Sleep state monitoring energy-saving old-age-aiding system based on deep learning image recognition

The invention discloses a sleep state monitoring energy-saving old-age aiding system based on deep learning image recognition, and the system consists of front-end equipment, a small server, a power supply control module, and rear-end equipment, and can collect images and upload the images to the small server through a network. An image processing system in the small server analyzes the image by using a human eye state recognition detection technology and a limb recognition detection technology, and sends a control signal to a power supply control module after it is judged that the old personfalls asleep; a Bluetooth communication device in the power supply control module receives the control signal of the small server and generates a corresponding infrared control signal, and an infraredtransmitting device in the module controls the volume, the brightness and the power supply of the television, so that the purpose of energy conservation is achieved; and the small server also regularly sends a report to the r