HDR Image Synthesis via Luminance Boundary Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing high dynamic range (HDR) image synthesis methods face challenges in capturing images with different exposure amounts using the same or different image pickup elements, leading to difficulties in managing photographic conditions, particularly with moving objects and varying image pickup factors.
Innovation Solution
An image processing system that receives multiple images with different luminance components, selects a standard image, and performs multivalued processing to identify and complement too bright and too dark regions using matching regions from other images, aligning boundary luminances to generate a synthetic HDR image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multiple images with different exposure amounts are captured using the same image pickup element, then HDR image synthesis quality is improved, but the time interval between captures increases making it difficult to satisfy consistent photographic conditions
Solution Approach 1:
The patent segments the image capture process by using multiple image pickup elements (cameras) that simultaneously capture images at different exposure amounts. This divides the temporal sequence into parallel simultaneous captures, eliminating the time interval problem while maintaining the benefit of different exposure levels for HDR synthesis.
Solution Approach 2:
The patent transitions from a temporal dimension (sequential captures at different times) to a spatial dimension (simultaneous captures by multiple elements). By adding the dimension of multiple image pickup elements, the system achieves different exposure amounts without time delays, resolving the contradiction between capture quality and timing consistency.
2Loss of time
If multiple images are captured using different image pickup elements, then images can be obtained at roughly the same time, but strict management of photographic conditions such as angle of field and focal length becomes necessary
Solution Approach 1:
The patent applies universal calibration data that can be used across different image pickup elements and photographic conditions. This multi-functional calibration approach allows the system to handle various cameras and optical configurations without requiring separate coordination procedures for each element, reducing device complexity while maintaining synchronization.
Solution Approach 2:
The patent changes the approach from managing physical parameters (angle of field, focal length) of each image pickup element to using computational parameters (luminance values, exposure amounts) that can be processed and aligned after capture. This parameter transformation simplifies the coordination requirement while achieving consistent HDR synthesis results.
3Manufacturing precision
If images are captured sequentially with increasing time intervals to obtain more exposure variations, then HDR dynamic range is improved, but satisfying consistent photographic conditions becomes increasingly difficult
Solution Approach 1:
The patent segments the dynamic range capture into multiple simultaneous images taken by different image pickup elements, each capturing a different exposure level. This segmentation into parallel captures eliminates the time-based degradation of condition consistency while maintaining the full dynamic range coverage needed for high-quality HDR synthesis.
Data Source
AI summary
Provided is an information processing system configured to: receive a plurality of images; select a standard image; search, in extracting partial images for complementing a too bright region/too dark region in the standard image from a reference images, using correction images obtained by subjecting each of the images to multivalued processing for each division of luminance components, for regions having matching shapes of respective regions included in the correction image of the standard image and the correction images of the other images; and complement image portions corresponding to the too bright region and/or too dark region in the correction image of the standard image using image portions, which are matching regions and correspond to appropriately bright regions, to thereby generate a synthetic image. Each image is dividable into at least three values: a too bright region, an appropriately bright region, and a too dark region, and has a relationship in which a boundary luminance between the too bright region and the appropriately bright region aligns with a boundary luminance between the appropriately bright region and the too dark region in a darker image having an adjacency relationship therewith.