作者：桂若霞，程进，朴林华，邹小平，李晓彤
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：TP391.41              
文献标识码：A              
文章编号：1006-883X(2022)02-0015-04
收稿日期：2021-09-25

摘要：随着社会的发展，我国人口老龄化越来越严重，因而增多了关于养老服务的需求。但目前很多养老机构的管理不规范，设备老旧不智能，没有跟上时代的步伐。老年人一旦发生意外情况，很难及时发现和救助。对此，文章提出一种老人摔倒求助视频检测系统，系统由一个微型计算机以及其自带的摄像头和一台上位机构成，通过摄像头对养老院老人进行拍照，微型计算机将拍摄的照片上传至上位机，在上位机中设计一个卷积神经网络模型用于检测老人是否摔倒和求救，便于及时提醒监护人施以援手。
关键词：姿势识别；人工智能；图像分类；微信提醒

Video Detection System for the Elderly in Nursing Home Fall and Seek Help
GUI Ruoxia, CHENG Jin, PIAO Linhua, ZOU Xiaoping, LI Xiaotong,
(Beijing Sensor Key Laboratory, Beijing Information Science & Technology University, Beijing 100101, China)
Abstract: With the development of society, the aging of Chinas population is becoming more and more serious, thus increasing the demand for pension services. But at present, the management of many pension institutions is not standardized and the equipment is old and not intelligent, which has not kept pace with the Times. Once the elderly happen accident, it is difficult to discover and help in time. To this, the paper puts forward a kind of the old man fell down turn video detection system, the system consists of a microcomputer and its own camera and into a table a institutions, through the camera take photo for the old man nursing home, the microcomputer to upload photos first machine, in the upper machine design a convolutional neural network model used to detect whether the old man fell down and ask for help, it is convenient to remind the guardian to help in time.
Key words: gesture recognition; artificial intelligence; image classification; WeChat remind

阅读全文

备注：2022年 第28卷 第02期