Responsive image
博碩士論文 etd-0804113-151545 詳細資訊
Title page for etd-0804113-151545
論文名稱
Title
掃描式聲納對潛水員之偵測與辨識之研究
Using Scanning Sonar to Detect and Identify Divers
系所名稱
Department
畢業學年期
Year, semester
語文別
Language
學位類別
Degree
頁數
Number of pages
134
研究生
Author
指導教授
Advisor
召集委員
Convenor
口試委員
Advisory Committee
口試日期
Date of Exam
2013-08-15
繳交日期
Date of Submission
2013-09-05
關鍵字
Keywords
掃描式聲納、灰度共現矩陣、水下安全監測、貝氏分類法、潛水員追蹤
Bayesian classification, tracking, underwater surveillance, GLCM, scanning sonar
統計
Statistics
本論文已被瀏覽 5716 次,被下載 0
The thesis/dissertation has been browsed 5716 times, has been downloaded 0 times.
中文摘要
潛水員於水下作業時,岸上人員無法確切掌握作業中潛水員之動向,所得到的水下環境資訊多半仰賴潛水員以口述方式進行記憶性描述。實際上,潛水員驗證之目標物是否符合岸上人員所要求之標的物,則無從得知。水下環境混濁且光線於水中衰減迅速,其有效涵蓋範圍可能僅數公尺,因此光學儀器無法有效偵測水下環境狀態及掌握潛水員的動向。利用聲脈波於水體中穿透性佳且可涵蓋大範圍之特性,使用掃描式聲納系統對水下環境狀態進行偵測,可達到監測作業中潛水員動向之目的,並可根據潛水員作業情況與搜尋目的,調整掃描所用之斜距及頻率,以滿足現場資料收集的需求。
本研究主要以雙園大橋救溺任務及鹽埔漁港潛水員潛水複訓兩個實際案例,進行資料收集與分析。現場救溺案例中(雙園大橋),初期以溺水者搜尋為主,掃描斜距設定為5公尺、頻率採用1MHz進行高解析度影像資料收集;對潛水員活動狀況及活動範圍進行偵測時,掃描斜距分別設定為10公尺、30公尺及50公尺、頻率採用675kHz進行影像資料收集,此設定可滿足潛水員水下安全監測及路徑追蹤之目的。
將收集到之聲納影像,結合以灰度共現矩陣及貝氏分類法為主體之自動辨識系統,對動態目標物進行辨識,可降低人為辨識所需花費的時間與人力以及因人員訓練程度高低所產生的誤差;為提高辨識系統之辨識成功率,採用動態目標物影像強化之方法將靜態背景去除,以達到突顯動態目標物影像之目的,降低辨識系統受到靜態目標物干擾導致誤判的情況發生。動態目標物影像強化即是將原始影像中屬於基底影像的元素去除,此方法能夠有效達到動態與靜態目標物分離的效果。
本研究之具體成果包括:
(1) 經實際量測,應用掃描式聲納系統於斜距50公尺的範圍內,能夠有效進行潛水員的偵測與辨識。
(2) 偵測過程中,對水下環境狀態進行的掃描,可掌握潛水員之活動狀態。
(3) 經過靜態背景去除之影像,應用以灰度共現矩陣與貝氏分類法為主體之自動化辨識系統,能夠有效進行動態目標物之辨識。
(4) 利用區塊成長法取得的目標物定量化資訊,可作為目標物存在之判別、目標物濾定及潛水員追蹤與搜索範圍評估之用途。
本研究現階段採用現場資料收集與後續影像辨識處理兩階段進行。自動化辨識系統若能與現場資料收集系統結合,將能有效達到即時辨識的效果,以及實際應用於入侵者偵測等議題中。
Abstract
When divers conduct underwater operations, supervisors at the surface are often unable to accurately collect behavior information of the divers predominate the tendency. The established information for underwater cases mostly are based on diver’s dictation. In fact, supervisors at the surface have no way of justify whether the target verified by divers is the target under investigation. Due to the existence of turbidity, light attenuates quickly in the water within an effective distance of meters or even cent-meters. Therefore, optical instruments are unable to effectively observe the diver's movements. On the other hand, based on good penetrating and large area coverage characteristics of sound pulse, acoustical instruments such as scanning sonar has the potential in monitoring underwater environment and in detecting divers and their behavior. Based on searching purposes, operation parameters of the scanning sonar can be adjusted to meet the needs of field operations.
This research was conducted by collecting and analyzing data collected at two cases, i.e., drowned person rescue mission at Shuang-Yuan Bridge and divers’ renewal training at Yen-Pu port. The initial purpose at Shuang-Yuan Bridge case was searching for drowned human body. The high image resolution is needed. So the scanning range was set to 5 meters with frequency at 1 MHz. As for detecting and monitoring divers’ activity, the scanning range was set to 10, 30, and 50 meters with frequency at 675 kHz in order to satisfy underwater surveillance and divers’ route tracking.
