Non-contact sleep monitoring method based on deep learning and multi-parameter fusion

• Main Authors: LIU JUANXIU, HE TAO, YAN BOYUN, WANG XIANGZHOU, LIU LIN, DU XIAOHUI, LIU YONG, ZHANG JING, SUN HAIXIN
• Format: Patent
• Language: Chinese; English
• Published: 02.09.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DIAGNOSIS; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; MEDICAL OR VETERINARY SCIENCE; PHYSICS; SURGERY

The invention discloses a non-contact sleep monitoring system based on deep learning and multi-parameter fusion, and belongs to the field of image processing and the field of deep learning. The system firstly segments an acquired sleep video image, then builds a deep convolutional neural network for physiological signal extraction and amplification, amplifies a heart rate signal of a forehead area and an eye movement frequency of an eye area by setting different amplification factors of the network, and finally obtains an eye movement signal of the forehead area and the eye movement frequency of the eye area. The forehead area video image with the amplified heart rate signal and the eye area video image with the amplified eye movement frequency are obtained, corresponding frequency spectrums are extracted through fast Fourier transform, and the frequency corresponding to the peak value of the frequency spectrums is found out to serve as the monitored heart rate signal and the monitored eye movement frequency.