Noise removal from UAV images using image processing techniques

Mohd Fauzey, Khadijah Amira (2016) Noise removal from UAV images using image processing techniques. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full text) Noise removal from UAV images using image processing techniques.pdf - Submitted Version Download (4MB)

Abstract

Digital images or images taken from camera are often corrupted with noise. The image needs to undergo processing before it can be used to certain purpose. Image denoising aims to produce a high quality of image by completely remove noise while preserving the edges. The main purpose of this research is to conduct a study on five types of noise such as Salt and Pepper noise, Gaussian noise, Localvar noise, Poisson noise and Speckle noise. Different noise densities of Salt and pepper noise have been removed between 0.1 to 1.0 by using four types of filters such as Average filter, Median filter, Wiener filter and Gaussian filter. To analyze the best method for removal of noise, Peak-Signal to Noise Ratio (PSNR) and Mean Square Errors (MSE) is used. Good Image can be identified when the PSNR value is high and the MSE value is lower. In this work, ten different of UAV images is used to observe the different values of PSNR and MSE obtained when the image is added by five types of noise and after applied filtering techniques. Standard deviation and average (mean) is calculated based on the values of PSNR and MSE of image corrupted with noise and image after removing noise.

Actions (login required)

View Item