Gabor Edge Detection Method Based on Bilateral Filter and Otsu Threshold for Noisy Ultrasound Image

摘要

Nowadays, ultrasound image is important and very useful in medical field. It is a technique used for visualizing body structures including tendons, muscles, joints, vessels and internal organs. However, it will cause some problem if the method of edge detection is not suitable. It can affect the visual diagnosis of medical personnel. Edge detection is important for segmentation, pattern recognition and image compression in ultrasound image. The purpose of edge detection is to capture the major feature at some specified spatial scale. This project proposes an edge detection method that could detect edges accurately without including noise by using the combination of the Bilateral Filter, Otsu Threshold and Gabor Filter. The Bilateral Filter is used to suppress the noise level in the input image. Otsu Threshold used to segment the regions into background and objects. The Gabor Filter is applied to detect edge accurately without including noise. From our simulation, the proposed method is capable to produce sharp edgemap image with less noise and detect edges accurately in comparison to conventional edge detection method for various types of ultrasound images.