In order to rapidly promote the development of medical industry, a robot system integrating intelligent monitoring and disinfection system emerges as the times require. The robot uses non-contact infrared induction temperature measurement to warn those whose body temperature exceeds the standard and upload face information to the cloud platform. It builds a EASY-IOT interface to monitor the working state of the robot in real time and obtains cloud background data from the background. The multifunctional disinfection robot effectively relieves the pressure of medical resources, reduce the risk of infection for health care workers and the public. This paper presents a monitoring system of EASY-IOT combined with intelligent image recognition. The designed robot increases the human-computer interaction function, which helps the administrator to check the working status of the robot at any time. The system mainly realizes: real-time monitoring of the local environment, autonomous image recognition, autonomous temperature measurement, and autonomous uploading of visitor information. The intelligent integration, multi-scene application of visual disinfection robot successfully solved the daily family, public places and other health protection problems. First, the original information from the sensor and the AI vision module is extracted from the target-related feature information, and then the feature parameters provided by the sensor are fused to obtain the classification and interpretation of the target. The monitoring system transmits data to the main system through RF, and the main system obtains whether to wear a mask and the temperature value, and at the same time prompts the corresponding information of the other party through the SYN6288 voice module. And through the EASY-IOT of environmental factors uploaded to the mobile phone small program or website, so as to achieve the purpose of monitoring and inspection. The overall system is independent and autonomous. 1. By analyzing the functional requirements of service robots in China, the monitoring system is studied and designed. The body temperature sensing system and active image recognition system are designed independently. Through the combination of the Internet of things, the local area network wireless communication system based on RF and the wide area narrowband communication network based on EASY- IOT are designed independently. 2. Real-time monitoring: choose "infrared temperature real-time monitoring system", "human body detection real-time monitoring system", "ultrasonic ranging real-time monitoring system", "buzzer real-time monitoring system", "starting rod real-time monitoring system", "whether the tested person enters the real-time monitoring system". These monitoring data are analyzed by fuzzy algorithm and presented by long-distance wireless transmission. 3. Depth image recognition: the system can convolve the input feature map to produce the corresponding output. Through a certain convolutional layer calculation, it can deal with a variety of image forms, which greatly improves the recognition accuracy and efficiency.