Parallax Image Ghost Reduction via Gain Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods require multiple image captures to effectively reduce unnecessary components like ghosts and flares, which is not suitable for still or moving object photography, especially when optical paths for these components are displaced from ideal pupil-divided paths.
Innovation Solution
An image processing method and apparatus that generate and process parallax images to determine unnecessary component intensity using gain distribution, allowing for their reduction in a single image capture by calculating difference information, determining gain distribution, and applying it to reduce unwanted components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple image captures are performed to detect and remove unnecessary components, then the quality of captured images is improved, but the productivity and ease of operation deteriorate due to the requirement of multiple captures
Solution Approach 1:
The patent segments the optical path into multiple pupil regions (first pupil region and second pupil region) that are displaced relative to each other. By capturing images through these separated optical paths simultaneously in a single shot, the system can detect unnecessary components through comparison without requiring multiple captures, thus resolving the contradiction between image quality and capturing efficiency
Solution Approach 2:
The patent introduces a spatial dimension by displacing the pupil regions in the optical path. This dimensional separation allows the system to obtain multiple perspectives (different pupil regions) simultaneously in a single image capture, enabling unnecessary component detection without multiple shots and thereby improving both quality and productivity
2Measurement precision
If multiple image captures are performed to detect unnecessary components, then the quality of captured images is improved, but the complexity of the device increases
Solution Approach 1:
The patent makes the single image pickup apparatus perform multiple functions: it captures images through displaced pupil regions, detects unnecessary components by comparing the images, and removes these components through gain distribution calculation. This multi-functionality in a single device avoids the need for complex multiple-capture systems while maintaining high image quality
3Measurement precision
If pupil-divided optical path is used to reduce unnecessary components, then the quality of captured images is improved, but the reliability decreases when the optical path for ghost is displaced from ideal pupil-divided path
Solution Approach 1:
The patent changes the parameter of pupil region displacement by intentionally creating a displaced optical path configuration. This displacement allows the system to capture images through different pupil regions that can effectively detect unnecessary components even when the ghost optical path is not ideal, thereby improving the reliability of ghost reduction while maintaining image 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
Effectively determines and reduces unnecessary components in captured images without the need for multiple captures, even when components pass through different pupil regions, improving image quality by minimizing ghosts and flares.
Implementation Method 1
an image pickup device configured to photoelectrically convert an optical image formed via an optical system to output a plurality of parallax images
Data Source
AI summary
An image processing apparatus (204) includes a generator (204a) which generates difference information relating to a plurality of parallax images, a gain distribution determiner (204b) which determines a gain distribution based on the difference information, an intensity determiner (204c) which determines an unnecessary component intensity based on the gain distribution, and a reducer (204d) which generates an unnecessary component reduction image in which an unnecessary component is reduced based on the parallax image and the unnecessary component intensity.