【摘要】 为了对口腔颌面部图像进行进一步定性定量的研究,从而预测生成治疗后的人脸图像,提出一套口腔颌面部图像预测生成系统设计方案。对各类因素导致口腔颌面部异常人脸图像使用计算机视觉技术与深度学习生成对抗网络技术进行治疗后图像预测生成。利用治疗前后不同组别患者图像,通过人脸图像预处理模块、人脸特征分析模块和人脸预测生成模块协同处理分析,最终预测生成口腔颌面部治疗后效果图。设计出一套口腔颌面部图像预测生成系统,本系统设计方案在一些医疗专科领域,如口腔颌面外科与医疗美容等领域都具有重要的临床参考价值和科研意义。更多还原

【Abstract】 In order to carry out further qualitative and quantitative research on oral and maxillofacial images, thereby predicting and generating face images after treatment, a set of architecture design schemes for the prediction and generation of oral and maxillofacial images is proposed. Computer vision and generative adversarial networks technology are used to predict and generate images after treatment for various factors that cause abnormal face images in the oral and maxillofacial region. Using the images of different groups of patients before and after treatment, the face image preprocessing module, the face features analysis module and the face prediction generation module are used for collaborative processing and analysis to finally predict and generate the face image after the oral and maxillofacial treatment. A set of oral and maxillofacial image prediction generation system is designed. The system architecture design has important clinical reference value and scientific research significance in some medical specialties, such as oral and maxillofacial surgery and medical cosmetology.更多还原