《電子技術(shù)應(yīng)用》
您所在的位置:首頁 > 人工智能 > 設(shè)計(jì)應(yīng)用 > 基于毫米波雷達(dá)三維點(diǎn)云的室內(nèi)跌倒檢測(cè)
基于毫米波雷達(dá)三維點(diǎn)云的室內(nèi)跌倒檢測(cè)
電子技術(shù)應(yīng)用
李偉1,李丹丹1,丁奇寧1,馬裕燚2,耿永福1
1.北方工業(yè)大學(xué) 信息學(xué)院;2.北方工業(yè)大學(xué) 電氣與控制工程學(xué)院
摘要: 全球老齡化時(shí)代的到來引發(fā)的老年人健康監(jiān)護(hù)問題不可忽視,而室內(nèi)跌倒對(duì)獨(dú)居的老年人有非常大的安全隱患。因此,為準(zhǔn)確檢測(cè)到跌倒動(dòng)作,使用毫米波雷達(dá)三維點(diǎn)云信息進(jìn)行室內(nèi)跌倒檢測(cè),并提出一種基于外部注意力機(jī)制的PointLSTM網(wǎng)絡(luò)實(shí)現(xiàn)三維點(diǎn)云在時(shí)序的分類。通過MIMO體制的毫米波雷達(dá)芯片采集人體動(dòng)作的回波信號(hào),利用集成雷達(dá)基帶處理器的微控制器實(shí)現(xiàn)信號(hào)處理的部分,可將原始數(shù)據(jù)實(shí)時(shí)轉(zhuǎn)換成三維點(diǎn)云,并提高點(diǎn)云處理中的計(jì)算速度及雷達(dá)硬件的整體性能?;谕獠孔⒁饬C(jī)制的PointLSTM網(wǎng)絡(luò)可實(shí)現(xiàn)點(diǎn)云在時(shí)空中的提取特征和分類識(shí)別,網(wǎng)絡(luò)改進(jìn)了PointLSTM幀間點(diǎn)信息的流失問題,并在信息提取中對(duì)所有數(shù)據(jù)實(shí)現(xiàn)特征聯(lián)系,外部注意力機(jī)制通過獨(dú)立的可學(xué)習(xí)參數(shù)優(yōu)化了網(wǎng)絡(luò)復(fù)雜度和識(shí)別精確率。實(shí)驗(yàn)結(jié)果表明,所提出的方法在室內(nèi)環(huán)境下檢測(cè)準(zhǔn)確率可以達(dá)到98.3%,可以有效區(qū)分動(dòng)作的類別,并驗(yàn)證了使用毫米波雷達(dá)三維點(diǎn)云檢測(cè)人體跌倒的可行性。
中圖分類號(hào):TN95 文獻(xiàn)標(biāo)志碼:A DOI: 10.16157/j.issn.0258-7998.245007
中文引用格式: 李偉,李丹丹,丁奇寧,等. 基于毫米波雷達(dá)三維點(diǎn)云的室內(nèi)跌倒檢測(cè)[J]. 電子技術(shù)應(yīng)用,2024,50(9):59-66.
英文引用格式: Li Wei,Li Dandan,Ding Qining,et al. Indoor fall detection based on millimeter-wave radar three-dimensional point cloud[J]. Application of Electronic Technique,2024,50(9):59-66.
Indoor fall detection based on millimeter-wave radar three-dimensional point cloud
Li Wei1,Li Dandan1,Ding Qining1,Ma Yuyi2,Geng Yongfu1
1.School of Information Science and Technology, North China University of Technology; 2.School of Electrical and Control Engineering, North China University of Technology
Abstract: The advent of the global aging era has brought critical issues concerning the elderly health care to light, and indoor falls pose a significant safety risk to seniors who live alone. Therefore, in order to accurately detect the action of falling, this paper uses millimeter-wave radar 3D point cloud data for indoor fall detection and introduces a PointLSTM network based on an external attention mechanism to classify 3D point clouds over time. The millimeter-wave radar chip of the MIMO system collects the echo signal of human movements, and the signal processing part is realized by using the microcontroller integrated with the radar baseband processor, which can convert the raw data into a three-dimensional point cloud in real time, and improve the computing speed in point cloud processing and the overall performance of radar hardware. The PointLSTM network based on external attention mechanism enables spatial and temporal feature extraction and classification of point clouds. The network addresses the loss of point information between frames in PointLSTM and links features across all data during information extraction. The external attention mechanism, with its independent learnable parameters, optimizes network complexity and recognition accuracy. Experimental results show that the proposed method achieves a detection accuracy of 98.3% in indoor environments, effectively differentiating between types of motions and confirming the feasibility of using millimeter-wave radar 3D point clouds for detecting human falls.
Key words : millimeter-wave radar;point cloud classification;signal processing;deep learning

