Research Of Expression Recognition Algorithm Facing The College Students' Psychological State

With the rapid development of computer technology and artificial intelligence,face recognition technology has become a hot issue in the society.From daily payment to intelligent monitoring,face recognition technology has penetrated into people's daily life and brought convenience to people.Government,society and business fields have been inseparable from face recognition technology.Face recognition technology is an interdisciplinary subject.Up to now,face recognition and facial expression recognition technology based on human facial biometrics has integrated many scientific knowledge,such as computer vision,artificial intelligence,human-computer interaction,in-depth learning and so on,involving many popular frontier scientific and technological knowledge.On the other hand,with the development of society,the number of college students in our country is increasing year by year,and the quality of college students varies from good to bad.With the reform of education,academic performance and scientific research level is not only one of the ability of college students,college students' psychological problems have also begun to be paid attention to by the society.In recent years,there are countless student accidents due to psychological problems,which also put forward a severe test for counselors and teachers who are responsible for the psychological problems of college students.To solve the students' psychological problems,the first thing is to find the problems in time.Face recognition technology and expression recognition technology in the university application field is only examinee identity authentication,access control management and so on,for concern about the mental health of students is still blank.This paper combines facial expression recognition technology with college students' psychological problems,through the study of appropriate facial expression recognition algorithm to solve college students' mental health problems,to facilitate college counselors and teachers to find the problem students in time to avoid accidents.The main contents of this paper are as follows:(1)Face detection system theory research.Face detection,as the first stage of face recognition and expression recognition,is a very important prophase work.This paper introduces the principle of some algorithms in face detection,include gray feature model,skin color model and artificial neural network model,Based on Haar feature,LBP feature and MB-LBP feature,combined with Adaboost classifier to achieve face detection,the three algorithms are compared,and finally come to the conclusion that MB-LBP feature method can more accurately detect the face,can adapt to changes in light and complex environmental changes,and has good robustness.(2)Image preprocessing.Image preprocessing is an important foundation of face recognition and expression recognition.This paper mainly introduces the image gradation,median filtering,binarization,histogram equalization and image noise reduction and removal in image preprocessing.In this paper,we use the above methods to work on the collected facial images,which effectively eliminates the influence of illumination,noise,size and other external factors on the recognition,and provides convenience for the recognition work.(3)Face recognition system theory research.This paper introduces several important methods of face recognition,including face feature-based method,template matching method,principal component analysis method,neural network method and support vector machine method,which provide theoretical support for subsequent facial expression recognition.(4)Research on expression recognition theory.In this paper,the expression recognition method based on feature block and LBP is used to test the performance of the algorithm.The performance of the algorithm based on JAFFE is compared with that of other algorithms,and the generalization ability and recognition ability of the algorithm are verified by experiments in natural scenes.(5)The study of psychological emotion based on facial expression.This paper studies the relationship between human expression and psychological emotion.Human facial expressions are divided into seven basic types:joy,surprise,sadness,contempt,disgust,fear and anger.This paper elaborates the types of facial expressions mapped by each kind of psychological emotion,and analyzes the changes of facial features of mouth,eyes,nose and other organs under various kinds of facial expressions in detail.(6)The relationship between expression recognition technology and psychology.Through heart-to-heart talks and mental health questionnaire to 30 college students,we can understand their mental state,and analyze whether they have mental health problems.At the same time to test the expression recognition based on the previous algorithm,the results of conversation,questionnaire and expression recognition are compared,the accuracy is extremely high,so we can use expression recognition method to control college students'psychology.