For economizing manpower and avoiding inaccurate judgments, this study developed an automatic sonar image processing system by using grey level co-occurrence matrix(GLCM) and Bayesian classification to identify mobile target in the sonar imagery. To improve the successfully detection rate of the processing system, a mobile target intensifying method was adopted. In this procedure, both static and mobile target were identified and mobile target was intensified by the removal of the static background.
The accomplishment of this study included:
(1) The scanning sonar used in this research can effectively detect the presence of divers at range setting of 50 meters.
(2) During detection processes, both information regarding underwater environment and divers’ activities were effectively collected.
(3) After a mobile target intensification procedure, the mobile target in the collected sonar imagery can be effectively identified through an automatic procedure dominated by GLCM and Bayesian classification.
(4) Mobile target quantitative information was collected by a region growing procedure. This quantitative information could be used as a criterion in identification of the existence or non-existence of mobile target, in tracking divers’ movement and in quantifying divers’ effective searching area.
At this stage, imagery collection by the scanning sonar and target identification through an automatic procedure were conducted separately. The application potential of this research could be greatly improved by the combination of these two major components together in time sequence. A plausible application is in the field of underwater invader detection and identification.
目次 Table of Contents
謝誌 i
摘要 ii
Abstract iii
目錄 v
表目錄 ix
圖目錄 x
第一章 緒論 1
1-1 前言 1
1-2 研究動機與目的 2
1-3 文獻回顧 3
1-3-1 目標物種類探討 3
1-3-2 聲學儀器應用 4
1-3-3 影像背景去除 6
1-3-4 自動辨識方法探討 6
1-4 論文架構與章節內容 8
第二章 水下探測原理與儀器設備 9
2-1 水下探測技術與聲納基礎工作原理 9
2-2 主動式聲納系統 11
2-2-1 多音束聲納系統 11
2-2-2 機械掃描式單音束聲納系統 12
2-2-3 指向性單音束聲納系統 13
2-3 掃描式聲納系統 14
2-3-1 工作原理 14
2-3-2 掃描式聲納系統之解析度 22
2-3-3 聲脈波對潛水員之有效偵測 26
2-4 目標物偵測準則 27
2-4-1 目標物偵測能力之影響參數 28
2-4-2 影響聲納影像品質之因素 33
第三章 搜救實例展示與初步分析 35
3-1 研究區概況與作業目的 35
3-2 聲學影像辨識基本準則 37
3-3 現場人為辨識結果與分析 39
第四章 影像分析原理 43
4-1 影像前置處理作業 44
4-1-1 聲納影像擷取 44
4-1-2 聲學影像裁剪 45
4-2 影像切割與滑動 46
4-2-1 影像切割 46
4-2-2 非滑動視窗 46
4-2-3 滑動視窗 46
4-2-4 紋理基元尺寸選取 47
4-3 灰度共現矩陣原理 49
4-3-1 影像分析 49
4-3-2 影像像素 50
4-3-3 影像紋理 50
4-3-4 紋理分析 50
4-3-5 灰度共現矩陣 51
4-4 參數選取 53
4-4-1 一階統計量 53
4-4-2 二階統計量 53
4-4-3 特徵函數選取 54
4-4-4 特徵函數性質 55
4-5 影像特徵函數分類 56
4-5-1 貝氏分類法 56
4-5-2 貝氏分類器 57
4-5-3 貝氏分類器設定 57
4-6 目標物類群及區塊 58
4-6-1 目標物類群選取 58
4-6-2 目標物區塊量化統計及邊界描繪 59
