頁籤選單縮合
題名 | 利用深度學習模型Mask R-CNN辨識鋼結構橋梁鏽蝕之研究=Research on Identifying Corrosion of Steel Structure Bridges Using Deep Learning Model Mask R-CNN |
---|---|
作者 | 黎俊成; 高書屏; 王豐良; 林志憲; | 書刊名 | 航測及遙測學刊 |
卷期 | 28:3 2023.09[民112.09] |
頁次 | 頁195-208 |
分類號 | 441.8 |
關鍵詞 | 鋼結構橋梁; 橋梁檢測; 無人機; 鏽蝕; Steel structure bridge; Bridge detection; UAV; Corrosion; Mask R-CNN; |
語文 | 中文(Chinese) |
DOI | 10.6574/JPRS.202309_28(3).0004 |
中文摘要 | 近幾年,鋼結構橋梁在臺灣逐漸盛行,但臺灣為容易發生鏽蝕的環境,因此橋梁的鏽蝕檢測必須投入大量成本進行維護。而本研究利用UAV結合Mask R-CNN進行鏽蝕辨識實驗,而結果表示該方法可以改善傳統橋梁檢測的缺點。而目前臺灣無適合本研究的鏽蝕數據集,故本研究自行建立了鏽蝕數據集;其匯入Mask R-CNN進行訓練,並經過本研究的實驗,揀選出最佳的超參數配置,即「優化器SGD搭配學習率1×10^(-3)」,而模型訓練的評估指標結果為:Recall可達97.1%、Precision可達90.4%、mAP可達91.0%以及mIoU可達89.0%;再經過辨識結果的分析,發現影像中的背景雜訊會影響鏽蝕的辨識。 |
英文摘要 | In recent years, steel structure bridges have become increasingly popular in Taiwan, but Taiwan is prone to corrosion, so a lot of costs must be invested in the maintenance of bridge corrosion detection. In this study, UAV and Mask R-CNN are used to conduct corrosion identification experiments, and the results show that this method can improve the shortcomings of traditional bridge detection. At present, there is no corrosion dataset suitable for this study in Taiwan, so this study has established a corrosion data set by itself; It is imported into Mask R-CNN for training, and through the experiment of this study, the best super-parameter configuration is selected, that is, "optimizer SGD collocation learning rate 1 × 10^(-3) ", and the evaluation index results of model training are: Recall can reach 97.1%, Precision can reach 90.4%, mAP can reach 91.0% and mIoU can reach 89.0%; After the analysis of the identification results, it is found that the background noise in the image will affect the identification of corrosion. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。