Image Processing Apparatus Two-Dimensional Search Area for Corresponding Pixel Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing technologies face challenges in detecting corresponding points between images with different optical parameters, leading to either failed detections or increased processing loads due to the need for precise optical parameter knowledge.

Innovation Solution

The implementation of a method that sets a two-dimensional search area based on the range of optical parameter variations, allowing for the detection of corresponding points with reduced processing load even when precise optical parameters are unknown, by considering changes in epipolar lines and optical axis directions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If corresponding pixels are searched only on the epipolar line based on optical parameters, then the processing load is reduced, but the detection precision deteriorates when optical parameters are not precisely known

Engineering Contradiction:
Improveprocessing loadVSAvoiddetection precision of corresponding pixels
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The invention transforms the one-dimensional epipolar line search into a two-dimensional search area. By defining a search area that extends perpendicular to the epipolar line, the system accounts for optical parameter variations while maintaining reduced processing complexity compared to searching the entire image.

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

Solution Approach 2:

The invention adjusts the search strategy by introducing a search area that varies based on optical parameter ranges. The search area's size and position are determined by the expected variation in optical parameters, allowing the system to adapt to parameter uncertainties without exhaustive searching.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If corresponding pixels are searched in the whole search image, then the detection precision is improved, but the processing load becomes excessively heavy

Engineering Contradiction:
Improvedetection precision of corresponding pixelsVSAvoidprocessing load
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The invention divides the search space into two components: the epipolar line (one-dimensional constraint) and the search area (two-dimensional region). This segmentation allows the search to be focused on a specific region rather than the entire image, significantly reducing processing load while maintaining detection precision.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention extends the search from the one-dimensional epipolar line to a two-dimensional search area. This dimensional expansion allows the system to account for optical parameter variations perpendicular to the epipolar line without requiring a full two-dimensional search of the entire image.

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

3Productivity

If the search area is limited to the epipolar line, then the processing load is reduced, but the reliability of corresponding point detection fails when optical parameters vary

Engineering Contradiction:
Improveprocessing loadVSAvoidreliability of corresponding point detection
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The invention prepares for optical parameter variations by defining a search area that acts as a buffer zone around the epipolar line. This search area is calculated based on expected parameter variations, providing a cushion that ensures corresponding pixels are not missed due to parameter uncertainties.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Solution Approach 2:

The invention makes the search area dynamic by basing its definition on optical parameter ranges. When optical parameters are expected to vary, the search area expands to accommodate these variations, ensuring reliable detection without requiring precise parameter knowledge.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3255888B1Image capturing apparatus, image processing method, and image processing program
Publication Date: 2023.02.01 CANON KK
  • EP3255888B1 patent drawingFigure 1A~1C
  • EP3255888B1 patent drawingFigure 2~3
  • EP3255888B1 patent drawingFigure 4~5

AI summary

An image processing apparatus processes a first image and a second image so as to detect a corresponding pixel in the second image which corresponds to a target pixel in the first image. The first image has a first parameter value, and the second image has a second parameter value different from the first parameter value. The first parameter value and the second parameter value are values of optical parameters of image capturing systems used to capture the first image and the second image. The image processing apparatus includes an area setter that sets a two-dimensional search area as a partial area in which the corresponding pixel is to be searched in the second image, based on a predetermined range in which each of the first and second parameter values can change, and a detector that detects the corresponding pixel by searching the two-dimensional search area.