本文選題：智慧家居　切入點(diǎn)：老人日常行為識(shí)別　出處：《杭州電子科技大學(xué)》2017年碩士論文

【摘要】：隨著人口老齡化情況加劇、人力資源短缺,養(yǎng)老成為當(dāng)前社會(huì)面臨的主要問題。家庭環(huán)境中自動(dòng)識(shí)別老人日常行為,發(fā)現(xiàn)老人行為異常,提高老人獨(dú)立生活能力和家庭健康護(hù)理水平,是緩解養(yǎng)老困境的可行方法。本文以華盛頓州立大學(xué)智能空間實(shí)驗(yàn)室的實(shí)驗(yàn)數(shù)據(jù)為基礎(chǔ),研究機(jī)器學(xué)習(xí)理論、數(shù)據(jù)處理技術(shù),結(jié)合智慧家居相關(guān)技術(shù),識(shí)別老人日常行為,了解老人的意圖,發(fā)現(xiàn)老人生活中的行為異常與環(huán)境異常。主要研究?jī)?nèi)容如下所示:(1)設(shè)計(jì)傳感器事件動(dòng)態(tài)分割與標(biāo)注方法。根據(jù)活動(dòng)執(zhí)行過程中所觸發(fā)傳感器之間的依賴關(guān)系,計(jì)算相鄰窗口的相似度,過濾行為邊界點(diǎn),保留相似度最低的點(diǎn),最后合并相鄰的且具有相同標(biāo)簽的序列片段。(2)構(gòu)建以較高精度識(shí)別老人多種日常行為的模型。根據(jù)老人日常行為特點(diǎn),構(gòu)建多維度特征框架,包括傳感器權(quán)重、傳感器類別特征、時(shí)間特征和頻率特征;使用遺傳算法和交叉驗(yàn)證對(duì)行為模型進(jìn)行訓(xùn)練與調(diào)優(yōu),建立基于支持向量機(jī)的日常行為模型。(3)設(shè)計(jì)老人異常行為檢測(cè)框架。結(jié)合日常行為模型,選擇重要特征,建立老人各類行為的高斯混合模型,計(jì)算異常樣本與所屬成分的偏離程度,對(duì)比各類特征與正常樣本特征之間的差異。日常行為識(shí)別是實(shí)現(xiàn)老人生活輔助、提升老人獨(dú)立生活能力、提高健康護(hù)理水平的前提。設(shè)計(jì)多組實(shí)驗(yàn)評(píng)估本文方法和模型的性能。實(shí)驗(yàn)結(jié)果表明序列分割方法能夠?qū)鞲衅餍蛄羞M(jìn)行合理分割,日常行為模型能準(zhǔn)確的識(shí)別多種行為,異常檢測(cè)方法能夠高效地發(fā)現(xiàn)老人行為中的異常。研究結(jié)果對(duì)居家養(yǎng)老智慧化發(fā)展具有一定促進(jìn)作用。
[Abstract]:With the aggravation of the aging population and the shortage of human resources, old-age support has become the main problem facing the society.It is a feasible method to identify the daily behavior of the elderly automatically in the family environment, find out the abnormal behavior of the elderly, improve the independent living ability of the elderly and the level of family health care, which is a feasible method to alleviate the plight of the aged.Based on the experimental data from the Intelligent Space Laboratory of Washington State University, this paper studies the theory of machine learning, data processing technology, combining with smart home related technology, to identify the daily behavior of the elderly and to understand the intention of the elderly.To find out the abnormal behavior and environment in the old people's life.The main contents of this paper are as follows: (1) Design the sensor event dynamic segmentation and tagging method.According to the dependencies between sensors triggered during the execution of the activity, the similarity of adjacent windows is calculated, the behavior boundary points are filtered, and the points with the lowest similarity are retained.Finally, the adjacent sequence fragment with the same tag is merged to construct a model to identify various daily behaviors of the elderly with high accuracy.According to the characteristics of daily behavior of the elderly, a multi-dimensional feature framework is constructed, including sensor weight, sensor category feature, time feature and frequency feature, and the training and tuning of the behavior model using genetic algorithm and cross-validation.A support vector machine (SVM) based daily behavior model is established to design a framework for the detection of abnormal behaviors of the elderly.Combined with the daily behavior model, the Gao Si mixed model of all kinds of behaviors of the elderly was established, and the deviations between abnormal samples and their components were calculated, and the differences between different characteristics and normal sample characteristics were compared.Daily behavior recognition is the premise to realize the elderly's life assistance, improve the elderly's independent living ability and raise the level of health care.Several experiments were designed to evaluate the performance of the proposed method and model.The experimental results show that the sequence segmentation method can segment the sensor sequences reasonably, the daily behavior model can accurately identify a variety of behaviors, and the anomaly detection method can efficiently detect the anomalies in the behavior of the elderly.The results of the study have a certain role in promoting the development of home-based endowment intelligence.
【學(xué)位授予單位】：杭州電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TU855;TP181

