Based On Fringe Projection 3D Imaging To Prescreen Patients With Marfan Syndrome

Marfan syndrome is an autosomal dominant inherited multi-system connective tissue disease.In terms of appearance,it is manifested as slender limbs,fingers and toes,asymmetrical limbs,finger distance greater than height,height significantly higher than ordinary people,lower body longer than upper body,accompanied by cardiovascular system abnormalities,the potential risk of cardiovascular diseases,leading to the possibility of sudden death.Currently,gene sequencing,color Doppler ultrasound diagnosis,echocardiography and other methods are mainly used for screening patients with Marfan syndrome,but these methods are inefficient and expensive.Therefore,developing a low-cost,rapid,pre-screening method is critical to diagnosing the disease.Fringe projection 3D imaging technology is a mature non-contact high precision technology,has been widely used in machine vision,industrial inspection,quality control,cultural relic protection and medical cosmetology and other fields.According to the characteristics of the external signs of Marfan syndrome,a feature detection system based on structured light measurement is designed in this thesis.The main research contents are as follows:Firstly,the detection technology of Marfan syndrome,face detection technology,structured light measurement technology and 3D anthropometry technology at home and abroad are reviewed.Combined with the appearance characteristics of the disease,the 3D human body measurement method of structured light is proposed.Secondly,this thesis constructs a set of infrared structured light system,and introduces the design scheme of the system in detail.In the aspect of three-dimensional measurement,gray code assisted phase shift technology is used to achieve the solution of absolute phase,and the system is calibrated to make its accuracy reach 0.1mm,which can meet the needs of anthropometry.At the same time,the three-dimensional reconstruction model and method are described in detail.The absolute phase can be converted into three dimensional information.In addition,this thesis also uses the DLIB face feature detection library to detect twodimensional face key features to obtain facial key feature information,and combines twodimensional semantic detection and recognition with three-dimensional reconstruction results to clarify the detection feature parameters,so as to analyze human feature data in advance and provide reference for medical diagnosis.To promote the application and popularization of structured light in the field of medical measurement.