Subpixel Image Resolution Using NNPD and PSF Deconvolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image capture technologies fail to recognize finer details smaller than the sensor pixel size, resulting in a loss of detail in captured images.
Innovation Solution
The method involves obtaining point spread function (PSF) data, determining subpixel PSF data, generating a filled subpixel sparse image, and applying nearest neighbor pixel deconvolution (NNPD) to enhance image resolution to a subpixel level.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image capture with sensor pixels is used, then the imaging system can capture images, but finer details smaller than the sensor pixel size cannot be recognized, resulting in loss of detail
Solution Approach 1:
The patent divides each sensor pixel into multiple subpixels (e.g., 2x2=4 subpixels per pixel) to increase the effective resolution. This segmentation allows the system to resolve finer details by treating each subpixel as an independent sampling point, thereby overcoming the limitation of the original pixel size without requiring additional physical sensors.
Solution Approach 2:
The patent introduces a subpixel dimension within each pixel by applying a smoothing function that distributes pixel values to multiple subpixel locations. This dimensional transformation from pixel-level to subpixel-level sampling enables the system to capture and reconstruct finer spatial details that would otherwise be lost due to the discrete pixel structure.
2Measurement precision
If image resolution is enhanced to subpixel level, then finer details can be recognized, but the processing complexity increases due to PSF data processing and deconvolution operations
Solution Approach 1:
The patent performs preliminary smoothing of the point spread function (PSF) data before deconvolution operations. By pre-processing the PSF to create a smoothed version, the system reduces the complexity of subsequent deconvolution calculations while maintaining the ability to recover subpixel-level details. This preliminary action simplifies the overall processing pipeline.
Solution Approach 2:
The patent introduces an intermediate filled subpixel sparse image as a mediator between the original pixel data and the final enhanced image. This intermediate representation allows the system to work with a simplified data structure that separates known pixel values from unknown subpixel values, making the deconvolution process more manageable and computationally efficient.
Data Source
AI summary
A method for providing enhanced subpixel resolution includes obtaining point spread function (PSF) data associated with an input image. The method also includes determining subpixel PSF data from the PSF data. The method further includes generating a filled subpixel sparse image from pixels of the input image. In addition, the method includes applying nearest neighbor pixel deconvolution (NNPD) to the subpixel PSF data and the filled subpixel sparse image to generate an enhanced subpixel image having an increased resolution.