Research And Implementation Image Processing System Of Corneal Topography

The digital image processing technique is applied to the system of corneal topography and useful to the analysis of corneal diseases. This thesis takes the placido image as the research object, mainly including image acquisition, image pretreatment, feature extraction technology and rendering technology. In addition, it also involves image denoising, edge extraction, etc.The content of this work is list as follows:1. The corneal physiological characteristics are analyzed. The general design method of corneal topography system and the realization method of each module are introduced.2. Image acquisition system is designed and realized. The control part is realized by the control cards and Step motor. Image acquisition part is realized by video capture card and CCD.3. Placido image feature extraction method based on gauss kernel is proposed. The median filtering is used for feature extraction, an improved Hough transform is used to detect the bright ring center roughly. Placido image is located in polar coordinate, the convolution characteristics of gray curve and gaussian kernel are used to detect the central line of bright ring, ellipse fitting is used to fix the center point, and the central line of bright ring is extracted finally.4. Calibration sphere algorithm is used to calculate the curvature radius of corneal surface, An Arc-step algorithm based on ellipse is used to calculate the corneal height. Rendering technology of corneal topography is researched, an improved reverse distance weighted method is proprosed to interpolate discrete points into successive points, color encoding method is used to connect corneal surface data with different color, and corneal surface data becomes visualized.