Stereo Image Parallax Calculation Using Motion Speed Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Stereo image processing devices face challenges in accurately determining parallax for identical objects due to the presence of multiple similar image regions and noise, leading to increased determination errors and decreased measurement accuracy.

Innovation Solution

The integration of information on the moving speed of imaging units into the stereo image processing method, which involves calculating predicted parallax and weighting similarity to suppress erroneous determinations, improves the accuracy of parallax calculation even in the presence of noise and multiple similar regions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If stereo matching is performed using only image similarity in a search range, then the process is simple, but matching errors increase when multiple similar regions exist or noise is present

Engineering Contradiction:
Improvematching process complexityVSAvoidparallax determination accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces motion speed as a new parameter to transform the static image matching problem into a dynamic one. By incorporating temporal information (motion between frames) into the matching criteria, the system can distinguish true matching regions from false similarities, thereby improving parallax determination accuracy without significantly increasing system complexity

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent performs preliminary motion estimation using optical flow or feature tracking before conducting stereo matching. This preliminary action provides expected motion compensation that guides the subsequent matching process, allowing the system to focus search efforts on regions with consistent motion patterns and reduce the impact of noise and false similarities

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the search range is expanded to cover more potential matching regions, then the likelihood of finding the real matching region increases, but the number of false similar regions also increases

Engineering Contradiction:
Improvematching success rateVSAvoidparallax determination accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

By adding motion speed as a filtering parameter, the patent enables the system to maintain a broad search range while effectively filtering out false matches. Regions that appear similar but have inconsistent motion patterns are eliminated from consideration, allowing the system to expand spatial search coverage without proportionally increasing false positive rates

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If noise filtering is applied to remove dirt and circuit noises from images, then measurement accuracy improves, but processing time and complexity increase

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

Solution Approach 1:

The patent performs noise filtering as a preliminary step before stereo matching, removing artifacts and inconsistencies from the input images. By addressing noise early in the processing pipeline, subsequent matching operations work with cleaner data, improving overall accuracy while the time penalty is paid once rather than repeatedly during iterative matching processes

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2910897B1Stereo image processing device and stereo image processing method
Publication Date: 2024.03.20 ASTEMO LTD
  • EP2910897B1 patent drawingFigure 1
  • EP2910897B1 patent drawingFigure 2
  • EP2910897B1 patent drawingFigure 3~4

AI summary

To provide a stereo image processing device and a stereo image processing method that can suppress a decrease in the determination accuracy of parallax for an identical object due to mixture of noise and the like, the device includes a pair of imaging units 101a, 101b; a similarity calculation unit 106b that calculates similarity for each parallax for the pair of images; a parallax calculation unit 106c that calculates parallax for an identical object on the basis of the similarity for each parallax; a parallax data buffer unit 105 that stores data on the parallax; a speed detection unit 107 that detects a moving speed of the pair of imaging units 101a,101b; and a parallax prediction unit 106a that calculates a predicted parallax value on the basis of the moving speed and past data on parallax stored in the parallax data buffer unit 105. The parallax calculation unit 106 calculates parallax for an identical object on the basis of the similarity for each parallax and the predicted parallax value.