透過您的圖書館登入
IP:3.15.202.214
  • 學位論文

以手持裝置進行鋼材鏽蝕即時影像辨識之系統開發

Development of Smartphone Application for Real-Time Steel Rust Recognition

指導教授 : 陳柏翰

摘要


由於台灣地理環境之因素,橋樑工程相關研究在台灣更顯重要,其中鋼橋樑因結構特性常使用於市區高架以及長垮徑橋樑,因此有不少鋼橋樑維護與檢測相關研究。 過去針對鋼橋樑鏽蝕檢測為人眼進行面積百分比之判斷,近年來相當多國內外研究,提出利用影像處理鋼橋梁鏽蝕檢測,準確率也逐漸提升,但辨識影像多以小畫數圖片居多,並未使用目前常見大小之影像做為研究對象。加上近年智慧型手機普及且進步,不論相機解析度或CPU運算速度都由明顯的提升,若能將鏽蝕影像辨識系統結合智慧型手機,可提升現場工程師或檢測人員之橋樑檢測效率,更可以減少設備相關費用。 鏽蝕辨識系統,透過邊緣檢測之技術,可以降低運算複雜度,提高運算效率減少時間,透過一系列的參數測試及調整,並使用實際鏽蝕影像測試,並將此鏽蝕辨識系統應用於智慧型手機即時檢測。

並列摘要


The island of Taiwan was formed at a complex convergent boundary between the Philippine Sea Plate and the Eurasian Plate. Due to geographical factors in Taiwan, bridge engineering is an important issue. The steel bridge often used in urban areas, attributed to its advantage such as lighter self-weight, easier constructability, so there are many research about its repairing and maintenance. Nowadays, corrosion on steel bridges is identified by human eye, but identifying through visual processing has already been proposed in previous studies. In past studies, they used small size image to recognize rust. In recent years, the popularity rate of smartphone has been raised, and smartphone has been equipped with high-pixel camera and good CPU. If this research can develop an application for smartphone, it can not only improve efficiency but also reduce the equipment costs for bridge engineers. This research develops a steel rust recognition system by using edge detected technology, in order to reduce image processing time and improve efficiency. By parameters testing and adjustment, the research used real rust image to test the system. Finally, research develops a smartphone application for real-time rust recognition.

並列關鍵字

rust recognition real-time edge detection smartphone

參考文獻


[2] 何文平(2012),「應用手持裝置於現地評估-以校舍耐震初步評估為例」,國立交通大學土木工程學碩士班碩士論文。
[3] 楊雅晴(2008),「運用智慧型彩色影像辨識於鋼橋生鏽檢測」,國立臺灣大學工學院土木工程學研究所碩士論文。
[1] 尹昱翰(2014),「鋼構件鏽蝕影片運用雲端辨識系統之初步開發」,國立臺灣大學工學院土木工程學研究所碩士論文。
[9] 黃隆明,陳文福,吳聲傑與吳昌翰(2006),「幾何校正法應用於無人載具空拍影像從事災區調查分析」,中華水土保持學報,37(2),111-124。
[5] 施旻汶(2015),「自動化無人飛行載具輔助擷取」,國立中央大學營建管理研究所碩士論文。

被引用紀錄


陳思愷(2019)。應用全卷積神經網絡:U-Net於鏽蝕語義分割影像辨識〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU201902786
白博升(2017)。結合擴增虛擬實境與即時影像辨識之工程應用---以鋼橋鏽蝕辨識為例〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU201701934

延伸閱讀