第五章 動態目標物自動化辨識與分析 62
5-1 動靜態目標物分離方法 62
5-1-1 基底影像建立 63
5-1-2 動態與靜態目標物分離 65
5-2 自動辨識系統 66
5-2-1 影像目標物存在判斷 66
5-2-2 目標物過濾方法 70
5-3 自動辨識實例分析 71
5-3-1 背景去除影像之辨識成果 72
5-3-2 目標物邊框描繪與目標物過濾 73
5-3-3 潛水員移動位置追蹤 74
5-4 潛水員辨識成效評估 76
第六章 討論、結論與未來研究方向 77
6-1 討論 77
6-1-1 偵測時水體狀況之影響 77
6-1-2 影像分析 80
6-1-3 雙園大橋資料收集與辨識之結果 82
6-1-4 鹽埔漁港資料收集與辨識之結果 83
6-1-5 探測方法 85
6-2 結論 86
6-3 未來研究方向 87
參考文獻 89
附錄一 靜態目標物分離與辨識 92
附錄二 視窗尺度與滑移量 97
附錄三 聲納影像圖集 110
參考文獻 References
1. 田文敏(1996),「側向掃描聲納系統之基本工理」,國立中山大學海洋環境及工程學系海上實習課程教材。
2. 田文敏(1998),「水下靜態目標物之搜尋、辨識、定位與監測」,國際海洋年「海洋海軍科技」研討會;海軍軍官學校,民國八十七年十一月十九日至十一月二十日。
3. 田文敏(2000),「水下靜態目標物之搜尋、辨識與定位」,海下技術季刊第十卷第二期,28頁-36頁。
4. 田文敏(2002),「中華搜救總隊水下探測技術訓練課程教材」,銳昌企業股份有限公司,114頁。
5. 田文敏(2006),「水下目標物搜尋作業原理與要點」,第15屆國防科技學術研討會暨95年度國防科技學術合作計畫成果發表會論文集,第147-154頁。
6. 田楷寅(2010),「水下目標物之偵搜與辨識原理」,第32屆海洋工程研討會論文集,第761-766頁。
7. 田文敏、林奕廷(2008),「溺水人員之偵搜辨識原理」,消防月刊,97年7月,第20-26頁。(內政部消防署 編著)。
8. 田文敏、李依倫、李英璋(2008),「水下靜態目標物搜尋與定位作業實例探討」,第十屆水下技術研討會暨國科會成果發表會論文集。第117-125頁。
9. 鄭勝文、邱逢琛(1997),「水下技術概論」,國立編譯館,398頁。
10. 沈志勇(2005),「水下靜態目標物搜尋、偵測、辨識與定位之研究」,國立中山大學海洋環境及工程研究所碩士論文,137頁。
11. 李英璋(2011),「側掃聲納影像人為目標物自動化辨識分析之研究」,國立中山大學海洋環境及工程研究所碩士論文,184頁。
12. 沈育佳(1999),「都會區空載多譜掃描影像紋理分析之研究」,國立中興大學土木工程學系研究所碩士論文,54頁。
13. 林俊宏(2003),「應用影像增加處理之邊界偵測演算法」,中華技術學院學報,第二十六期,310~317頁。
14. 林榮章(1999),「都會區多解像力遙測影像之紋理分析」,國立中興大學土木工程學系研究所碩士論文,71頁。
15. 林奕廷(2009),「溺水人員之偵搜與辨識原理」,國立中山大學海洋環境及工程學系碩士論文,102頁。
16. 唐德誠(2002),「灰度共現矩陣於多波段多極化SAR影像分類之研究」,國防大學中正理工學院軍事工程研究所碩士論文,114頁。
17. 郭孟維(2008),「應用側掃聲納影像進行海床靜態目標物自動化辨識與分析之研究」,國立中山大學海洋環境及工程學系碩士論文,136頁。
18. 曹士亮(2008),「墾丁現生珊瑚礁自動化分類與描繪之研究」,國立中山大學海洋環境及工程學系碩士論文,117頁。
19. 蔡營寬(2006)「水下靜態目標物之偵搜與辨識原理」,國立中山大學海洋環境及工程學系碩士論文,45頁。
20. 黃偉碩(2005),「利用貝氏分類與因子分析法於半導體製程錯誤偵測與診斷」,中華大學科技管理學系暨科技管理研究所碩士論文,51頁。
21. 鄭勝文、邱逢琛(1997),「水下技術概論」,國立編譯館,398頁。
22. 蕭博元(2005),「水下目標物搜尋作業原理與要點」,國立中山大學海洋環境及工程學系碩士論文,198頁。
23. 劉金源(2001),「水中聲學ー水聲系統之基本操作原理」,國立編譯館,619頁。
24. A.P.M. Michel, K.L. Croff, K.W. McLetchie, J.D. Irish (2002), “A Remote Monitoring System for Open Ocean Aquaculture”, OCEANS '02 MTS/IEEE, vol. 4, pp. 2488 – 2496.
25. Bradley G. Stevens, “Use of Video and Sector Scannign Sonar for Studying Tanner Crab Aggregations”, Report of the National Marine Fisheries Service Workshop on Underwater Video Analysis August 4-6, 2004, pp. 6-7 pages.
26. Bjarke Pedersen and Mark V. Trevorrow (1999), “Continuous monitoring of fish in a shallow channel using a fixed horizontal sonar”, J. Acoust. Soc. Am. 105(6), pp. 3126-3135.
27. Debby L. Burwen, Steve J. Fleischman, James D. Miller (2007), “ Evaluation of a Dual-Frequency Imaging Sonar for Detecting and Estimating the Size of Migrating Salmon”, Fishery Data Series No.07-44.
