Multi-Resolution Image Denoising for Global and Local Detail
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Noise corruption in digital images affects image displaying effects and analysis, necessitating improved image denoising methods.
Innovation Solution
An image denoising method involving obtaining multiple images with decreasing resolutions from an original image, extracting features hierarchically, and using convolutional neural networks to fuse and process these features for accurate denoising.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional single-resolution denoising methods are used, then the processing is simple and fast, but the denoising effectiveness is insufficient and global information is lost
Solution Approach 1:
The patent divides the original image into multiple versions with different resolutions (first resolution, second resolution, third resolution). Each resolution version is processed separately through the denoising network to extract features at different scales. This segmentation allows the system to capture both global structures (from lower resolutions) and local details (from higher resolutions), thereby improving denoising effectiveness while managing complexity through systematic multi-scale processing.
2Loss of information
If multiple resolution images are processed separately, then global information is retained, but the processing time and computational load increase
Solution Approach 1:
The patent merges the features extracted from multiple resolution images through feature fusion modules. The denoised features from first, second, and third resolution images are combined to produce the final denoised output. This merging strategy allows the system to retain global information from all resolution levels while consolidating the processing results, thereby reducing the overall computational burden compared to maintaining separate processing pipelines for each resolution.
3Measurement precision
If high-resolution processing is performed directly, then local details are preserved, but global context and noise patterns are missed
Solution Approach 1:
The patent introduces a resolution dimension by processing the same image at multiple different resolutions. This dimensional transformation allows the system to perceive global context from lower resolutions and local details from higher resolutions. By adding this resolution dimension to the processing, the network can simultaneously access both global and local information that would be inaccessible when processing only at a single high resolution, thereby improving overall denoising performance.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This application provides an image denoising method and apparatus, and relates to the artificial intelligence field and specifically relates to the computer vision field. The method includes: performing resolution reduction processing on a to-be-processed image to obtain a plurality of images whose resolutions are lower than that of the to-be-processed image; extracting an image feature of a higher-resolution image based on an image feature of a lower-resolution image to obtain an image feature of the to-be-processed image; and performing denoising processing on the to-be-processed image based on the image feature of the to-be-processed image to obtain a denoised image. This application can improve an image denoising effect.