Camera Array Pixel Distribution Adjustment for Region-Specific Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing super-resolution image acquisition methods using camera arrays struggle to meet diverse user demands for imaging quality across different regions of a scene, as they uniformly distribute pixels, failing to prioritize high-resolution areas like human faces over scenery.
Innovation Solution
The method adjusts pixel point distribution in camera array image sensors to concentrate more pixels in regions of interest, allowing for higher resolution and signal-to-noise ratio (SNR) in those areas, enhancing imaging quality by acquiring multiple images with slight vision deviations and combining them to form super-resolution images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pixel points are uniformly distributed across the image sensor, then the imaging coverage is comprehensive, but the resolution and SNR in specific regions of interest cannot be improved
Solution Approach 1:
The patent applies local quality by adjusting pixel point distribution to be non-uniform across the image sensor. Specifically, pixel points are concentrated in regions of interest (such as human faces detected through eye position detection) while maintaining appropriate coverage in other areas. This allows higher resolution and signal-to-noise ratio in critical regions without completely sacrificing overall scene coverage, directly resolving the contradiction between uniform coverage and region-specific quality enhancement.
2Measurement precision
If more pixel points are concentrated in specific regions, then the resolution and SNR in those regions improve, but the pixel coverage of other regions decreases
Solution Approach 1:
The patent employs dynamics by making the pixel point distribution adjustable and adaptive rather than fixed. The system dynamically determines the optimal pixel distribution based on real-time scene analysis, particularly through eye position detection to identify regions of interest. This dynamic adjustment allows the system to concentrate pixels in critical areas when needed while maintaining broader coverage when regions of interest are not present, thus resolving the contradiction between localized quality enhancement and overall coverage.
3Measurement precision
If a camera array is used to acquire super-resolution images, then the overall resolution improves, but the ability to meet diverse user demands for different regions is limited
Solution Approach 1:
The patent implements feedback by using eye position detection to identify regions of interest and then adjusting pixel point distribution accordingly. The system continuously monitors the scene, detects human eyes, determines the pupil center positions, and uses this information to dynamically adjust where pixel points should be concentrated. This feedback loop enables the camera array to adapt to diverse user demands by prioritizing regions that are most likely to contain subjects of interest, thereby enhancing super-resolution capability where it matters most while maintaining flexibility for different shooting scenarios.
Data Source
AI summary
Embodiments of the present application disclose various super-resolution image acquisition methods and acquisition apparatus, wherein one of super-resolution image acquisition methods comprises: determining a to-be-adjusted region of an image sensor of at least one camera in a camera array, the to-be-adjusted region being an imaging region of the image sensor of the at least one camera corresponding to a first region, and the first region being at least a local part of a scene; adjusting pixel point distribution of the image sensor of the at least one camera, to increase the number of pixel points distributed in the to-be-adjusted region; acquiring images of the scene respectively by cameras of the camera array after adjustment; and acquiring a super-resolution image of the scene according to the acquired images. The embodiments of the present application improve imaging quality of a sub-image in the super-resolution image corresponding to the first region.