Under-Display Camera Narrowband Filtering for Diffraction Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Under-display camera systems face image restoration challenges due to diffraction issues caused by incoherent natural light, which leads to lost information and reduced image resolution, as conventional image sensors struggle to capture and process the specific point spread function of all wavelengths.
Innovation Solution
Incorporating narrowband filters in the image sensor to detect and transmit narrowband information to a model unit, which reconstructs edge information using a database of diffraction patterns, and combining it with color information to restore images effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If narrowband filters are added to detect narrowband information, then image resolution is improved, but device complexity increases
Solution Approach 1:
The pixel array is segmented into different functional regions: color filter photodiodes for color information and narrowband filter photodiodes for edge information. This segmentation allows each region to specialize in capturing specific types of optical information, improving overall image resolution while maintaining manageable system complexity through functional division.
Solution Approach 2:
The image sensor is designed with multi-functionality by integrating both color filter photodiodes and narrowband filter photodiodes within the same sensor structure. This allows the sensor to simultaneously capture both color information and edge/diffraction information, enabling comprehensive image restoration without requiring separate sensing systems.
2Measurement precision
If deep neural network is used to reconstruct edge information, then image quality is improved, but processing time increases
Solution Approach 1:
Diffraction patterns are pre-calculated and stored in a database for various wavelengths and display configurations. During image restoration, the deep neural network can directly query and apply these pre-computed patterns rather than calculating them in real-time, significantly reducing processing time while maintaining high image quality through accurate diffraction compensation.
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 improves image resolution by resolving diffraction problems and enhancing the performance of under-display camera systems by accurately reconstructing edge and color information, resulting in a higher quality image.
Implementation Method 1
The plurality of first narrowband filters are disposed on a plurality of second photodiodes, respectively, and configured to generate a first narrowband information in accordance with the raw data. A first spectrum linewidth of the plurality of the first narrowband filters is in a range from 5 nm to 70 nm.
Implementation Method 2
An image generated by the imager sensor has diffraction problems when the light passes through the display. The database unit of the point spread function is configured to store a plurality of diffraction patterns.
Implementation Method 3
The plurality of color filters are disposed on a plurality of first photodiodes, respectively, and configured to generate a color information in accordance with the raw data.
Data Source
AI summary
An operating method of an under-display camera system includes: providing a raw data by a pixel array; generating, by a plurality of color filters respectively disposed on a plurality of first photodiodes of the pixel array, a color information in accordance with the raw data; generating, by a plurality of first narrowband filters respectively disposed on a plurality of second photodiodes of the pixel array, a first narrowband information in accordance with the raw data, wherein a spectrum linewidth of the plurality of first narrowband filters is in a range from 5 nm to 70 nm; reconstructing an edge information from the first narrowband information based on one of a plurality of diffraction patterns provided by a database unit of a point spread function; and obtaining an image by combining the edge information with the color information.