Design hybrid filter technique for mixed noise reduction from synthetic aperture radar imagery
| Autor: | Ekbal Hussain Ali, Ahmed Hameed Reja, Layla H. Abood |
|---|---|
| Rok vydání: | 2022 |
| Předmět: |
Control and Optimization
Computer Networks and Communications Speckle noise Nonlinear filter Fuzzy classical filter Hardware and Architecture Control and Systems Engineering Additive noise Computer Science (miscellaneous) Impulse noise (salt and pepper) Mixed noise Electrical and Electronic Engineering Instrumentation Information Systems |
| Zdroj: | Bulletin of Electrical Engineering and Informatics. 11:1325-1331 |
| ISSN: | 2302-9285 2089-3191 |
| DOI: | 10.11591/eei.v11i3.3708 |
| Popis: | For military and civilian applications, synthetic aperture radar (SAR) imaging is an essential instrument for obtaining images of the Earth's surface. Speckle noise, a form of noise that is multiplicative, generated by conflicting echoes returned from each pixel, has a significant impact on the SAR picture. On SAR pictures, a hybrid filter for mixed noise reduction is used to remove the mixed noises that are present in the data during capture and transmission. Specifically, speckle noise and salt and pepper noises from SAR images. Both are being worked on at the same time to minimize mixed noise in SAR pictures without revealing edges or other features. This study proposes a technique that combines a hybrid filter derived from a statistics filters with nonlinear functions (SFNF). When comparing to mean, median as no adaptive filters, and frost filter, a lee filter, and fuzzy filters as adaptive filters, this hybrid filter produces good results. MATLAB was used to carry out the simulation. To illustrate the filtering technique's performance, quantitative measurements like signal to noise ratio (SNR) procedure, the mean square error (MSE) method, and edge measurement (β) mechanization are used. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|