Research On Periodontal Disease Detection Of Mobile Device Oral Image Based On Deep Learning

Periodontal disease is a disease of the influence oral health people,has the characteristics of the high incidence and phase.It is a serious condition that can lead to complications such as cardiovascular disease and diabetes.Unfortunately,the shortage of professional stomatologists and the uneven distribution of medical resources between urban and rural areas in China make it challenging to diagnose periodontal disease.Therefore,it is of great theoretical and practical significance to study the method of automatic diagnosis of periodontal disease.With the popularity of mobile devices,based on a mobile device oral periodontal disease detection image research has become possible.These studies are not perfect.They only achieve automatic detection of periodontal disease,but do not carry out studies on the problems of small detection targets and similar detection targets in oral imaging periodontal disease detection.The article proposes a deep learning-based method for detecting periodontal disease in oral images on mobile devices.For mobile equipment oral image blur and the imbalance of the sample,the details of the problems,the article puts forward a kind of oral optimization strategy,image data fusion and data of image enhancement strategy to adjust oral image optimization.On this basis,combining with the existing similar between classes,periodontal disease detection target smaller problems,the article proposes a periodontal disease detection algorithm based on YOLO,through the design coordinates attention mechanism of double sampling cross the fusion feature network layer and the scale,the realization of mobile equipment oral periodontal disease image automatic detection.In order to verify the validity and rationality of our method,the article carried out an experiment on the detection method of periodontal disease.The experimental results show that m AP of the detection method of periodontal disease is 75.92%,and FPS is 21.All the evaluation indicators meet the need of periodontal disease detection.In addition,a YOLO-based wechat applet for the detection of periodontal disease in oral images on mobile devices is developed and designed,which provides the possibility for the transformation of demonstration method research into practical application.