跳到主要內容

臺灣博碩士論文加值系統

(216.73.216.240) 您好!臺灣時間:2026/06/13 11:18
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:劉泓岑
研究生(外文):HONG-CEN LIU
論文名稱:結合一維和二維卷積神經網路進行跌倒偵測
論文名稱(外文):Combining 1D & 2D Convolution Neural Networks For Fall Detection
指導教授:謝尚琳
指導教授(外文):Shang-Lin Hsieh
口試委員:謝尚琳
口試委員(外文):Shang-Lin Hsieh
口試日期:2019-07-22
學位類別:碩士
校院名稱:大同大學
系所名稱:資訊工程學系(所)
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2019
畢業學年度:107
語文別:中文
論文頁數:66
中文關鍵詞:跌倒偵測加速度計神經網路
外文關鍵詞:fall detectionaccelerometerneural network
相關次數:
  • 被引用被引用:2
  • 點閱點閱:281
  • 評分評分:
  • 下載下載:99
  • 收藏至我的研究室書目清單書目收藏:0
目前跌倒偵測機制依照感測器放置的位置能分為兩類,分別為「放置於環境」與「放置於人體上」,「放置於環境」方法的感測器通常對擺放的場地有所要求,且感測器的價錢也不會太便宜;相較之下「放置於人體上」方法的感測器不僅價格實惠且對擺放的場地也沒有限制。「放置於人體上」方法通常會使用智慧型手機當成感測器,因為智慧型手機除了普及率高之外,本身就可以完成偵測、判斷、發出通知。早期使用智慧型手機的研究方法,大多要求使用者須將手機固定擺放於身體某處,但現實情況下,每個人擺放手機的習慣都不同,因此就必須考量到使用者會將手機擺放到不同位置,而不是固定於單一位置。
有鑒於此,本文提出一種使用加速度計與兩種神經網路進行跌倒偵測的機制,先利用第一種神經網路排除非跌倒事件,再利用第二種神經網路確認跌倒事件。此外,本機制不限制使用者需將手機固定擺放於單一位置,而允許使用者將手機放置於衣服、褲子或外套口袋中。根據實驗結果,本機制的特異度跟準確度皆優於比較的三個方法,而靈敏度只小輸其中一個方法,而優於其他方法。
Fall detection mechanism can be sorted by the position where the sensor is placed. One is “Placed in the environment” and the other is “Attached on the human body”. The sensor of the first method usually required to be placed in specific place, and the price of the sensor is expensive. In contrast, the sensor of the second method is not only affordable but also no restrictions on the place. The second method usually uses a smartphone as a sensor, because the smartphone can complete the detection, determination, and notification by itself. The initial research in detection methods of smartphone had to place smartphone somewhere specifically on the body. But in reality, phone users have different habits of smartphone placement. It is important to consider that users will place their smartphone in different positions instead of being placed a single position.
In view of this, this paper proposes a mechanism for using accelerometer and two neural networks for fall detection. First, using the first neural network to exclude non-fall event. The second neural network will confirm the fall event. In addition, the mechanism does not require users to place the smartphone on a single body part allowing users to place the smartphone in a garment, trousers or jacket pocket. According to experimental results
in this paper, the Specificity and Accuracy of this mechanism are better than the three other methods. The Sensitivity of this method is slightly lower than one of three method, but better than other two.
致謝 i
摘要 ii
ABSTRACT iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 導論 1
1.1 研究背景 1
1.2 研究動機與目的 4
1.3 論文架構 4
第二章 背景知識與相關文獻 5
2.1 加速度計 5
2.2 訊號向量強度 5
2.3 低通濾波 8
2.4 快速傅立葉轉換 10
2.5 Convolution Neural Network 11
2.6 李昕容之跌倒偵測研究 15
2.7 Nur Syazarin Natasha Abd Aziz等人之跌倒偵測研究 17
2.8 Nantakrit Yodpijit等人之跌倒偵測研究 18
第三章 基於卷積神經網路的跌倒偵測 19
3.1 偵測流程 19
3.2 公式說明 21
3.3 五個階段的機制 22
第四章 實驗結果與討論 32
4.1 實驗設備與項目 32
4.2 實驗數據與分析 34
4.3 實驗結果 43
第五章 結論與未來展望 60
5.1 結論 60
5.2 未來展望 61
參考文獻 62
[1] 國家發展委員會「人口推估 高齡化時程」,2019年7月2日,檢自:
https://www.ndc.gov.tw/Content_List.aspx?n=695E69E28C6AC7F3
[2] 行政院主計處「國情統計通報」,2019年7月2日,檢自:
https://www.stat.gov.tw/public/Data/8928102859N41P4GNS.pdf
[3] Rinat Khusainov, Djamel Azzi, Ifeyinwa E. Achumba and Sebastian D. Bersch, “Real-Time Human Ambulation, Activity, and Physiological Monitoring: Taxonomy of Issues, Techniques, Applications, Challenges and Limitations,” Sensors 2013, Volume 13, Issue 10, pp. 12852-12902, Jul. 2013.
[4] Jia Dongyao, Viocu Groza, Liu XiaoHui, Liu Xu, and Zhu Huaihua, Qingsheng Zeng, “A Method of Detecting Human Body Falling Action in a Complex Background,” 2016 IEEE 20th Jubilee International Conference on Intelligent Engineering Systems (INES), pp. 51-56, Jul. 2016.
