基于4D點(diǎn)云和航跡信息的人員跌倒檢測方法
電子技術(shù)應(yīng)用
周楊,李劍鵬,王知雨,梁慶真
四川啟??丝萍加邢薰荆拇?成都 610041
摘要: 在家庭和醫(yī)療健康監(jiān)護(hù)中,及時(shí)準(zhǔn)確對老人跌倒進(jìn)行實(shí)時(shí)監(jiān)護(hù)是一個(gè)重要課題,而毫米波雷達(dá)在非接觸式人員跌倒檢測方面表現(xiàn)出巨大應(yīng)用潛力。傳統(tǒng)方法主要利用毫米波雷達(dá)獲取的多普勒信息作為跌倒檢測特征,但對于某些場景,諸如人員跌倒時(shí)與雷達(dá)的距離較遠(yuǎn)或者跌倒方向與雷達(dá)天線法向垂直,多普勒信息特征的有效性就會(huì)降低,進(jìn)而導(dǎo)致跌倒檢測準(zhǔn)確率下降。針對上述問題,提出了一種基于4D點(diǎn)云和航跡信息的毫米波雷達(dá)跌倒檢測方法,利用點(diǎn)云的空間信息、時(shí)間信息和微動(dòng)信息全方位地表征人體動(dòng)作特征,有效緩解距離和姿態(tài)多樣性對雷達(dá)跌倒檢測的影響,同時(shí)結(jié)合跟蹤航跡信息帶來的濾波器效應(yīng)進(jìn)一步抑制環(huán)境雜波的影響,提高了跌倒檢測準(zhǔn)確性。
中圖分類號(hào):TP391.9 文獻(xiàn)標(biāo)志碼:A DOI: 10.16157/j.issn.0258-7998.234228
中文引用格式: 周楊,李劍鵬,王知雨,等. 基于4D點(diǎn)云和航跡信息的人員跌倒檢測方法[J]. 電子技術(shù)應(yīng)用,2024,50(1):120-124.
英文引用格式: Zhou Yang,Li Jianpeng,Wang Zhiyu,et al. Human fall detection method based on 4D point cloud and track information[J]. Application of Electronic Technique,2024,50(1):120-124.
中文引用格式: 周楊,李劍鵬,王知雨,等. 基于4D點(diǎn)云和航跡信息的人員跌倒檢測方法[J]. 電子技術(shù)應(yīng)用,2024,50(1):120-124.
英文引用格式: Zhou Yang,Li Jianpeng,Wang Zhiyu,et al. Human fall detection method based on 4D point cloud and track information[J]. Application of Electronic Technique,2024,50(1):120-124.
Human fall detection method based on 4D point cloud and track information
Zhou Yang,Li Jianpeng,Wang Zhiyu,Liang Qingzhen
Sichuan Cric Technology Co., Ltd., Chengdu 610041, China
Abstract: In family and medical health monitoring, timely and accurate real-time monitoring of elderly falls is an important topic, and millimeter wave radar has shown great application potential in non-contact fall detection. The traditional method mainly uses the Doppler information obtained by millimeter wave radar as the fall detection feature. However, for some scenarios, such as the distance between the person and the radar or the direction of the fall is perpendicular to the normal of the radar antenna, the effectiveness of the Doppler information feature will be reduced, and the fall detection accuracy will be reduced.
Key words : millimeter wave radar;4D point cloud;fall detection
引言
我國已進(jìn)入快速老齡化階段且老年人口基數(shù)大,因跌倒帶來的健康及經(jīng)濟(jì)損失已成為重大的社會(huì)問題[1]。隨著雷達(dá)芯片技術(shù)的進(jìn)步,毫米波雷達(dá)在非接觸式人員跌倒檢測方面表現(xiàn)出巨大應(yīng)用潛力[2]。毫米波雷達(dá)在跌倒檢測領(lǐng)域具有獨(dú)特的優(yōu)勢,與視頻設(shè)備相比,避免了用戶隱私泄露的風(fēng)險(xiǎn),與陀螺儀、加速度計(jì)等可穿戴設(shè)備相比,不影響用戶的日常生活,不會(huì)出現(xiàn)忘記佩戴設(shè)備導(dǎo)致設(shè)備失效的情況[3]。
本文詳細(xì)內(nèi)容請下載:
http://theprogrammingfactory.com/resource/share/2000005846
作者信息:
周楊,李劍鵬,王知雨,梁慶真
(四川啟??丝萍加邢薰?,四川 成都 610041)
此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權(quán)禁止轉(zhuǎn)載。