Viewpoint Image Shading Correction for Phase Detection Autofocus

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image capturing apparatuses face issues with image quality degradation due to flaws such as shading caused by pupil division, saturation signals, and defects in viewpoint images, leading to suboptimal performance in focus detection and image processing.

Innovation Solution

An image processing apparatus and method that corrects viewpoint images by performing shading correction and defect correction using a captured image, generating improved viewpoint images through pixel sub-pixel signal processing and polynomial function-based shading correction, thereby enhancing image quality and focus detection accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If pupil division is performed to enable focus detection, then focus detection capability is improved, but shading and image quality degradation occur in viewpoint images

Engineering Contradiction:
Improvefocus detection accuracyVSAvoidviewpoint image quality
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

A shading correction unit is introduced as an intermediary component that processes viewpoint images to remove shading effects. This unit receives viewpoint images containing shading artifacts and outputs corrected images, thereby resolving the contradiction between maintaining pupil division for focus detection and eliminating shading degradation in viewpoint images.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The shading component is extracted and separated from the viewpoint images through dedicated shading correction processing. By identifying and removing the shading artifact as a distinct element, the system preserves the useful focus detection information while eliminating the harmful shading effect that degrades image quality.

Inventive Principle:
Principle #2Taking out (Extraction)

2Adaptability or versatility

If multiple divided photoelectric conversion units are used to capture light from different pupil regions, then phase difference focus detection is enabled, but saturation signals and defects are generated

Engineering Contradiction:
Improvefocus detection functionalityVSAvoidviewpoint image quality
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

A defect correction unit implements feedback processing by analyzing viewpoint images for saturation signals and defects, then correcting these issues in subsequent processing stages. This feedback mechanism ensures that focus detection functionality is maintained while image quality degradation from saturation and defects is systematically addressed and corrected.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Shading correction and defect correction are performed as preliminary processing steps before final viewpoint image generation and focus detection. By addressing shading and defects early in the processing chain, the system prevents these issues from propagating through subsequent stages, thereby maintaining both functionality and image quality.

Inventive Principle:
Principle #10Preliminary action

3Loss of information

If viewpoint images are generated from divided photoelectric conversion units, then light field information is obtained, but shading correction is required to improve image quality

Engineering Contradiction:
Improvelight field information retentionVSAvoidviewpoint image quality
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The shading correction unit serves as an intermediary processing stage that preserves light field information while removing shading artifacts. It processes the raw viewpoint images containing both valuable light field data and unwanted shading effects, outputting corrected images that retain the essential light field information without the degradation caused by shading.

Inventive Principle:
Principle #24Intermediary (Mediator)

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 solution effectively improves the quality of viewpoint images by correcting shading and defect issues, leading to enhanced focus detection and image processing capabilities, resulting in higher image fidelity and accuracy.

Implementation Method 1

an image sensor that generates a captured image and a plurality of viewpoint images by photoelectric conversion of light

Methodology Applied
Scientific EffectPhotoelectric conversion: Photoelectric Effect

Data Source

PatentEP3261328B1Image processing apparatus, image processing method, and computer-readable storage medium
Publication Date: 2021.10.13 CANON KK
  • EP3261328B1 patent drawingFigure 1
  • EP3261328B1 patent drawingFigure 2
  • EP3261328B1 patent drawingFigure 3A~3B

AI summary

An image processing apparatus comprises acquisition means configured to acquire a first viewpoint image (from subpixel 201) corresponding to a first partial pupil region of an exit pupil of an imaging optical system divided into a plurality of partial pupil regions in a first direction (X-direction, fig.2), and a captured image (from subpixels 201+202) corresponding to the exit pupil. The image processing apparatus comprises correction means configured to correct shading of a first pixel of a first pixel group based on a first ratio of a sum of the first pixel group of the first viewpoint image, said pixel group being arranged in a second direction (Y direction) orthogonal to the first direction to a sum of a pixel group of the captured image corresponding to a position of the first pixel group.