[5] Bo Yu Su, K. C. Ho, Marilyn J. Rantz, and Marjorie Skubic, ”Doppler Radar Fall Activity Detection Using the Wavelet Transform,” IEEE Transactions on Biomedical Engineering, Volume 62, Issue 3, pp. 865-875, Mar. 2015.
[6] Andreas Ejupi, Chantel Galang, Omar Aziz, Edward J. Park, and Stephen Robinovitch, “Accuracy of a Wavelet-based Fall Detection Approach Using an Accelerometer and a Barometric Pressure Sensor,” 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 2150-2153, Jul. 2017.
[7] Edgar T. Horta, Ivo C. Lopes, Joel J. P. C. Rodrigues, and Mario L. Proença, “A Mobile Health Application for Falls Detection and Biofeedback Monitoring,” 2013 IEEE 15th International Conference on e-Health Networking, Applications and Services (Healthcom 2013), pp. 36-40, Oct. 2013.
[8] Nuth Otanasap, “Pre-Impact Fall Detection Based on Wearable Device Using Dynamic Threshold Model,” 2016 17th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT), pp. 362-365, Dec. 2016.
[9] 李昕容,「可擺放於不同衣褲口袋之智慧型手機跌倒偵測機制」,大同大學資訊工程所碩士論文,2018年。
[10] Anagha Purushothaman, K.V. Vineetha, and Dhanesh G. Kurup, “Fall Detection System Using Artificial Neural Network,” 2018 Second International Conference on Inventive Communication and Computational Technologies (ICICCT), pp. 1146-1149, Apr. 2018.
[11] Nur Syazarin Natasha Abd Aziz, Salwani Mohd Daud, Hafiza Abas, Sya Azmeela Shariff, and Nur Qamarina Mohd Noor, “Fall Detection System: Signal Analysis in Reducing False Alarms Using Built-in Tri-axial Accelerometer,” 2018 2nd International Conference on BioSignal Analysis, Processing and Systems (ICBAPS), pp. 70-74, Jul. 2018.
[12] Gao Tongyue, Yang Jia, Huang Kaida, Hu Qingxuan, and Zhao Fengshou, “Research and Implementation of Two-Layer Fall Detection Algorithm,” 2018 5th International Conference on Systems and Informatics (ICSAI), pp. 70-74, Nov. 2018.
[13] Nantakrit Yodpijit, Teppakorn Sittiwanchai, and Manutchanok Jongprasithporn, “The Development of Artificial Neural Networks (ANN) for Falls Detection,” 2017 3rd International Conference on Control, Automation and Robotics (ICCAR), pp. 547-550, Apr. 2017.
[14] Numpy API的FFT說明,2019年7月2日,檢自:
https://docs.scipy.org/doc/numpy/reference/routines.fft.html#rfb1dc64dd6a5-nr
[15] Cooley-Tukey演算法,2019年7月2日,檢自:
https://zh.wikipedia.org/wiki/%E5%BF%AB%E9%80%9F%E5%82%85%E9%87%8C%E5%8F%B6%E5%8F%98%E6%8D%A2
[16] 簡易CNN架構圖,2019年7月2日,檢自:
https://chtseng.wordpress.com/2017/09/12/%E5%88%9D%E6%8E%A2%E5%8D%B7%E7%A9%8D%E7%A5%9E%E7%B6%93%E7%B6%B2%E8%B7%AF/
[17] Tensorflow,2019年7月2日,檢自:
https://www.tensorflow.org/
[18] Keras,2019年7月2日,檢自:
https://keras.io/
[19] Conv1D,2019年8月1日,檢自:
https://keras-cn.readthedocs.io/en/latest/layers/convolutional_layer/
[20] Conv2D,2019年8月1日,檢自:
https://keras-cn.readthedocs.io/en/latest/layers/convolutional_layer/#conv2d
[21] Pooling Layer,2019年8月1日,檢自:
https://keras-cn.readthedocs.io/en/latest/layers/pooling_layer/
[22] OverFitting,2019年8月1日,檢自:
https://morvanzhou.github.io/tutorials/machine-learning/ML-intro/3-05-overfitting/
[23] Relu Function,2019年8月1日,檢自:
https://zh.wikipedia.org/wiki/%E7%BA%BF%E6%80%A7%E6%95%B4%E6%B5%81%E5%87%BD%E6%95%B0
[24] Activation Function,2019年8月1日,檢自:
https://www.zhihu.com/question/22334626
[25] 梯度消逝問題,2019年8月1日,檢自:
https://zh.wikipedia.org/wiki/%E6%A2%AF%E5%BA%A6%E6%B6%88%E5%A4%B1%E9%97%AE%E9%A2%98
[26] Relu Function示意圖,2019年7月2日,檢自:
https://zh.wikipedia.org/wiki/%E7%BA%BF%E6%80%A7%E6%95%B4%E6%B5%81%E5%87%BD%E6%95%B0
[27] SoftMax公式,2019年7月2日,檢自:
https://zh.wikipedia.org/wiki/Softmax%E5%87%BD%E6%95%B0
[28] MobileNet,2019年8月1日,檢自:
https://ai.googleblog.com/2017/06/mobilenets-open-source-models-for.html
[29] MobileNetV2,2019年8月1日,檢自:
https://ai.googleblog.com/2018/04/mobilenetv2-next-generation-of-on.html
[30] Matplotlib,2019年7月2日,檢自:
https://matplotlib.org/
[31] 交叉驗證,2019年7月2日,檢自:
https://zh.wikipedia.org/wiki/%E4%BA%A4%E5%8F%89%E9%A9%97%E8%AD%89
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