Expert Reasoning System Based On Microexpression Recognition

Face facial expressions contain complex inner emotions and are an important way for people to communicate.In recent years,they have been the hot topics in the field of computer vision,human-computer interaction and pattern recognition.With the improvement of computing performance,the algorithm has been improved,and human-computer interaction has been partially realized from the traditional peripherals such as mouse,keyboard and display touch screen to the multimedia-based conversion of images,audio and video.Emoticon recognition has great recognition difficulty and research significance of micro-facial expressions.It is the realistic feedback of human psychological intentions and physiological needs.As China's demographic structure is about to enter the aging stage,it will be an enormous challenge to the society,families and individuals.It will be a heavy and time-consuming labor care and guardianship for the elderly with disabilities.At present,China is already the country with the largest population aging in the world,reaching over 150 million.However,the aging-related industries in our country are still very backward and it is difficult to meet the demand for related services in the aging society.It is necessary and urgent to build a medical expert reasoning system with high degree of automation and intelligent reasoning and remote monitoring.Therefore,it is of great significance to build a system that can solve the problem of social aging and provide the elderly with a more humane and intelligent system on the platform of intelligent rehabilitation care beds.Through the camera real-time monitoring and identification of the user's micro-facial expressions,The system can take the micro-facial expressions as an important basis for their psychological activity and physiological needs.Through the expert reasoning system,which stores a large number of expert experiences and knowledge,Micro-facial expression expert reasoning system is performed to make decisions that meet the real needs of users.Firstly,capture images from high-speed video stream and save them.After preprocessing the image,Ada Boost-Haar algorithm is used to identify and calibrate the human face in the video,and the captured face image is used as the source image for micro-facial expression recognition.PCA Principal Component Analysis(PCA)method is used to construct the feature space of the micro-facial expression.Micro-facial expression recognition is performed on the source image.At the same time,the parameters of the main facial expression of the micro-facial expression are obtained,that is,the opening of eyes and mouth.Depth learning technology is used to do the same micro-facial expression recognition and compared to The PCA algorithm.Secondly,we establish an expert reasoning system that takes micro-facial expressions as input,and the system acquires the knowledge and experience provided by experts in specialized fields and simulates the decision-making process of experts in this field.We use neural network expert reasoning system mechanisms.Expert system reasoning machines,databases and explanations Etc.jointly formed into an expert reasoning system.Thirdly,build the hardware system and write the software system.The hardware system mainly includes Hikvision high-speed video surveillance camera,LAN,computer,lower computer control system and robot bed.The software system mainly includes face capture module,micro facial expression recognition module and Expert reasoning system module,and the docking interface between the modules.After the system is set up,the system design is debugged,verified and optimized through experiments.