跳到主要內容

臺灣博碩士論文加值系統

(44.211.26.178) 您好!臺灣時間:2024/06/15 02:05
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:呂金和
研究生(外文):LU, CHIN-HO
論文名稱:以YOLO深度學習模型應用於偵測危險物品標誌車輛
論文名稱(外文):Detection of Vehicles with Dangerous Goods Sign Based on YOLO
指導教授:莊鎮嘉莊鎮嘉引用關係
指導教授(外文):CHUANG, CHEN-CHIA
口試委員:陳松雄蕭志清許駿飛陶金旺
口試委員(外文):CHEN, SONG-SHYONGHSIAO, CHIH-CHINGHSU, CHUN-FEICHIN-WANG TAO
口試日期:2021-07-12
學位類別:碩士
校院名稱:國立宜蘭大學
系所名稱:電機資訊學院碩士在職專班
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2021
畢業學年度:109
語文別:中文
論文頁數:46
中文關鍵詞:YOLO車牌辨識危險物品深度學習
外文關鍵詞:YOLOVehicle recognitionDangerous goodsDeep Learning
相關次數:
  • 被引用被引用:8
  • 點閱點閱:849
  • 評分評分:
  • 下載下載:234
  • 收藏至我的研究室書目清單書目收藏:1
因大貨車車種多且體積龐大,再加上所運送物品之種類及特性複雜,當事故發生時造成危害亦較嚴重,須更長排除及善後復原時間,為確保降低大貨車通行後,對於行車動線及管理所產生之影響,於蘇花改各段端點均設有地磅站,規定所有大貨車均須進入地磅站受檢,並禁止載運危險物品,以及超長、超寬、超高及超重車輛行駛,且受儀控系統管制以避免大貨車連續行駛。
本研究係以YOLOv3物件偵測演算法,搭配Darknet神經網路框架、OpenCV影像處理函式庫及Tesseract光學影像辨識功能,透過介接既有攝影機串流,以實現即時影像物件辨識功能,當有危險物品標誌車輛進入時,將於對應地磅站產生告警並顯示車輛照片及車牌號碼,於實際驗證辨識危險物品標誌準確率達95%以上,可有效產生預告警示效果,維護用路人行車安全。
Due to variety of large trucks and complexity of types and characteristics of goods being transported, the hazards will be more serious when an accident occurs. It might take longer to eliminate accident and recover afterwards. In order to reduce the impact of the traffic line and management after the large truck being allowed to go through Suhuagai, Weigh stations (total of six) has been set up. All large trucks are required to enter weigh station for inspection and carriage of dangerous goods is being prohibited. Vehicles which exceed the length, width, or height limit and overweight are regulated by ramp metering system in order to avoid continuous driving through Suhuagai.
This research is based on the YOLOv3 object detection algorithm, combined with the Darknet neural network framework, OpenCV image processing library and Tesseract optical image recognition function. By interacting with the existing camera stream, a real-time image object recognition function can be achieved. When a vehicle with dangerous goods (DG) sign enters weighbridge, an alarm will be generated with vehicle photo and license plate number showing in displayer inside of weighbridge. The accuracy of identifying the DG sign by actual verification is above 95%. This research seen effectively provide the pre-warning effect in order to maintain the safety of pedestrian.
摘要 I
Abstract II
誌謝詞 III
目錄 IV
表目錄 VI
圖目錄 VII
第一章 緒論 1
1.1 前言 1
1.2 研究動機 3
1.3 研究目的 4
1.4 預期成果 5
第二章 研究背景及相關知識 6
2.1 防救災整體資源分布 6
2.2 防救災演練 7
2.3 交控中心營運管理 9
2.3.1 平時營運 9
2.3.2 緊急事故 10
2.4 通行車種管制 10
2.4.1 載運危險物品車輛通行管制 10
2.4.2 大貨車過磅及儀控管制作業 13
2.5 行車管制與交通配套 14
2.5.1 行車安全距離保持 14
2.5.2 大貨車地磅管制辦法 14
2.5.3 超長車輛處置 15
第三章 研究工具 16
3.1 Labelimg 16
3.2 YOLO 16
3.3 Google Colaboratory 20
3.4 卷積神經網路 21
3.5 Tesseract 22
第四章 偵測危險物品標誌車輛系統 23
4.1 資料蒐集及處理 24
4.2 YOLO訓練 24
4.3 影像介接 31
4.4 危險物品辨識 31
4.5 車牌偵測 32
4.6 影像前處理 37
4.7 光學文字辨識 38
4.8 實驗環境及評估方式 41
4.9 辨識準確率 42
第五章 結論與未來展望 43
5.1 結論 43
5.2 未來展望 43
參考文獻 44
[1]Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi, "You Only Look Once: Unified, Real- Time Object Detection," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 779-788, 2016.
[2]J. Redmon, A. Farhadi, "YOLO9000: Better, Faster, Stronger," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 6517-6525, 07 2017.
[3]J. Redmon, A. Farhadi, "YOLOv3: An Incremental Improvement," arXiv:1804.02767 [cs.CV], 2018.
