HDR RAW File Structure for Image Type Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies, such as Japanese Patent Laid-Open No. 2005-323162, do not facilitate easy identification of the type of RAW images included in a single file, complicating the specification and management of high dynamic range (HDR) synthesis in imaging devices.
Innovation Solution
An information processing device and method that acquires and processes multiple RAW images with different brightness, generates developed images, and records them in a file with metadata specifying the type of images, allowing for HDR synthesis and easy identification of image types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple RAW images are recorded in one file, then storage efficiency is improved, but identification of image types becomes difficult
Solution Approach 1:
The patent segments the file structure into distinct components: RAW image data areas and separate information areas. Each information area contains metadata specifically associated with its corresponding RAW images, including tags that identify the type of RAW images (e.g., L-RAW, H-RAW). This segmentation allows multiple RAW images to be stored efficiently while maintaining clear identification of each image type through structured metadata.
Solution Approach 2:
The patent introduces information areas as intermediaries between the RAW image data and the user/system. These information areas contain structured metadata that mediates the identification and management of RAW image types. The metadata acts as an intermediary layer that provides easy specification of image types without requiring direct analysis of the RAW data itself.
2Measurement precision
If alignment processing is performed on two images, then synthesis accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by performing alignment processing specifically when synthesizing images from time-division exposure (two separate exposure times). For images captured simultaneously by the DGO sensor, alignment is not required. This selective preliminary alignment approach ensures synthesis accuracy is improved only where necessary, while avoiding unnecessary processing time for already-aligned images.
3Adaptability or versatility
If synthesized RAW image is recorded, then flexibility for post-processing is improved, but file size increases
Solution Approach 1:
The patent applies partial action by selectively recording only certain synthesized images in the file, rather than recording all possible synthesized variations. The system records RAW images at different exposure levels (L-RAW, H-RAW) and their corresponding synthesized images, but allows users to perform additional post-processing on the RAW data as needed. This approach provides sufficient flexibility for most post-processing needs while avoiding the excessive file size that would result from storing all possible synthesized versions.
Data Source
AI summary
An information processing device executes image processing of generating a plurality of developed images by performing development processing on the plurality of RAW images, and executes recording control processing of generating one file including the plurality of RAW images, information of the plurality of RAW images, and the plurality of developed images, and record the file in a storage medium, wherein in the image processing, it is capable of generating a synthesized image having gradation properties different from gradation properties of the plurality of developed images by performing synthesis processing of the plurality of RAW images or the plurality of developed images, and in the recording control processing, a synthesized image after the synthesis processing and after the development processing is recorded in the one file, but a synthesized RAW image generated by the synthesis processing of the plurality of RAW images is not recorded.