论文部分内容阅读
面部表情运动估计是面瘫早期诊断和治疗评价等领域即将出现的一种重要的关键技术。本文提出了一种基于遗传算法的光流方法 (OpticalFlowMethodbasedonGeneticAlgorithm ,GAOF)进行早期面瘫运动估计。该方法从面部动态图像序列中估计面部微小运动矢量分布、特征参量及动态模式 ,用于早期面瘫的定位诊断与治疗定量评价。对一组正常被试者与一组面部神经肌肉运动异常病人进行实验 ,结果表明 :该方法与其他诊断方法相结合 ,对于早期面瘫定位诊断和恢复程度定量评价非常有效。与其他方法相比具有定位客观 ,定量检测精度高和便于进行动态模式评价等优点
Facial expression motion estimation is an important and crucial technology in the field of early diagnosis and treatment evaluation of facial paralysis. This paper presents a genetic algorithm based optical flow method (OpticalFlowMethodbasedonGeneticAlgorithm, GAOF) for early facial paralysis motion estimation. This method estimates facial micro motion vector distribution, characteristic parameters and dynamic mode from facial dynamic image sequence for the quantitative diagnosis of the diagnosis and treatment of facial paralysis. Experiments were performed on a group of normal subjects and a group of patients with facial neuromuscular abnormalities. The results show that this method combined with other diagnostic methods is very effective for the quantitative evaluation of the early diagnosis and recovery of facial paralysis. Compared with other methods, it has the advantages of objective positioning, high quantitative detection accuracy and easy dynamic model evaluation