M.Sc. Tezi Görüntüleme
Periodic and quasi-periodic noises are effects which are come out because of electromagnetic interference and cause to arise a periodic pattern on the image. This can result from the power frequency used in the instrumentation and recording systems during image acquisition. For example, vibration, halftone and moire patterns are periodic noises. The resolution of the scanner used to scan the image affects the high frequency noise pattern in the acquired image and can produce moire patterns. It is also characteristic of gray scale images obtained from single-chip video cameras. It is also arise on biomedical images.
It is known that periodic and quasi-periodic noise results peaks in image spectrum amplitude. Periodic noise coordinates in the frequency domain can determined as benefited from this property.
The theme of Removing Periodic Noise Effects from Video Images, taken up within this study and a filtering technique which is applied to the frequency amplitude spectrum of the image, has developed. Periodic and quasi-periodic noise coordinates in the frequency domain is determined automatically with median filter masks for all frames. The operation of removing noise is performed a Butterworth multi-notch filter which is adapted according to centre of notches and numbers of perceived noise.
Key Words:Removing Periodic and Quasi-Periodic Noise, Median Filter, Multi-Notch Filter