RYYB Image Sensor Guided Filtering Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image sensors, particularly CMOS image sensors, face challenges in reducing noise in generated images, which affects image quality, especially in applications like smartphones and medical microcameras.
Innovation Solution
The implementation of an image processing system that includes an image sensor using RYYB or RWWB color filter arrays, interpolation logic to generate guide images, and guided filtering logic to improve image quality by reducing noise, specifically by interpolating and filtering R, Y, W, and B image portions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If RYYB or RWWB color filter arrays are used in image sensors, then light sensitivity and signal-to-noise ratio are improved, but noise in the generated images increases
Solution Approach 1:
The patent segments the image processing into multiple stages: first separating the image into different color components (R, Y/W, B), then selectively applying interpolation to specific color portions (Y or W portions for R image generation, W portions for B image generation), and finally combining the processed components. This segmented approach allows noise reduction in specific color channels while preserving the high light sensitivity benefits of RYYB/RWWB arrays.
Solution Approach 2:
The patent introduces an intermediary guide image (generated from interpolated Y or W portions) that mediates the filtering process. This guide image is used in guided filtering to reduce noise in the R and B image portions while preserving edges and important features. The intermediary guide image acts as a reference that enables selective noise reduction without degrading the overall image quality.
2Loss of information
If conventional interpolation methods are used to generate missing color portions, then image completeness is improved, but image quality and noise performance deteriorate
Solution Approach 1:
The patent applies different processing qualities to different parts of the image. Specifically, it performs interpolation on Y portions to generate R image data, and on W portions to generate B image data, while using guided filtering with locally adapted guide images. This local quality approach ensures that each color channel is reconstructed with appropriate detail and noise characteristics, improving overall image quality compared to uniform interpolation methods.
Solution Approach 2:
The patent performs preliminary interpolation to generate guide images (Y image for R portion filtering, W image for B portion filtering) before applying the final guided filtering operation. This preliminary action creates reference images that contain edge and structure information, which are then used to guide the noise reduction process and preserve important image features during reconstruction.
Data Source
AI summary
An image processing system includes: an image sensor suitable for generating an RYYB (Red Yellow Yellow Blue) bayer image by applying an RYYB color filter array; an interpolation logic suitable for generating a Y image at a position of an R image portion in the RYYB bayer image by interpolating a YY image portion in the RYYB bayer image; and a guided filtering logic suitable for guided-filtering the R image portion in the RYYB bayer image by using, as a guide image, the Y image at the position of the R image portion.