An eyeball region segmentation method in an ultrasonic image

The invention discloses a method for segmenting an eyeball region in an ultrasonic image. The method comprises the following steps of: constructing a training set by utilizing a certain number of ophthalmic ultrasonic images which are subjected to eyeball region marking by experts; respectively calculating a directed symbol distance field matrix of each ophthalmic ultrasonic image in the trainingset, and forming a directed symbol distance field matrix of the training set by all the images in the training set; Training a pre-built convolutional neural network SS-by using a training set and a directed symbol distance field matrix Unet; inputting the to-be-segmented ophthalmic ultrasound image into the trained convolutional neural network SS- And Unet, thereby finishing eyeball region segmentation. According to the method, the eyeball region segmentation efficiency can be improved, and the segmentation accuracy is ensured. 本发明公开了一种超声图像中眼球区域分割方法，包括：利用一定数量的、且经专家进行眼球区域标注的眼科超声图像构建训练集；分别计算训练集中的每一眼科超声图像的有向符号距离场矩阵，由训练集