[4]Cheng-Hsuan Hsiao, “Increasing Recognition Accuracy by Supplementing Dataset with YOLO Labeling,” 國立中興大學, 學術論文. 07, 2020.
[5]CHEN, SHU-XIAN, “A weed identification system based on YOLO,” 私立南臺科技大學, 學術論文. 07, 2020.
[6]Ton-mei Lee, “A study on the assessment of carrying perilous goods to pass highway long tunnel― Ba-Guah-Shan long tunnel as an illustrative example,” 私立逢甲大學, 學術論文. 2007.
[7]LAI, JIAN-LIN, “High-Efficient License Plate Recognition System Based on Deep Learning,” 國立雲林科技大學, 學術論文. 2019.
[8]JHANG,WUN-YI, “Real-time Vehicle and License Plate Recognition Using Deep Learning Networks,” 國立雲林科技大學, 學術論文. 2020.
[9]交通部公路總局蘇花公路改善工程處, “台9線蘇花公路山區路段改善計畫(蘇澳~東澳、南澳~和平、和中~大清水)環境影響說明書,” 交通部公路總局, 2011.
[10]交通部公路總局蘇花公路改善工程處, “台9線蘇花公路山區路段改善計畫隧道事故暨整體防救災應變計畫(第一次修訂),” 交通部公路總局, 2020.
[11]交通部公路總局第四區養護工程處, “台9線蘇花公路山區路段改善計畫開放大貨車通行評估報告書(蘇澳~東澳段),” 交通部公路總局, 2018.
[12]交通部公路總局第四區養護工程處, “台9線蘇花公路山區路段改善計畫開放大貨車通行評估報告書(南澳至和平段、和中至大清水段),” 交通部公路總局, 2020.
[13]" Train YOLO to detect a custom object (online with free GPU)," [Online]. Available: https://pysource.com/2020/04/02/train-yolo-to-detect-a-custom-object-online-with-free-gpu/.
[14]" 採用YOLOv3模型的自助餐菜色辨識結帳系統," [Online]. Available: https://notes.andywu.tw/2020/%E6%8E%A1%E7%94%A8yolov3%E6%A8%A1%E5%9E%8B%E7%9A%84%E8%87%AA%E5%8A%A9%E9%A4%90%E8%8F%9C%E8%89%B2%E8%BE%A8%E8%AD%98%E7%B5%90%E5%B8%B3%E7%B3%BB%E7%B5%B1/.
[15]" yolo---參數解釋之訓練log中各參數," [Online]. Available: https://www.cnblogs.com/carle-09/p/12192231.html.
[16]" 【AI_Column】如何以YOLOv3訓練自己的資料集 ─ 以小蕃茄為例," [Online]. Available: https://makerpro.cc/2019/12/train-your-dataset-with-yolov3/.
[17]" Yolo:基於深度學習的物件偵測 (含YoloV3)," [Online]. Available: https://mropengate.blogspot.com/2018/06/yolo-yolov3.html.
[18]" 雪隧火燒車2死25傷," [Online]. Available: https://tw.appledaily.com/headline/20120508/IL472URNHK4IOEOFN57UUBDDWA/.
[19]"【論文解讀】Yolo三部曲解讀——Yolov3," [Online]. Available: https://zhuanlan.zhihu.com/p/76802514.
[20]"蘇花改通車宣導," [Online]. Available: https://thbu4.thb.gov.tw/page?node=e4d3a824-f304-49f8-82e2-1c2e96e7644e.
[21]"Day03 YOLOv3 (即時物件偵測)," [Online]. Available: https://www.coderbridge.com/series/d4b5a1a1565e4e7a9cd14618ffe6146f/posts/7ac8de3dbb1b441ab1b2788386a3c349.
[22]"Tesseract-OCR 4.00簡介," [Online]. Available: https://www.jianshu.com/p/726d4ece4031.
[23]"車牌偵測與辨識 (by Yolov4/Tesseract)," [Online]. Available: https://github.com/shihyung/Yolov4_car_plate_detection_recognition.
[24]"A comprehensive guide to OCR with Tesseract, OpenCV and Python," [Online]. Available: https://nanonets.com/blog/ocr-with-tesseract/#ocrwithpytesseractandopencv.
[25]"使用 OpenCV 及 Tesseract 進行 OCR 辨識(2)-使用 OpenCV 進行影像前處理, "[Online]. Available: https://medium.com/pivot-the-life/%E4%BD%BF%E7%94%A8-opencv-%E5%8F%8A-tesseract-%E9%80%B2%E8%A1%8C-ocr-%E8%BE%A8%E8%AD%98-2-%E4%BD%BF%E7%94%A8-opencv-%E9%80%B2%E8%A1%8C%E5%BD%B1%E5%83%8F%E5%89%8D%E8%99%95%E7%90%86-cd18ddd4fef0.
[26]"YOLOv4 產業應用心得整理 - 張家銘," [Online]. Available: https://aiacademy.tw/yolo-v4-intro/.
[27]危險物運輸標示, CNS6864 Z5071, 2006.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