Low-Light Video Image Processing Through RGB-Domain Downsampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods for videos shot in dark environments face challenges in reducing noise while maintaining image quality, as they often reduce image size, affecting definition and potentially destroying the Bayer structure.
Innovation Solution
Convert raw-domain image streams to RGB-domain or YUV-domain, perform downsampling and resampling on the RGB-domain image to reduce size, and then process the reduced image without affecting definition or Bayer structure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If image size is reduced for raw-domain denoising, then operation performance and power consumption are improved, but image definition is reduced
Solution Approach 1:
The patent performs demosaicing algorithm on the raw-domain image stream before downsampling to convert it to RGB-domain or YUV-domain. This preliminary action ensures that the Bayer structure is maintained during size reduction, allowing subsequent downsampling to reduce image size without sacrificing definition or destroying the Bayer structure.
2Power
If image size is reduced for raw-domain denoising, then operation performance and power consumption are improved, but Bayer structure is destroyed
Solution Approach 1:
The patent applies demosaicing algorithm before downsampling to convert the raw-domain image to RGB-domain or YUV-domain. This preliminary conversion preserves the Bayer structure information, allowing subsequent downsampling to reduce image size while maintaining the structural integrity and color information of the original image.
Solution Approach 2:
The patent uses RGB-domain or YUV-domain as an intermediary representation between the raw-domain Bayer structure and the final processed image. By converting through this intermediary color space before downsampling, the patent maintains the Bayer structure's color information while enabling size reduction.
3Quantity of substance
If downsampling is performed on raw-domain image, then image size is reduced, but pseudo color is introduced
Solution Approach 1:
The patent performs demosaicing algorithm before downsampling to convert the raw-domain image to RGB-domain or YUV-domain. This preliminary action ensures that color information is properly interpolated and maintained, preventing pseudo-color artifacts from being introduced during the subsequent downsampling process.
Solution Approach 2:
The patent uses RGB-domain or YUV-domain as an intermediary that properly represents color information before downsampling. This intermediary representation allows for accurate color preservation during size reduction, avoiding the pseudo-color artifacts that would occur if downsampling were applied directly to the raw-domain Bayer data.
Data Source
AI summary
An image processing method and an electronic device are provided. The image processing method includes: displaying a first interface, where the first interface includes a first control; detecting a first operation on the first control; obtaining a first image stream in response to the first operation, where the first image stream is an image stream of first color space; converting the first image stream into a second image stream of second color space according to a demosaicing algorithm; performing downsampling and resampling on the second image stream to obtain a third image stream, where the third image stream is an image stream of the first color space, and a size of the third image stream is less than that of the first image stream; and performing image processing on the third image stream to obtain a fourth image stream.