Accurate interpretation of endoscopic image is the firm basis of discovering thepatients' pathological changes, from which sound diagnosis is able to be made for taking further clinical measures , including minimally invasive surgery etc . Therefore, It's vital that effective processing acquired endoscopic image in a well-structured manner.Wavelet transform is an information processing tool which is one of the most widelyused, and plays an important role in image processing . The study is focusing on the application of wavelet transform in the endoscopic image de-noising .The thesis includes the following three parts :1. In-depth analysis and study of the image of the endoscope and developed the general processing framework in terms of remove image highlight , optical distortion geometric correction , motion blur, image de-noising and image enhancement.2. Based on specific experiments , the traditional variety of image de-noising methods were analyzed and compared .3. The key part of study lies on the application of wavelet transform in the endoscopic image de-noising . A new de-noising method is proposed : firstly, in the airspace using the median filtering method for the image ; secondly , in the frequency using wavelet transform to decompose the image ; thirdly , for the wavelet coefficients using the median filtering method ; finally , using wavelet inverse transform for image restoration . It's testified this method is better in comparison with other de-noising algorithms , in the Gaussian white noise, impulse noise handling . |