Stereo Distance Detection with Selective Region Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing object distance detection systems face a trade-off between computational load and accuracy, with reduced image resolution leading to potential accuracy loss and increased processing requirements.

Innovation Solution

An object distance detection device utilizing multiple imaging units with image correction and stereo distance detection, combined with selective region processing and monocular distance detection, to enhance accuracy while reducing computational load.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If image resolution is reduced to decrease computational load, then processing speed improves, but object detection accuracy deteriorates

Engineering Contradiction:
Improveprocessing speedVSAvoidobject detection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The imaging area is divided into multiple regions, with different processing strategies applied to each region. High-priority regions maintain full resolution for accurate detection, while other regions use reduced resolution to decrease computational load overall.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different image processing qualities are applied to different regions of the image. Critical regions receive full-resolution processing for high accuracy, while non-critical regions use reduced resolution, optimizing the balance between processing speed and detection accuracy locally.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If full-resolution image processing is used to improve object detection accuracy, then measurement precision improves, but computational load increases

Engineering Contradiction:
Improveobject detection accuracyVSAvoidcomputational load
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The image processing task is segmented into multiple regions, allowing the system to apply full-resolution processing only where necessary while using reduced resolution elsewhere, thereby maintaining accuracy where needed while reducing overall computational load.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of processing the entire image at full resolution, the system applies full-resolution processing partially only to specific regions of interest, reducing the total computational load while maintaining sufficient accuracy for detection purposes.

Inventive Principle:
Principle #16Partial or excessive action

3Device complexity

If stereo distance detection is performed on reduced images, then computational load decreases, but reliability of distance measurement deteriorates

Engineering Contradiction:
Improvecomputational loadVSAvoiddistance measurement reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The system segments the image processing into different resolution levels, performing stereo distance detection on strategically selected regions. This segmentation allows reduced computational load while maintaining reliable distance measurements in critical areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different processing qualities are applied locally across the image, with high-quality stereo processing applied to regions where accurate distance measurement is critical, while other regions use reduced processing, optimizing the balance between computational load and measurement reliability.

Inventive Principle:
Principle #3Local quality

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

The device achieves improved object detection accuracy and decreased computational load by optimizing image processing through selective region detection and monocular methods, maintaining accuracy across varying distances.

Implementation Method 1

stereo distance detection is performed by dividing a region with an image whose resolution is not reduced

Methodology Applied
Scientific EffectParallax: Parallax

Data Source

PatentEP3690812B1Object distance detection device
Publication Date: 2025.12.17 ASTEMO LTD
  • EP3690812B1 patent drawingFigure 1
  • EP3690812B1 patent drawingFigure 2
  • EP3690812B1 patent drawingFigure 3~4

AI summary

The present invention provides an object distance detection device with which object detection accuracy can be improved and computational load can be decreased. The present invention comprises: a stereo distance detection unit 105 that performs a search, within a reference image captured by an imaging unit 102, for an image element corresponding to a specific image element within a reference image captured by another imaging unit 101, and that detects an external-environment object and the distance to the object on the basis of parallax obtained by the search; and a distance detection unit 107 that, on the basis of the results detected by the stereo distance detection unit 105, detects the distance to an object in a region selected using a processing region setting unit 106 and selected from external-environment information obtained by an imaging unit 103.