Light Field Image Synthesis via Phase Difference Elimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing techniques fail to effectively generate focused images at a desired depth using multiple light field data obtained through various exposures based on the scene shot or purpose of shooting.
Innovation Solution
An image processing apparatus that obtains multiple images of the same subject at different positions and timings, detects phase differences between them, and synthesizes the images to eliminate these differences based on set generation conditions, allowing for the generation of a favorable reconstructed image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple LF data are obtained through multiple exposures to enhance dynamic range, then image quality is improved, but a method for usefully processing multiple pieces of LF data based on the scene and purpose has not been proposed
Solution Approach 1:
The patent applies dynamics by making the image synthesis process adaptive and flexible. The synthesis unit dynamically selects and processes LF data based on detected phase differences, correlation values, and user-specified conditions (such as depth of field, exposure time, and shooting scenario). This allows the system to adaptively process multiple LF data obtained through multiple exposures, enhancing dynamic range while maintaining image quality across various shooting scenarios.
2Measurement precision
If images are synthesized to eliminate phase differences, then image clarity is improved, but the complexity of detecting and eliminating phase differences increases
Solution Approach 1:
The patent applies feedback by using the detected phase difference information to guide the synthesis process. The phase difference detection unit measures phase differences between LF data, and this measurement feedback is used by the synthesis unit to adjust the synthesis parameters and eliminate phase differences. This feedback mechanism enables automatic adaptation to different shooting scenarios while maintaining image clarity.
Solution Approach 2:
The patent replaces complex manual phase difference detection and elimination procedures with automated computational methods. Instead of manual alignment and synthesis, the system uses correlation-based phase difference detection and automated synthesis algorithms that select and combine LF data based on detected phase differences and user-specified conditions, significantly reducing the complexity of the process.
3Reliability
If multiple images are synthesized to eliminate phase differences, then reconstructed image quality is improved, but processing time and computational load increase
Solution Approach 1:
The patent applies partial action by selectively processing only the necessary LF data based on user-specified conditions and detected phase differences. The synthesis unit does not process all available LF data uniformly, but rather selects and processes specific subsets based on correlation values, depth of field requirements, and shooting scenarios. This selective processing reduces computational load and processing time while maintaining reconstructed image quality.
Data Source
AI summary
An image processing apparatus obtains a plurality of images obtained by shooting the same subject at different positions, at each of a plurality of timings. Then, a first phase difference between images obtained at the same time, and a second phase difference between images obtained by shooting at the same position but at different timings, are detected. The image processing apparatus then selects at least some of the obtained images as images to be used in synthesis in accordance with a set generation condition, synthesizes the images so as to eliminate at least one of the first and second phase differences, and outputs the resulting image.