A Fast Blind Noise Estimation Algorithm for High ISO Images Based on Local Variance

For the problem that traditional noise estimation algorithms do not take the noise characteristic under actual conditions into consideration, a fast blind noise estimation algorithm for high ISO images based on local variance is proposed. Firstly, random sampling positions are generated, and ineffec...

Full description

Saved in:
Bibliographic Details
Published in2022 IEEE 12th International Conference on Electronics Information and Emergency Communication (ICEIEC) pp. 22 - 26
Main Authors Liu, Yinghan, Song, Zhengyu, Wang, Xing
Format Conference Proceeding
LanguageEnglish
Published IEEE 15.07.2022
Subjects
Analytical models
blind estimation
Conferences
Estimation
Fitting
Image edge detection
ISO
local variance
noise estimation
noise level
Simulation
Online AccessGet full text

Cover

More Information
Summary:For the problem that traditional noise estimation algorithms do not take the noise characteristic under actual conditions into consideration, a fast blind noise estimation algorithm for high ISO images based on local variance is proposed. Firstly, random sampling positions are generated, and ineffective sampling positions are removed by using the improved edge detection algorithm based on canny operator. Then, the gray level and the average local variance of smooth image regions is calculated according to the gray level and the local variance of image blocks. Finally, the fitting function of the average local variance and the gray level is constructed to obtain the estimate of noise level. The simulation results show that the proposed algorithm is faster and more accurate than the traditional algorithms.
ISBN:9781665407533
1665407530
ISSN:2377-844X
DOI:10.1109/ICEIEC54567.2022.9835100