隨著近年來載板效能越來越強，物聯網已不再是早期單純數據傳輸的低效能產
品了，加上近年來AI 研究的興盛，結合兩者功能使其能夠便利人們的生活已成為趨
勢，兩者結合的產物現在有了新的名詞－人工智慧物聯網(AIOT)，而如何在效能與
功能間作取捨成為了人工智慧物聯網討論的最重要主題之一。
本論文可分為兩個部分，第一個部分是與我們平常生活息息相關的居家系統，居家
系統我們將劃分為「環境」與「安全」做討論，第二部分是與居住處相關的大樓管
理系統。
「環境」的部分我們使用了感測器來檢測我們生活周遭的生活環境，以幫助使
用者能夠清楚知道個人生活周遭之汙染源，並使用資料庫儲存，在檢測物質超標時，
也有緊急回報系統，讓使用者可以即時處置。「安全」部分只有在使用者設定開啟時
才開啟，我們使用了人體傳感器當作人體監測系統，一旦偵測到人體訊號，便會發
送給使用者，我們同時也使用移動偵測系統，影像偵測到移動物體會對其做紀錄，
將紀錄之資料，與資料庫內使用者設定資料做辨識，辨識的方式為KNN 臉部辨識，
一旦辨識為資料庫內沒有之陌生人將發送警告給使用者。
大樓管理部分，我們分為社區之電鈴，以及社區之環境偵測，電鈴部份我們使
用了Twilio 作為電信服務，以供使用者就算不在家也可以收到他人來訪的通知，同
時也將設置相機，將按電鈴者的照片傳輸給使用者，讓其知道誰為按電鈴者，環境
部分我們使用了PMS5003，用以偵測大樓之空氣品質，順便也可偵測是否有人在禁
菸區抽菸。

With the powerful of small single-board computers in recently. IOT were no longer
just designed for the basic data translation. Because of the AI research getting popular these
days, the combination between AI and IOT to help people become the most recent trends.
The combination between both also has a new proper noun-“AIOT”. How to get balance
between efficacy and functional is the most important object in AIOT.
This paper be can separate into two parts. The first part we will discuss is “Home
system” which around to us. “Home system” includes the “Environment detection system”
and “Security system”. The second part we will discuss about is the ”Community managing
system”.
“Environment detection system” was built by multiple senor to detect the
environmental data. User can use the data to get information from the pollution material.
The data will be saved in the database. Once the number of pollution material has raise
higher than the safety line, E-mail will send a mail to announce the user. Let user has time
to deal with it. “Security system” will start only if user turn it on. We used PIR as human
body detector. Once human body been detected we will sent a warning to the user. We also
built a motion detector based on image processing. Once system detecting something is
moving, we will save the image. The image will send to user by E-mail. It will also operate
in our face recognition system which based on KNN. If people were detected as a stranger,
the system will send another notification to user.
”Community managing system” can be also separated into two parts. One of it is
environment detection based on PMS5003, another is the community doorbell system
based on Twilio. Environmental detection can catch people who smokes illegal indoor. The
purpose of doorbell is to let outdoor user can get a phone call from visitor who is outside
and waiting.

