Method and system for obtaining retina structure from optical coherent tomography image

The invention provides a method and system for obtaining a retina structure from an optical coherent tomography image. The method comprises that a region of interest (ROI) image of a source image is calculated; Gaussian filtering is carried out on the ROI image; a multi-resolution method is used to calculate a first estimated boundary position of a first layer boundary and a second estimated boundary position of a second layer boundary; a simplified active contour model is used to cut the first layer boundary and the second layer boundary initially according to the first estimated boundary position and the estimated boundary position, and a first initial contour position of the first layer boundary and a second initial contour position of the second layer boundary are obtained; and the first initial contour position and the second initial contour position are smoothed by utilizing a filtering method. According to a segmentation position of the first layer boundary and a segmentation position of the second layer boundary, a segmentation position of a residual layer boundary in the ROI image is obtained. The method and system can improve the calculation speed and reduce the calculation complexity substantially.