Research on intelligent elderly care service information system based on machine learning

The aging of the population is becoming more and more serious, and intelligent pension is becoming more and more important. Today, machine learning and artificial intelligence are widely used in all walks of life, and facial recognition based on machine learning is applied to the pension service system, which will make the pension system intelligent. The elderly identity verification through face recognition has the characteristics of direct and convenient, not easy to lose. In recent years, with the research and application of machine learning in the field of artificial intelligence such as computer vision, face recognition as an important part of computer vision technology has also been greatly developed, all kinds of face recognition algorithms based on machine learning emerge endlessly, and the effect is far higher than the traditional method. This paper designs and implements an artificial intelligence elderly care service information system based on machine learning, and studies the face recognition technology of intelligent elderly care system. The paper carried out feasibility analysis and demand analysis of the system research, discussed the functions to be designed and implemented in the research, analyzed the work flow of the research, and according to the work flow, the research was divided into five functional modules: image acquisition, face detection, feature extraction, identity recognition and result output. The machine learning model YOLO is used for face detection, ERT algorithm is used for face alignment, and 128-dimensional feature vector is extracted by residual neural network to realize face recognition. The input and output of real-time images are realized by OpenCV. The function of the algorithm is tested on the flw data set, and the accuracy of the algorithm can reach more than 98%. It realizes the research of face recognition system in artificial intelligence elderly care service information system based on machine learning, which can achieve high recognition accuracy and fast operating efficiency, and can meet the requirements of practical applications.