Omnidirectional Image Rectification for Vehicle Cross Traffic

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Omnidirectional camera images used in vehicle imaging systems are distorted, making it difficult for drivers to accurately determine the distance and speed of surrounding vehicles, limiting safety benefits.

Innovation Solution

A computer-implemented method and system that transforms omnidirectional images into rectilinear images using a panel model, reducing distortion and allowing for more accurate distance and speed assessments by mapping omnidirectional image pixels to rectilinear image pixels and world coordinates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of moving object

If an omnidirectional camera is used to capture ultra-wide angle views, then the field of view is improved and cross-traffic detection is enhanced, but image distortion increases making distance and speed determination difficult

Engineering Contradiction:
Improvefield of viewVSAvoiddistance and speed determination
Core Design Contradiction:
Area of moving objectVSMeasurement precision

Solution Approach 1:

The omnidirectional image is divided into multiple regions of interest (ROIs), each corresponding to a specific viewing direction. Each ROI is transformed independently into a separate rectilinear image, allowing precise measurement in each segment while maintaining the overall wide field of view coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A coordinate transformation system acts as an intermediary between the omnidirectional camera output and the display system. The system uses spherical coordinate system conversions and panel models to mathematically transform distorted omnidirectional images into undistorted rectilinear images, preserving measurement accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If the omnidirectional image is displayed directly to the driver, then the system complexity is reduced, but the driver's ability to assess distance and speed accurately deteriorates

Engineering Contradiction:
Improvesystem complexityVSAvoiddistance and speed assessment
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The system performs coordinate transformations and image rectification in advance, before displaying the images to the driver. By pre-processing the omnidirectional images into rectilinear format with accurate spatial relationships, the system eliminates the need for complex real-time calculations during driver interaction while ensuring measurement precision.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If multiple rectilinear images are generated from different regions of interest, then the measurement accuracy is improved, but the processing time and computational load increase

Engineering Contradiction:
Improvedistance gauging accuracyVSAvoidimage processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The omnidirectional image is segmented into multiple regions of interest, each processed independently into a rectilinear image. This segmentation allows parallel processing of different ROIs, reducing overall processing time while maintaining high measurement accuracy in each region through dedicated coordinate transformations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system processes only the necessary regions of interest at any given time rather than the entire omnidirectional image. By focusing computational resources on relevant ROIs based on vehicle context and driver needs, the system achieves sufficient measurement accuracy with reduced processing time and computational load.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3326146B1Rear cross traffic - quick looks
Publication Date: 2023.03.01 ROBERT BOSCH GMBH
  • EP3326146B1 patent drawingFigure 1
  • EP3326146B1 patent drawingFigure 2~3
  • EP3326146B1 patent drawingFigure 4~5

AI summary

System and method of performing various transformations between an omnidirectional image model and a rectilinear image model for use in vehicle imaging systems. In particular, a panel transform system and method for transforming an omnidirectional image from an omnidirectional camera positioned on a vehicle to a rectilinear image. The rectilinear image is then displayed to the driver for use while performing vehicle maneuvers. The panel transform system and method also provide a rectilinear image model based on the omnidirectional camera for use with existing rectilinear imaging processing systems. The rectilinear image model is determined based on a variable set of input parameters that are defined by both automatic and manual system inputs such as a steering wheel angle sensor and a user interface.