摘要
对于通过经验和肉眼观察的传统诊断方法,可能引发误诊和漏诊的情况。为了精确的诊断口腔疾病,设计了一种基于卷积神经网络的口腔牙齿疾病诊断系统。该系统利用基于STM32的图像显示平台采集口腔图像,并使用GoogLeNet网络模型对图像进行特征提取与识别。采用TensorFlow框架对模型进行训练,结果表明,相比于传统的全连接网络,所提出的系统具有较高识别率。
Traditional methods of diagnosis through experience and visual observation may lead to misdiagnosis and missed diagnosis.A dental diagnosis system based on convolutional neural network is proposed for improving the recognition accuracy of dental diseases. The system uses the platform based on STM32 to get the oral image, and uses the GoogLeNet to obtains the image feature. The network model is trained by using the Tensor Flow deep-learning framework. simulation results show the proposed system has relatively higher recognition rate compared with the fully connected network.
作者
王丹峰
陈超波
马天力
李长红
Wang Danfeng;Chen Chaobo;Ma Tianli;Li Changhong(School of Electronic Information Engineering,Xi'an Technological University,Xi'an 710021,China)
出处
《国外电子测量技术》
2019年第6期93-97,共5页
Foreign Electronic Measurement Technology
基金
陕西省国际科技合作计划项目(2017KW-009)资助
