| 研究生: |
賴真珠 Lai, Chen-Chu |
|---|---|
| 論文名稱: |
應用人工智慧對泳池安全管理及泳客危險行為之預警系統 Apply artificial intelligence to swimming pool safety management and early warning system of dangerous behaviors of swimmers |
| 指導教授: |
蔡明田
Tsai, Ming-Tien |
| 學位類別: |
碩士 Master |
| 系所名稱: |
工學院 - 工程管理碩士在職專班 Engineering Management Graduate Program(on-the-job class) |
| 論文出版年: | 2024 |
| 畢業學年度: | 112 |
| 語文別: | 中文 |
| 論文頁數: | 47 |
| 中文關鍵詞: | 泳池安全 、人工智慧 、深度學習 、行為辨識 |
| 外文關鍵詞: | artificial intelligence, deep learning, machine vision, behavior recognition |
| 相關次數: | 點閱:101 下載:10 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
游泳池是人們夏季消暑娛樂或運動休閒的地方,同時也存在一定的安全隱患之場所。然而泳池事故時有發生,尤其是泳客危險的行為,例如:溺水、泳池邊跳水、泳池邊追逐嬉戲等行為可能導致事故發生而產生嚴重後果。
本研究提出一種基於人工智慧技術的泳池安全管理系統及透過即時影像和智能演算法,來針對泳客或特定場域裡的人員之行為的監測和預警。首先是泳池安全管理系統架構,此系統由多個模組組成,包括機器視覺之攝影機的部署,用於影像的捕捉將實時影像傳送到後端伺服器處理,影像行為辨識模組透過深度學習演算法與資料庫比對進行分類,如溺水、跳水、水中停留時間過長等,辨識出危險行為,即時發出警報通知救生員或相關人員,消弭事故發生於無形中。
通過在實際游泳池環境中,我們驗證了安全管理系統的即時性和有效性。實驗結果表明此系統能夠準確識別泳客及人員的危險行為並即時發出預警,有效提高了泳池的安全性。
The swimming pool is a place where people go to cool off in the summer for entertainment or sports and leisure. At the same time, it is also a place where there are certain safety hazards. However, swimming pool accidents happen from time to time, especially dangerous behaviors of swimmers, such as drowning, swimming pool diving, pool chasing and playing, etc., which may lead to accidents with serious consequences.
This study proposes a swimming pool safety management system based on artificial intelligence technology and uses real-time images and intelligent algorithms to monitor and warn the behavior of swimmers or people in specific areas. The first is the architecture of the swimming pool safety management system. This system consists of multiple modules, including the deployment of machine vision cameras, which are used to capture images and send real-time images to the back-end server for processing. The image behavior recognition module uses deep learning to The algorithm compares with the database to classify, such as drowning, diving, staying in the water for too long, etc., identifies dangerous behaviors, and immediately sends an alarm to notify lifeguards or related personnel, eliminating accidents from happening invisibly.
We verified the immediacy and effectiveness of the safety management system in an actual swimming pool environment. Experimental results show that this system can accurately identify dangerous behaviors of swimmers and personnel and issue immediate warnings, effectively improving the safety of the swimming pool.
[1] 朱育霆,"應用人工智慧技術於運動監控",吳鳳科技大學電機工程系碩士論文,2023
[2] 羅婕芸,"人工智慧藥品研發與專利法之研究",國立成功大學法律學系碩士論文,2023
[3] 柯俞帆,"人工智慧徵才與組織吸引力之關聯",國立臺灣科技大學管理學院MBA碩士論文,2023
[4] 教育部體育署「游泳池管理規範」
[5] 國家海洋研究院,"智慧化海域環境監測系統建置之研究",研究報告
[6] 唐景章,"應用機器視覺技術於車輛輪胎檢驗系統之研究",國立成功大學工學院工程管理碩專班碩士論文,2022