Dual Imaging Unit Exposure Control for Road Sign Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional on-vehicle image processing devices struggle to recognize road signs and other objects of varying brightness without interfering with stereo processing, leading to issues like blown-out highlights and blocked shadows, which affects recognition accuracy and compatibility with multiple recognition applications.

Innovation Solution

The device employs a dual-camera system with exposure control units that capture images for stereo processing and road sign recognition using different exposure control settings, allowing for simultaneous recognition of road signs with varying brightness without disrupting each other's processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If exposure control is optimized for recognizing internally illuminated signs, then recognition accuracy of bright signs improves, but stereo processing quality deteriorates due to blocked shadows

Engineering Contradiction:
Improverecognition accuracyVSAvoidstereo processing quality
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent divides the imaging function into two separate imaging units: one dedicated to monocular processing for road sign recognition and another dedicated to stereo processing for distance calculation. This segmentation allows each unit to use exposure settings optimized for its specific function without compromising the other.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts the road sign recognition function from the stereo processing function by using a separate imaging unit. This extraction enables independent optimization of exposure control for each function, resolving the conflict between recognizing bright internally illuminated signs and maintaining quality for stereo processing.

Inventive Principle:
Principle #2Taking out (Extraction)

2Measurement precision

If exposure control is optimized for recognizing signboard signs, then recognition accuracy of dark signs improves, but recognition accuracy of internally illuminated signs deteriorates due to blown-out highlights

Engineering Contradiction:
Improverecognition accuracyVSAvoidblown-out highlights
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent segments the imaging function so that road sign recognition uses a dedicated imaging unit with exposure optimized for various brightness conditions, while stereo processing uses a separate unit. This allows the monocular unit to capture both dark signboard signs and bright internally illuminated signs without the compromise affecting stereo processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different exposure control strategies to different imaging units based on their specific functions. The monocular imaging unit uses exposure settings suitable for recognizing road signs of varying brightness, while the stereo imaging unit uses settings optimized for depth perception, allowing each to maintain optimal quality for its purpose.

Inventive Principle:
Principle #3Local quality

3Device complexity

If a single imaging unit is used for both stereo processing and road sign recognition, then device complexity is reduced, but processing speed deteriorates due to conflicting exposure requirements

Engineering Contradiction:
Improvenumber of imaging unitsVSAvoidprocessing speed
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent segments the imaging system into two parallel processing paths with dedicated imaging units, enabling simultaneous execution of stereo processing and road sign recognition without exposure conflicts. This segmentation increases device complexity but resolves the processing speed bottleneck caused by sequential or conflicting operations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent combines multiple imaging units into a unified system that processes different image types simultaneously. By merging the monocular and stereo imaging capabilities into one integrated device, it achieves both high processing speed and comprehensive functionality despite the increased hardware complexity.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3637758B1Image processing device
Publication Date: 2024.09.04 ASTEMO LTD
  • EP3637758B1 patent drawingFigure 1~2
  • EP3637758B1 patent drawingFigure 3(a)~4
  • EP3637758B1 patent drawingFigure 5~6

AI summary

The purpose of the present invention is to provide an image processing device capable of obtaining a captured image in which various objects different in brightness can be recognized without exerting a bad influence on other applications. This image processing device is characterized by being provided with: an imaging unit for capturing an image; one exposure control unit for performing exposure control of the imaging device using a first control amount; another exposure control unit for performing exposure control of the imaging unit using a control amount based on the first control amount; and an exposure control switching unit for switching between performing exposure control of the imaging device by means of the one exposure control unit and by means of the other exposure control unit.