作者：陈馨瑶，程进，邹小平，姜超，邬润杰，廖晶晶
单位：北京信息科技大学北京市传感器重点实验室，北京100101
中图分类号：TP277                
文献标识码：A                
文章编号：1006-883X(2022)01-0010-04
收稿日期：2021-09-22

摘要：婴儿看护是每个家长承担的责任，但在大多数家长需要兼顾家庭和工作的今天，传统的看护方式不再满足各个家庭的需求，“三胎”政策的开放更是加剧了婴儿看护的压力。为了能减轻家长的负担以及降低看护风险，论文提出了一种婴儿睡眠监控系统来辅助传统看护。该系统的核心为微型电脑（树莓派），在其控制摄像头对婴儿睡眠情况进行实时监控的同时，通过调用OpenCV等库实现了对监控视频流图像的识别，在识别到婴儿有吐奶、睡觉张嘴、口鼻被遮挡等异常状态时，会通过微信小程序发送消息与婴儿图片提醒家长。该系统能及时针对不同的异常情况发出不同的提醒，并且程序移植到了嵌入式平台，体积小巧，操作简便。
关键词：婴儿看护；微型电脑；图像识别

Baby Sleep Monitoring System Based on Artificial Intellignce
CHEN Xinyao, CHENG Jin, ZOU Xiaoping, JIANG Chao, WU Runjie, LIAO Jingjing
(Beijing Key Laboratory of Sensors, Beijing Information Science and Technology University, Beijing 100101, China)
Abstract: Babysitting is the responsibility of every family, but with parents having to juggle family and work, traditional care is no longer sufficient to meet the needs of each family. In order to reduce the burden on parents and the risk to the family, the paper proposes a baby sleep monitoring system to complement traditional caregiving. The core of the system is a microc-omputer (Raspberry Pi). While it controls the camera to monitor the babys sleep in real time, it also can recognize the surve-illance video streams image by calling OpenCV and other libraries. When an abnormal state such as a baby vomiting milk, sleeping mouth, mouth and nose being blocked, etc. is recognized, it will send messages and baby pictures to remind parents through the WeChat applet. The system can send out corresponding reminders in time for different abnormal situations, and the program is transplanted to the embedded platform, which is small in size and easy to operate.
Key words: babysitting; microcomputers; image recognition

阅读全文

备注：2022年 第28卷 第01期