[1]台北市政府，台北市智慧城市推動概念與機制。取自:
https://drive.google.com/file/d/14UYCuGkfk25rkOlEUat99r5qIbZMR9qw/view
[2]J.Gutiérrez Bayo,”International Case Studies of Smart Cities
Santander Spain”，Inter-American Development Bank，June 2016
[3] 8051、8052 比較。取自:《微電腦控制: 8051/8052 專題實習(C 語言)》，全華
圖書，作者:王振裕，2012 年
[4]Wikipedia，Raspberry Pi。取自:
https://zh.wikipedia.org/wiki/%E6%A0%91%E8%8E%93%E6%B4%BE
[5]STM 官網，STM32F1 MCU Series。取自:
https://www.st.com/content/ccc/fragment/product_related/series_information/series
_level_diagram/06/79/fc/3f/ab/01/44/2e/STM32F1_series_SS1031.jpg/files/STM32
F1_series_SS1031.jpg/_jcr_content/translations/en.STM32F1_series_SS1031.jpg
[6]Wikipedia，Linux Distro Timeline。取自:
https://wiki.sars.tw/lib/exe/fetch.php?media=linux:linux_distro_timeline-7.2.png
[7]Wikipedia，OpenCV。取自:
https://zh.wikipedia.org/wiki/OpenCV
[8]University of Texas at Tyler Department of Electrical Engineering，
The basic structure of OpenCV。取自:
https://www.researchgate.net/figure/The-basic-structure-of-
OpenCV_fig2_252973402
[9]Wikipedia，LAMP。取自:
https://zh.wikipedia.org/wiki/LAMP
[10]MySQL，MySQL LOGO。取自:
https://cdn.worldvectorlogo.com/logos/mysql-3.svg
[11]Apache，Apache LOGO。取自:
https://hostingwiki.org/wp-content/uploads/2019/09/apache.png
76
[12]Wikipedia，PHP。取自:
https://zh.wikipedia.org/wiki/PHP#PHP%E5%AE%98%E7%BD%91%E8%A2%A
B%E9%A7%AD%E5%85%A5%E4%BA%8B%E4%BB%B6
[13]Python 官網，Python Logo。取自:
https://www.python.org/community/logos/
[14]Wikipedia，藍芽。取自:
https://zh.wikipedia.org/wiki/%E8%97%8D%E7%89%99#%E5%B9%B2%E6%93
%BE
[15]Allion，Bluetooth Logo。取自:
https://www.allion.com.tw/
[16]Miro Medium，LORA LOGO。取自:
https://miro.medium.com/max/833/1*3CYmP-gCt-xXp7QcGGybXg.png
[17]Wikipedia，星型網。取自:
https://zh.wikipedia.org/wiki/%E6%98%9F%E5%9E%8B%E7%BD%91
[18]Miro Medium，Twillio LOGO。取自:
https://miro.medium.com/max/2600/1*RgTiFJX6Aa7ftVjqXHcm2Q.png
[19]Ohhoon Kwon ; Kern Koh，”Swap space management technique for
portable consumer electronics with NAND flash memory”，IEEE Transactions on
Consumer Electronics，28 October 2010
[20]Longtask，Swap 交換。取自:
https://www.longtask.net/css/pics/2018-11-15-swap.png
[21]NCKU，I2C。取自:
http://wiki.csie.ncku.edu.tw/embedded/I2C
[22]Wikipedia，SMTP。取自:
https://zh.wikipedia.org/wiki/%E7%AE%80%E5%8D%95%E9%82%AE%E4%BB
%B6%E4%BC%A0%E8%BE%93%E5%8D%8F%E8%AE%AE
[23] V. Garcia, E. Debreuve, F. Nielsen, and M. Barlaud, “K-nearest
neighbor search: Fast GPU-based implementations and application to highdimensional
feature matching”, 2010 IEEE International Conference on Image
Processing, 2010
77
[24] G. B. Huang, M. Ramesh, T. Berg, and E. L. Miller,“ Labeled Faces
in the Wild: A Database for Studying Face Recognition in Unconstrained
Environments”,Workshop on faces in’Real-Life’Images: detection alignment and
recognition,2008
[25] U. Nanda and S. K. Pattnaik, "Universal Asynchronous Receiver and
Transmitter (UART)", 2016 3rd International Conference on Advanced Computing
and Communication Systems (ICACCS), pp. 1-5, 2016.
[26]King Tzeng，PIR。取自:
https://ithelp.ithome.com.tw/articles/10225375
[27]Gentec-Eo，PYROELECTRIC THERMAL DETECTORS，取自:
https://www.gentec-eo.com/discrete-pyroelectric-sensors
[28]Edmundoptics，Frenzel Lens。取自:
https://www.edmundoptics.com.tw/globalassets/knowledge-center/app-notes/fig-1-
afl.gif
[29]Blogspot，How PIR Sensor Works and How To Use It with Arduino。取自:
https://1.bp.blogspot.com/-
YLk9eGLur2Y/XZoQVKszr6I/AAAAAAAAUJY/RcFM06lXBUQtGXiag2sm6MI
Qqu8N3Ls_ACLcBGAsYHQ/s1600/
[30]Cubie，HC-05 與HC-06 藍牙模組補充說明。取自:
https://swf.com.tw/?p=693
[31]Twillio，Twillio Graph。取自:
https://www.twilio.com/docs/taskrouter/contact-center-blueprint/dashboards
[32]Semtech，各個傳輸方式之比較。取自:
https://www.semtech.cn/uploads/images/LoRa_Why_Range.png