28. Haralick, R.M., Shanmugam, K. , Dinstenin I. (1973), “Texture Features for Image Classification”, IEEE Transactions on System, Vol-SMC-3, No.6 Nor 1973, pp. 610-621.
29. Imagenex Technology Corp. (2002), “Sonar Theory and Applications”, excerpt from Imagenex model 855 color imaging sonar user’s manual, 8 pages.
30. Imagenex Technology Corp. (2003), “Imagenex model 881A Digital Multi-Frequency Imaging Sonar user’s manual”, 54 pages.
31. Konsberg Mesotech, Ltd. (2002), “Use of Scanning and Side-scan Sonar for Body Recovery”, Technical Notes, V1.1, pp. 11-14 pages.
32. Kongsberg Mesotech, Ltd. (2008), “Sonar for Harbour Surveillance and Intruder Detection”, Technical Notes, V1.1, 20 pages.
33. Mario Zampolli, Finn B. Jensen, Alessandra Tesei (2008), “Review of target strength information for waterside security applications” , NURC Reprint Series.
34. Mazel ,C. (1985), “Side Scan Sonar Training Manual”, Klein Associates, Inc. pp 144.
35. Mark V. Trevorrow (1998), “Salmon and herring school detection in shallow waters using sidescan sonars”, Fisheries Research 35, pp. 5-14 pages.
36. Mark V. Trevorrow and Ross R. Claytor (1998), “Detection of Atlantic herring (Clupea harengus) schools in shallow waters using high-frequency sidescan sonars”, Can. J. Fish. Aquat. Sci. 55, pp. 1419-1429.
37. M.J. Chantler, D.M. Lane, D. Dai, N. Williams (1996), “Detection and tracking of returns in sector-scan sonar image sequences”, IEE Proc.-Radar, Sonar Navig., Vol.143, No. 3, pp.157-162.
38. Nils Olav Handegard and Kresimir Williams (2008), “Automated tracking of fish in trawls using the DIDSON (Dual frequency Identification SONar)”, ICES Journal of Marine Science. 65, pp. 636-644.
39. Paul Baxter (2006), “3D Sonar in Object Detection”, Archive, Vol. 10, No.6
40. R.M. Starr and R.E. Thorne (1998), “Acoustic assessment of squid stocks”, FAO Fisheries Technical Paper No.376 , pp. 181-198 pages
41. Ronald T. Kessel and Reginald D. Hollett (2006), “Underwater Intruder Detection Sonar for Harbour Protection: State of the Art Review and Implications”, NATO Undersea Research Centre.
42. Reed, T.B. and Hussong, D.(1989), “Digital Image Processing Techniques for Enhancement and Classifition of SeaMARCII Side Scan Sonar Imagery”, Journal of Geophysical Research, Vol. 94, no. B6, pp. 7469~7490.
43. Stephane G. Conti (2006), “Acoustical monitoring of fish density, behavvior, and growth rate in a tank”, Aquaculture Vol. 251, pp. 314-323.
44. Thomas Edgar Curtis (2011), “Underwater Sureillance”, United States Patent Application Publication.
45. Vasilis Trygonis, Stratis Georgakarakos, E. John Simmonds (2009), “An operational system for automatic school identification on multibeam sonar echoes”, ICES journal of Marine Science, pp.935-949.
46. W. Tesler, R. Kieser, J.M. Preston (2004), “Use of seabed classification methodology to obtain species information from fisheries acoustic data”, ICES CM.
47. Wikantika, K., Harto, A. B., Tateishi, R., Wihartini, J. Tetuko S. S., Park, J. H.(2000), “An Investigation of Textual Characteristics Associatedwith Spectral Information for Land Use Classification,” InternationalGeoscience and Remote Sensing Symposium, pp. 2915-2917.
48. Weszka, J. S., Dyer C. R., Rosenfeld, A.(1976), “A ComparativeStudy of Texture Measures for Terrain Classification,” IEEETrans. on Syst., Man and Cybern., SMC, 6(4), pp. 269-285.
電子全文 Fulltext
本電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
論文使用權限 Thesis access permission:自定論文開放時間 user define
開放時間 Available:
校內 Campus:永不公開 not available
校外 Off-campus:永不公開 not available

您的 IP(校外) 位址是 3.141.8.247
論文開放下載的時間是 校外不公開

Your IP address is 3.141.8.247
This thesis will be available to you on Indicate off-campus access is not available.

紙本論文 Printed copies
紙本論文的公開資訊在102學年度以後相對較為完整。如果需要查詢101學年度以前的紙本論文公開資訊,請聯繫圖資處紙本論文服務櫃台。如有不便之處敬請見諒。
開放時間 available 永不公開 not available

QR Code