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉一偉;;互補(bǔ)還是替代:“社會(huì)養(yǎng)老”與“家庭養(yǎng)老”——基于城鄉(xiāng)差異的分析視角[J];公共管理學(xué)報(bào);2016年04期

2 吳菲菲;張亞茹;黃魯成;欒靜靜;;面向老年人的環(huán)境輔助生活技術(shù)研究態(tài)勢(shì)分析[J];科技管理研究;2016年20期

3 張會(huì)君;孫鶴;;基于SWOT分析法M機(jī)構(gòu)護(hù)理服務(wù)價(jià)值分析[J];管理世界;2016年09期

4 白玫;朱慶華;;智慧養(yǎng)老現(xiàn)狀分析及發(fā)展對(duì)策[J];現(xiàn)代管理科學(xué);2016年09期

5 賀炎;王科;王忠民;;用戶無關(guān)的多分類器融合行為識(shí)別模型[J];西安郵電大學(xué)學(xué)報(bào);2016年05期

6 張佳驥;趙海英;;智慧養(yǎng)老計(jì)算環(huán)境研究狀況的回顧[J];計(jì)算機(jī)應(yīng)用與軟件;2016年08期

7 李鵬;;基于數(shù)據(jù)挖掘的大型企業(yè)人力資源需求預(yù)測(cè)研究[J];人力資源管理;2016年06期

8 林宇;黃迅;淳偉德;黃登仕;;基于ODR-ADASYN-SVM的極端金融風(fēng)險(xiǎn)預(yù)警研究[J];管理科學(xué)學(xué)報(bào);2016年05期

9 翟振武;陳佳鞠;李龍;;中國(guó)人口老齡化的大趨勢(shì)、新特點(diǎn)及相應(yīng)養(yǎng)老政策[J];山東大學(xué)學(xué)報(bào)(哲學(xué)社會(huì)科學(xué)版);2016年03期

10 李鋒;潘敬奎;;基于三軸加速度傳感器的人體運(yùn)動(dòng)識(shí)別[J];計(jì)算機(jī)研究與發(fā)展;2016年03期

相關(guān)會(huì)議論文 前1條

1 張香云;;基于EM算法缺失數(shù)據(jù)下混合模型的參數(shù)估計(jì)[A];第十三屆中國(guó)管理科學(xué)學(xué)術(shù)年會(huì)論文集[C];2011年

相關(guān)博士學(xué)位論文 前2條

1 董寶玉;支持向量技術(shù)及其應(yīng)用研究[D];大連海事大學(xué);2016年

2 仝鈺;基于條件隨機(jī)場(chǎng)的智能家居行為識(shí)別研究[D];大連海事大學(xué);2015年

相關(guān)碩士學(xué)位論文 前2條

1 張偉;智能空間下基于非視覺傳感器數(shù)據(jù)的人體行為識(shí)別方法研究[D];山東大學(xué);2015年

2 胡龍;基于智能手機(jī)的用戶行為識(shí)別技術(shù)研究與應(yīng)用[D];電子科技大學(xué);2015年

，

本文編號(hào)：1722057