Research On Dynamic Human Ear Recognition Method Based On Deep Learning

With the development of the field of machine learning,biometric recognition technology has also developed rapidly.In recent years,the human ear,as a biological feature,has gradually attracted the attention of related researchers due to its stable,non-invasive capture process,expressionless images,and significant differences in characteristics between individuals and value.However,most of the current research work on human ear recognition is carried out on static human ear images,and there is no public dynamic human ear database.Therefore,the establishment of a dynamic human ear database and the realization of dynamic human ear recognition are one of the human ear subjects.An important research direction is of great significance to the development of biometric recognition technology.This paper studies the dynamic ear recognition method based on deep learning.The main research contents include:(1)Aiming at the currently undisclosed dynamic human ear database,in order to better carry out the research on the dynamic human ear recognition method,the CCU-DE small sample human ear database is constructed.The database fully considers the various complex situations and posture changes of human ear images,such as translation angle,rotation angle,illumination change,occlusion and interference,etc.,making the research of dynamic human ear recognition closer to the complex actual situation.The applicability of human ear dynamic recognition.(2)Constructed the YOLO?v3 dynamic ear recognition model based on deep learning.The model adopts the migration learning method for pre-training.In the CCU-DE small sample ear database,it is used for multi-posture changes under different contrast,translation and rotation motions.A simulation study was conducted with or without occlusion.The simulation results show that the model can perform dynamic ear recognition well under complex conditions such as multi-posture changes,interference,and occlusion,thus verifying the feasibility and feasibility of dynamic ear recognition.Effectiveness and robustness.It has important reference and reference value for the research in the field of human ear recognition and other biometric recognition fields.