including a start point of the lane line in accordance with the optical flow image; determining a center of gravity of the initial search region; determining a new search region through an offsetting operation on the center of gravity; determining a center of gravity of the new search region; repeating the steps of determining a new search region and determining a center of gravity of the new search region iteratively to acquire centers of gravity of a plurality of search regions; and determining the lane line in accordance with the centers of gravity of the plurality of search regions.