Camera Device Road Shape Estimation via 3D Object Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing vehicle safety systems struggle to accurately detect road curves and determine the need for deceleration when white lines or roadside three-dimensional objects are difficult to detect, leading to potential accidents.

Innovation Solution

A camera device with multiple image capturing units that detect three-dimensional objects near the vanishing point of the road and estimate road shapes, enabling accurate determination of curve presence and necessity for deceleration, even in situations where white lines or roadside objects are hard to detect.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If white line detection is used to detect road shape, then detection accuracy is improved when white lines are visible, but detection reliability deteriorates when white lines are absent or difficult to recognize

Engineering Contradiction:
Improveroad shape detection accuracyVSAvoiddetection reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces three-dimensional objects (such as roadside objects, vehicles, or environmental features) as intermediary reference points to detect road shape. When white lines are absent or difficult to recognize, the system uses these three-dimensional objects as alternative mediators to calculate road curvature and shape, ensuring continuous and reliable detection under various road conditions.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system dynamically changes the detection parameter from two-dimensional white line features to three-dimensional object features based on detection conditions. When white line detection fails or reliability is low, the system switches to using depth information and spatial relationships of three-dimensional objects, changing the fundamental parameter basis for road shape detection.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If in-vehicle camera is used to detect distant white lines, then road shape detection is possible, but detection accuracy deteriorates at high traveling speeds due to difficulty in detecting far white lines

Engineering Contradiction:
Improvedetection capabilityVSAvoidcurvature detection accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent transitions from two-dimensional image plane detection to three-dimensional spatial detection by incorporating depth information from multiple image capturing units. This dimensional change allows the system to accurately measure distances and positions of three-dimensional objects, enabling precise curvature calculation even for distant road features at high speeds where traditional 2D detection fails.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system compensates for the limitations of single-camera detection by using multiple image capturing units that provide counterbalancing perspectives. This multi-view approach compensates for the reduced accuracy of distant object detection, allowing the system to maintain high measurement precision for road curvature even when detecting far-away features at high traveling speeds.

Inventive Principle:
Principle #8Anti-weight (Counterweight)

3Reliability

If roadside three-dimensional object detection is used, then detection reliability is improved when objects are present, but detection accuracy deteriorates when no three-dimensional objects exist by the roadside

Engineering Contradiction:
Improvedetection reliabilityVSAvoidroad shape detection accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent creates a universal detection system that can function using multiple different types of features - both two-dimensional white lines and three-dimensional objects. The system is designed to handle various road environments uniformly, switching between detection methods based on availability, ensuring both reliability and accuracy across diverse conditions whether white lines, three-dimensional objects, or both are present.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Device complexity

If single image capturing unit is used, then device complexity is reduced, but measurement precision of distant road shape deteriorates

Engineering Contradiction:
Improvesystem complexityVSAvoiddistant curvature detection precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

By adding the dimension of depth perception through multiple image capturing units, the system achieves accurate three-dimensional reconstruction of distant road features. This dimensional enhancement allows precise measurement of distant curvature that cannot be achieved with single-camera 2D imaging, as the multi-unit system can triangulate positions and calculate true spatial relationships.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP2372669B1Camera device
Publication Date: 2020.08.19 HITACHI AUTOMOTIVE SYST LTD
  • EP2372669B1 patent drawingFigure 1
  • EP2372669B1 patent drawingFigure 2~3
  • EP2372669B1 patent drawingFigure 4~5

AI summary

Provided is a camera device which is capable of estimating a road shape ahead of a target vehicle or capable of determining whether or not the target vehicle needs to be decelerated by controlling a brake before a curve, even in the situation where a white line of a traveling road or a roadside three-dimensional object is difficult to detect. A camera device 105 including a plurality of image capturing units 107 and 108 which each take an image of a traveling road ahead of a target vehicle 106, includes: a three-dimensional object ahead detection unit 114 which detects three-dimensional objects ahead 101 existing in a vicinity of a vanishing point of the traveling road 102 on the basis of the images picked up by the plurality of image capturing units 107 and 108; and a road shape estimation unit 113 which estimates a road shape of a distant portion on the traveling road 102 on the basis of a detection result detected by the three-dimensional object ahead detection unit 114.