Photometric Normalization in Array Cameras
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital cameras face challenges in capturing the full dynamic range of a scene due to differences in imaging components, leading to photometric imbalances and noise in fused images, particularly caused by veiling glare and manufacturing tolerances, which affect the accuracy of parallax detection and image quality.
Innovation Solution
The method involves designating a reference imaging component and alternate components in an array camera, applying a photometric normalization process that includes determining a nominal parallax shift, applying a low-pass filter, and computing gain and offset parameters to align pixel information, thereby reducing photometric imbalances and normalizing image data across components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple imaging components are used in an array camera, then the dynamic range and image quality are improved, but photometric imbalances and noise are introduced due to manufacturing tolerances and veiling glare
Solution Approach 1:
The patent applies preliminary photometric normalization to image data from multiple imaging components before fusion. This involves computing normalization factors based on reference imaging data and applying these factors to alternate imaging components' data, thereby pre-correcting photometric imbalances and veiling glare effects before the images are combined, ensuring consistent numerical representation across all components
Solution Approach 2:
The patent changes photometric parameters (gain and offset) of individual imaging components based on computed normalization factors. By adjusting these parameters, the patent compensates for manufacturing tolerances and veiling glare, ensuring that all imaging components produce numerically consistent representations of the same scene points
2Manufacturing precision
If correction factors are applied to compensate for manufacturing differences, then imaging consistency is improved, but scene-related errors such as veiling glare cannot be compensated
Solution Approach 1:
The patent transitions from static correction factors to dynamic photometric normalization. The normalization factors are computed adaptively based on the actual scene content captured in reference imaging data, allowing the system to compensate for scene-related errors like veiling glare that vary with different imaging conditions, rather than using fixed correction factors
Solution Approach 2:
The patent uses feedback from reference imaging components to compute normalization factors that are then applied to alternate imaging components. This feedback mechanism allows the system to continuously adapt to scene conditions and compensate for both manufacturing variations and scene-related errors by comparing actual captured data against reference data
3Device complexity
If individual imaging components have different sensitivities and characteristics, then device complexity is reduced, but parallax detection accuracy and noise levels are adversely affected
Solution Approach 1:
The patent applies preliminary photometric normalization to ensure that alternate imaging components produce numerically consistent representations with the reference component before parallax detection is performed. This pre-processing step corrects sensitivity differences and veiling glare effects, thereby improving parallax detection accuracy without requiring all imaging components to be physically identical
Data Source
AI summary
Systems and methods for performing photometric normalization in an array camera in accordance with embodiments of this invention are disclosed. The image data of scene from a reference imaging component and alternate imaging components is received. The image data from each of the alternate imaging components is then translated to so that pixel information in the image data of each alternate imaging component corresponds to pixel information in the image data of the reference component. The shifted image data of each alternate imaging component is compared to the image data of the reference imaging component to determine gain and offset parameters for each alternate imaging component. The gain and offset parameters of each alternate imaging component is then applied to the image data of the associate imaging to generate corrected image data for each of the alternate imaging components.