Enhancing the Quality of Fog/Mist Images by Comparing the Effectiveness of Kalman Filter and Adaptive Filter for Noise Reduction
Main Article Content
Article Sidebar
Abstract
The primary objective of this study is to enhance the precision of fog and mist noise reduction in photographs
by introducing a novel Kalman filter and comparing its performance to that of an Adaptive filter. Materials
and Methods: For this investigation, the research dataset was sourced from the Kaggle database system. Using
twenty iteration samples (ten for Group 1 and ten for Group 2), involving a total of 1240 samples, the efficacy
of fog and mist noise elimination with improved accuracy was assessed. This evaluation was conducted
employing a G-power of 0.8, a 95% confidence interval, and alpha and beta values of 0.05 and 0.2,
respectively. The determination of the sample size was based on the outcomes of these calculations. The novel
Kalman filter and the Adaptive filter, both utilizing the same number of data samples (N=10), were employed
for fog and mist noise removal from images. Notably, the Kalman filter exhibited a higher accuracy rate.
Results: The novel Kalman filter showcased a success rate of 96.34%, outperforming the Adaptive filter's
success rate of 93.78%. This difference in performance is statistically significant. The study's significance
threshold was set at p=.001 (p<0.05), confirming the significance of the hypothesis. This analysis was carried
out through an independent sample T-test. Conclusion: In conclusion, the proposed Kalman filter model,
achieving an accuracy rate of 96.34%, demonstrates superior performance compared to the Adaptive filter,
which yielded an accuracy rate of 93.78%. This comparison underscores the efficacy of the Kalman filter in
the context of image noise removal.
How to Cite
Article Details
Algorithm, Covariance Matrix, , Novel Kalman