Method and system for image denoising

By combining high-frequency and low-frequency denoising methods, and utilizing non-sampling Haar wavelet decomposition and sampling wavelet decomposition to dynamically select the denoising level, the problem of noise removal and detail preservation in image fusion is solved, achieving a highly efficient image denoising effect.

CN114626996BActive Publication Date: 2026-06-09SAMSUNG ELECTRONICS CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SAMSUNG ELECTRONICS CO LTD
Filing Date
2021-12-08
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing technologies struggle to effectively remove noise in image fusion without losing details and edges, especially in multi-frame image processing, leading to a decline in image quality.

Method used

A hybrid denoising method combining high-frequency noise reduction (HFNR) and low-frequency noise reduction (LFNR) is adopted. The method dynamically selects the denoising level through adaptive spatial domain denoising of high-frequency and low-frequency bands, uses a hybrid fusion weight map to control the denoising result, and combines non-sampling Haar wavelet decomposition and sampling wavelet decomposition to achieve efficient denoising.

Benefits of technology

It significantly reduces noise without losing image details and edges, improving the quality and efficiency of multi-frame image processing and reducing system complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114626996B_ABST
    Figure CN114626996B_ABST
Patent Text Reader

Abstract

Methods and systems for image denoising are provided. The methods include receiving an input frame, performing high frequency noise reduction (HFNR) on the received input frame, performing low frequency noise reduction (LFNR) on the received input frame, and obtaining a hybrid denoised clean frame by fusing an output of the performed HFNR and an output of the performed LFNR.
Need to check novelty before this filing date? Find Prior Art