Vehicle Surround Alert System Using Stereo Camera Overlap

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional camera-based driver assistance systems have limitations in capturing a wide field of view and require multiple cameras, which increases complexity and cost, and often rely on stereo camera pairs that are limited in their field of vision and require additional hardware for accurate distance estimation.

Innovation Solution

A system using multiple wide-angle cameras positioned at strategic locations on a vehicle, including outer positions, captures overlapping images to determine object positions stereoscopically, allowing for precise warning generation without the need for additional hardware, and can be integrated into existing surround view systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If stereo camera pairs with narrow field of view are used to achieve high degree of distance vision, then measurement precision of distance is improved, but area of stationary object (field of vision) deteriorates

Engineering Contradiction:
Improvedistance vision precisionVSAvoidfield of vision
Core Design Contradiction:
Measurement precisionVSArea of stationary object

Solution Approach 1:

The system divides the monitoring task into multiple segments by using several wide-angle cameras positioned at different locations (front, rear, left, right) to cover different areas, collectively achieving complete surrounding coverage while maintaining wide field of view for each camera

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from using two cameras in a stereo pair configuration to using multiple cameras arranged in a spatial distribution around the vehicle, adding dimensional coverage in all directions (front, rear, left, right) while maintaining wide field of view, thereby resolving the contradiction between coverage area and depth perception capability

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

2Area of stationary object

If multiple wide-angle cameras are used to capture large field of view, then area of stationary object (field of vision) is improved, but device complexity increases

Engineering Contradiction:
Improvefield of visionVSAvoidcamera system complexity
Core Design Contradiction:
Area of stationary objectVSDevice complexity

Solution Approach 1:

The wide-angle cameras serve multiple functions simultaneously: they capture wide field of view for surround view display, enable distance estimation through image overlap analysis, and provide collision warning data, thereby reducing the need for separate specialized sensors and systems

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

Solution Approach 2:

The system uses the image data from wide-angle cameras for multiple purposes including distance estimation and collision warning without requiring additional specialized hardware, allowing the existing camera system to serve itself for multiple safety functions

Inventive Principle:
Principle #25Self-service

3Measurement precision

If stereo camera pairs are used for distance estimation, then measurement precision of distance is improved, but device complexity (additional hardware) increases

Engineering Contradiction:
Improvedistance estimation accuracyVSAvoidhardware requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system creates a virtual stereo effect by using images from widely separated wide-angle cameras that capture overlapping areas, effectively copying the distance estimation function of traditional stereo cameras without requiring precise mechanical alignment or specialized stereo camera hardware

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The system changes the operational parameters by using wide-angle lenses with large opening angles and positioning cameras at maximum distance from each other, which fundamentally alters how distance estimation is achieved compared to traditional narrow-field stereo cameras, enabling distance measurement without specialized hardware

Inventive Principle:
Principle #35Parameter changes

4Measurement precision

If classifiers are used for distance estimation in wide-angle cameras, then measurement precision is improved, but computational effort increases

Engineering Contradiction:
Improvedistance estimation accuracyVSAvoidcomputational effort
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system replaces complex computational classification methods with a geometric approach based on triangulation and image overlap analysis, substituting algorithmic complexity with straightforward mathematical calculations from multiple camera perspectives

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach provides reliable and precise warnings about vehicle surroundings, especially for distant objects, with reduced computational effort and hardware requirements, enhancing safety by minimizing unnecessary warnings and activating appropriate safety systems.

Implementation Method 1

The position information in the first overlapping area is determined stereoscopically based on the images from the first and second cameras

Methodology Applied
Scientific EffectStereoscopy: Parallax

Data Source

PatentEP3078015B1Method and device for generating an alert by means of two images of a vehicle environment obtained via cameras
Publication Date: 2017.12.13 ROBERT BOSCH GMBH
  • EP3078015B1 patent drawingFigure 1
  • EP3078015B1 patent drawingFigure 2
  • EP3078015B1 patent drawingFigure 3

AI summary

The invention relates to a device and a method for generating an alert. To this end, a first image of a vehicle environment of a vehicle (10) is obtained via a first camera (1R) from a first camera perspective, and a second image of the vehicle environment of the vehicle (10) is obtained via a second camera (1L) from a second camera perspective. The first camera (1R) and the second camera (1L) are two separate cameras arranged in external positions on the vehicle (10), and arranged at a distance from one another in a longitudinal direction (L) or a transverse direction (Q) of the vehicle (10). The first and the second images overlap in a first overlapping region (6, 6'). An item of positional information, defining a position of an object or a expected position of the object in relation to the vehicle (10), is provided in the first overlapping region (6, 6') based on the images of the first and second cameras (1R, 1L). An alert is emitted if the positional information falls in a predetermined alert region in relation to the vehicle (10).