Sparse Color Image Sensor With Multi-Frame Color Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic image sensors with color filter arrays (CFAs) face inefficiencies in light capture, leading to reduced resolution, increased noise, and decreased image quality, particularly in low-light conditions, due to wavelength-specific filters blocking significant light and causing chromatic aberrations.
Innovation Solution
A sparse color filter array (CFA) with a majority of panchromatic pixels and a minority of wavelength-filtered pixels is used to capture multiple frames, which are then computationally aligned and merged to generate a single color image, undersampling chromatic data and compensating for spatial offsets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a Color Filter Array (CFA) with wavelength-specific filters is used to capture chromatic information, then color image creation is enabled, but optical efficiency is significantly reduced and light reception is blocked
Solution Approach 1:
The CFA is segmented into multiple subunits, each containing a mix of wavelength-filtered pixels and panchromatic pixels. This segmentation allows different regions to capture different types of information (chromatic and luminance) simultaneously, resolving the contradiction between capturing color information and maintaining light efficiency.
Solution Approach 2:
Different pixels within the CFA have different local qualities - some pixels have wavelength-specific filters for chromatic information while others are panchromatic for luminance information. This local differentiation allows the system to capture both color and light efficiency in different locations, resolving the overall contradiction.
2Loss of information
If wavelength-specific filters are disposed on individual pixels in a repeating pattern, then chromatic information is captured, but spatial resolution is reduced and noise increases
Solution Approach 1:
The CFA structure is divided into subunits with specific patterns of filtered and panchromatic pixels. This segmentation creates a more efficient sampling pattern that improves spatial resolution while maintaining chromatic information capture, compared to traditional repeating patterns.
Solution Approach 2:
The patent introduces temporal dimension by capturing multiple image frames over time. This allows the system to overcome spatial sampling limitations by combining information from multiple temporal instances, thereby improving effective spatial resolution while maintaining chromatic accuracy.
3Manufacturing precision
If multiple image frames are captured and processed computationally, then image resolution and quality are enhanced, but processing complexity increases
Solution Approach 1:
The CFA is pre-configured with specific subunit patterns that optimize the capture of both chromatic and luminance information. This preliminary structural arrangement reduces the complexity of subsequent processing by ensuring that complementary information is captured in an organized manner from the outset.
Solution Approach 2:
Multiple identical or similar subunit patterns are replicated across the CFA structure. This copying approach allows the system to capture redundant information in different locations, which simplifies processing by providing multiple sources for the same type of data, making the computational merging more straightforward.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances image resolution and detail while maintaining high light sensitivity, reducing noise, and improving image quality by computationally combining chromatically undersampled and shifted information from multiple frames.
Implementation Method 1
each subunit including a minority of one or more wavelength-filtered adjacent pixels and a majority of remaining panchromatic pixels
Implementation Method 2
the majority of pixels of each subunit are panchromatic pixels... Whereas the panchromatic pixels provide luminance information for a captured image frame
Implementation Method 3
electronic image sensors in mobile and other imaging systems typically include a Color Filter Array (CFA)... To capture chromatic information
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and techniques are described for generating a color image by computationally combining chromatically undersampled and shifted information present in multiple component image frames captured via a sparse color filter array that includes a minority of wavelength-filtered picture elements and a remaining majority of panchromatic picture elements. A burst capture is initiated of multiple image frames via a color filter array comprising a plurality of subunits, each subunit including a minority of one or more wavelength-filtered adjacent pixels and a majority of remaining panchromatic pixels. Each of the multiple image frames is processed to generate a resulting color image.