引言

據(jù)世界衛(wèi)生組織報(bào)道,世界各國老年人的數(shù)量和占比都出現(xiàn)上升趨勢(shì),而老年人因跌倒而出現(xiàn)重傷和死亡的風(fēng)險(xiǎn)最大[1]。跌倒不僅對(duì)老年人造成身體傷害,也會(huì)引起消極恐懼的情緒,能夠及時(shí)檢測(cè)到跌倒并作出提醒尤為重要。因此,本文使用毫米波雷達(dá)三維點(diǎn)云進(jìn)行人體跌倒檢測(cè),通過對(duì)人體姿態(tài)的分類準(zhǔn)確地檢測(cè)出人體跌倒的行為,增強(qiáng)獨(dú)居老年人的安全保障。

實(shí)現(xiàn)跌倒檢測(cè)的方法有基于可接觸式和非接觸式設(shè)備。常見的人體跌倒檢測(cè)的可接觸式設(shè)備是基于加速度傳感器[2]和重力傳感器[3]等,但需隨身攜帶,影響日常生活,還會(huì)因未及時(shí)充電產(chǎn)生誤報(bào)現(xiàn)象。非接觸式設(shè)備主要有紅外、Wi-Fi、攝像機(jī)和雷達(dá)等,基于攝像機(jī)的方法容易侵犯?jìng)€(gè)人隱私,易受環(huán)境、信號(hào)的影響,激光雷達(dá)和超寬帶雷達(dá)易受極端天氣的影響,且價(jià)格比較昂貴,而毫米波雷達(dá)測(cè)量精度高、可全天時(shí)全天候工作,性價(jià)比高。因此本文使用毫米波雷達(dá)進(jìn)行人體跌倒檢測(cè)。

目前基于深度學(xué)習(xí)的毫米波雷達(dá)跌倒檢測(cè)方法大多是使用深度卷積神經(jīng)網(wǎng)絡(luò)(CNN)對(duì)二維圖像進(jìn)行空間特征提取[4]。相比雷達(dá)的二維圖像,三維點(diǎn)云圖更直觀形象,包含更豐富的有用信息?;趩螏c(diǎn)云的識(shí)別中,Pointnet網(wǎng)絡(luò)[5]可通過排列不變的最大池化實(shí)現(xiàn)全局特征提取,而Pointnet++[6]在前者的基礎(chǔ)上利用分層分組實(shí)現(xiàn)局部特征的提取,不斷迭代實(shí)現(xiàn)全局特征提取。Pantomime網(wǎng)絡(luò)[7]中結(jié)合Pointnet++和LSTM網(wǎng)絡(luò),提取所有幀的全局特征實(shí)現(xiàn)對(duì)手勢(shì)的分類。FlickerNet模型[8]修改了分組操作,從相鄰幀中提取運(yùn)動(dòng)和結(jié)構(gòu)特征,但缺乏捕獲長期關(guān)系的能力。PointLSTM模型[9]在此基礎(chǔ)上,提出一種關(guān)于無序點(diǎn)云的新型LSTM單元,用于捕獲點(diǎn)級(jí)別的長期關(guān)系。

受到以上方法的啟發(fā),本文提出一種基于外部注意力機(jī)制的PointLSTM網(wǎng)絡(luò)結(jié)構(gòu)實(shí)現(xiàn)點(diǎn)云在時(shí)空中的特征提取和分類識(shí)別,網(wǎng)絡(luò)改進(jìn)了PointLSTM幀間點(diǎn)信息的流失問題,并在信息提取中對(duì)所有數(shù)據(jù)實(shí)現(xiàn)特征聯(lián)系,外部注意力機(jī)制通過獨(dú)立的可學(xué)習(xí)參數(shù)優(yōu)化了網(wǎng)絡(luò)復(fù)雜度和識(shí)別精確率。


本文詳細(xì)內(nèi)容請(qǐng)下載:

http://theprogrammingfactory.com/resource/share/2000006143


作者信息:

李偉1,李丹丹1,丁奇寧1,馬裕燚2,耿永福1

(1.北方工業(yè)大學(xué) 信息學(xué)院,北京 100043;

2.北方工業(yè)大學(xué) 電氣與控制工程學(xué)院,北京 100043)


Magazine.Subscription.jpg

此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權(quán)禁止轉(zhuǎn)載。