整合可見光影像與熱影像重建三維熱紋理模型於橋梁檢測之研究

Research on Integrating Visible Image and Thermal Image to Reconstruct 3D Thermal Texture Model for Bridge Inspection

陳漢霖(Han-Lin Chen) ； 高書屏(Szu-Pyng Kao) ； 王豐良(Feng-Liang Wang) ； 林志憲(Jhih-Sian Lin)

《航測及遙測學刊》 27卷3期 (2022/09) Pp. 165-178

https://doi.org/10.6574/JPRS.202209_27(3).0003

摘要

橋梁具有縮短縣市間距離之重要性，定期之橋梁檢測可以儘早發現劣損，並加以修繕，防止突發性的斷裂及崩塌等嚴重災情。近年來無人機（Unmanned Aerial Vehicle, UAV）應用於橋梁檢測發展迅速，其具備機動性高、時效性高、且可搭載各種儀器使檢測效能提升。本研究中透過無人機搭載數位相機及紅外線熱像儀拍攝橋梁影像，並設計一套影像處理之流程，將校準處理後之熱影像進行影像三維模型重建（Image-Based Modeling, IBM），以利於判識內部劣損位置及三維空間之量測。研究中，於目標區域以打音診斷法獲得實際劣損位置，研究顯示三維模型空間幾何精度達到公厘級別，且於劣損區域之面積測量準確率優於96%。

關鍵字

無人機 ； 熱影像 ； 影像三維模型重建 ； 橋梁檢測

並列摘要

Bridges have the importance of shortening the distance between counties and cities. Regular bridge inspections can detect bad damage as soon as possible, and repair them to prevent serious disasters such as sudden ruptures and collapses. In recent years, the application of Unmanned Aerial Vehicle (UAV) in bridge inspection has developed rapidly. It has high mobility, high timeliness, and can be equipped with various instruments to improve the inspection efficiency. In this study, the bridge image was captured by a drone equipped with a camera system, and a set of image processing procedures were designed to reconstruct a 3D thermal texture model from the calibrated image, so as to facilitate the identification of the bad damage location and the measurement of the 3D space. In the research, the actual location of the defect was obtained by the impact acoustic method in the target area. The results showed that the spatial geometric accuracy of the 3D model reached the millimeter level, and the measurement accuracy of the surface of the inferior area was better than 96%.

並列關鍵字

UAV ； Thermal Image ； Image-Based Modeling ； Bridge Inspection

參考文獻

Dlesk, A., Vach, K., and Pavelka, K., 2021. Transformations in the photogrammetric co-processing of thermal infrared images and RGB images, Sensors, 21(15): 5061.

Ellenberg, A., Kontsos, A., Moon, F.L., and Bartoli, I., 2016. Bridge deck delamination identification from unmanned aerial vehicle infrared imagery, Automation in Construction, 72: 155-165.

林宏昱，2017。整合紅外線熱影像與可見光影像之熱影像三維模型重建，國立中興大學土木工程學系碩士學位論文。[Lin, H.Y., 2017. Integration of infrared thermal image and visible image for 3D thermal model reconstruction, Master Thesis, National Chung Hsing University, Taiwan, ROC. (in Chinese)]

高鵬雅，2020。無人機搭載紅外線熱像儀於混凝土橋梁檢測，國立中興大學土木工程學系碩士學位論文。[Kao, P.Y., 2020. UAV with infrared thermal imager for concrete bridge inspection, Master Thesis, National Chung Hsing University, Taiwan, ROC. (in Chinese)]

張榮鑑，2022。基於擴張空間感知編碼器之異質雙影像語意切割網路設計與 FPGA 晶片實現，國立雲林科技大學電子工程系碩士學位論文。[Chang, J.C., 2022., Heterogeneous image sementation network architecture design and FPGA implementation based on dilated spatial encoder, Master Thesis, National Yunlin University of Science & Technology, Taiwan, ROC. (in Chinese)]