Image segmentation method, device and equipment and storage medium

• Main Authors: WEI RUNJIE, GAO QI, LIU DONGLIN
• Format: Patent
• Language: Chinese; English
• Published: 09.02.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATIONTECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING ORPROCESSING OF MEDICAL OR HEALTHCARE DATA; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

The invention discloses an image segmentation method and device, equipment and a storage medium, and the method comprises the steps: improving the network structure of a full convolution network FCN,building the residual connection between all feature extraction units of a feature extraction network of the FCN, carrying out training of the structure-improved FCN based on sample data, and obtaining a trained FCN. Therefore, when the to-be-segmented image is processed through the trained FCN, the segmentation accuracy of the pixel level of the to-be-segmented image can be improved, and the to-be-segmented image and the sample data are medical image data. 本申请公开了一种图像分割方法、装置、设备以及存储介质，在本申请中，改进了全卷积网络FCN的网络结构，在FCN的特征提取网络的每一特征提取单元之间建立残差连接，并基于样本数据对结构改进的FCN进行训练，得到训练完成的FCN，如此，通过训练完成的FCN对待分割图像处理时，可以改善对待分割图像的像素级别的分割精准度，其中，待分割图像以及样本数据，均为医疗图像数据。