High Dynamic Range Image Synthesis via Depth-Field Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing high dynamic range (HDR) image synthesis methods struggle to produce clear images at each depth of field, leading to overexposure or underexposure due to shallow depth of field and insufficient dynamic range.
Innovation Solution
A method that obtains multiple images with different depths of field and exposure values during an HDR photographing operation, synthesizes images with the same exposure value to create full depth-of-field images, and then uses an HDR algorithm to generate a high dynamic range image that is clear at each depth of field.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If a larger aperture is used to improve light intake and dynamic range, then the dynamic range is improved, but the depth of field becomes shallower causing background blur
Solution Approach 1:
The patent segments the image processing into multiple depth-of-field layers. It identifies foreground and background regions at different depths, processes them separately with appropriate focus settings, and composites them together. This allows the foreground to be captured with a larger aperture for better dynamic range while the background is separately focused and combined, resolving the contradiction between aperture size and depth of field.
2Illumination intensity
If multi-frame HDR synthesis is used to improve dynamic range, then the dynamic range is improved, but image clarity is lost due to misalignment and blur across frames
Solution Approach 1:
The patent performs preliminary actions by capturing multiple pre-focused images at different depths before the final HDR synthesis. These pre-focused images serve as aligned reference frames that are already sharp at their respective depth planes. When HDR synthesis is performed, these pre-aligned focused images are combined, ensuring both dynamic range improvement and image clarity is maintained without the misalignment problems of traditional multi-frame HDR.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
In the technical solutions of a high dynamic range image synthesis method and an electronic device provided in embodiments of this application, a plurality of images with different depths of field in a current photographing scene are obtained based on an HDR photographing operation entered by a user, and each image has an exposure value. A plurality of images with a same exposure value are synthesized to generate a full depth-of-field image. Full depth-of-field images with a plurality of exposure values are synthesized by using an HDR algorithm to generate a high dynamic range image. Therefore, a high dynamic range image that is clear at each depth of field can be obtained, and a problem that a shallow depth of field leads to a blurred background and an insufficient dynamic range, and then results in overexposure or underexposure of a high dynamic range image can be resolved.