Klasyfikacja konturów znaczników z wykorzystaniem miary zmienności na obrazie z sonaru sektorowego

• Autorzy: Borawski M. , Łatuszyńska A.

Warianty tytułu

EN

Classification of marker's contours using a measure of variability on a sector-scan sonar image

• Języki publikacji: PL

Abstrakty

PL

Podstawowym problemem przy wykonywaniu mozaiki z obrazów z sonaru sektorowego jest właściwe ułożenie obrazów względem siebie. Do tego celu trzeba wykorzystać systemy podwodnego pozycjonowania lub dopasować obrazy względem siebie na podstawie punktów charakterystycznych. Ze względu na to, że punkty charakterystyczne czasami nie występują, można tworzyć je sztucznie poprzez umieszczenie na dnie specjalnych znaczników. W artykule przedstawiono metodę klasyfikacji konturu znacznika w kształcie trójkąta. Przy czym, dzięki wykorzystaniu miary zmienności w klasyfikacji, brany jest pod uwagę stopień zniekształcenia obrazu znacznika.

EN

Image from the sector-scan sonar has a limited range. The most often it is not possible to obtain an image of the whole interesting area in one single sounding. It is necessary to make multiple soundings. To obtain a uniform image of the whole area, the composition of multiple soundings’ images called mosaic, is needed. In creating the mosaic, one problem occurs. It is its positioning in relation to each other. In standard equipment of the sector-scan sonar, the system of positioning in relation to the reference point cannot be found. In general, it is only possible to state the direction to north, on the basis of the embedded compass. The systems of more accurate positioning of sonar in relation to the reference point are very expensive, more costly than the sonar itself. In cheaper solutions, the position is stated on the basis of positioning systems mounted on the ship. Considering the fact that the sector-scan sonar does not have a fixed position in relation to the ship, the positioning of that kind is far from accuracy, to far to perform the automated mosaic creation. In cheap solutions, the automated creation of the mosaic needs the searching of reference points which can be used to "positioning", on images. Prof. Stateczny proposed to use the markers as landmarks. The marker is mounted on the stick which is set into the bottom in shallower waters and lowered on the line in the deeper water. Considering the big distortions of sonar images, the markers should be simple geometric figures, e.g. the equilateral triangle. The problem of finding such a marker is brought to the problem of finding the equilateral triangle on the image. Such a triangle will be, of course, distorted, because the sonar image is the projection from the 3D-space to the 2D-space. However, the distortion of that kind could be easily removed basing on the data attached to the line of sonar image. Much bigger problem is the noise distortion which does not have the fixed form. It depends on many factors. So, in some situations, the same object which resembles the triangle could be classified as the marker, if the noise distortions will be considerable, and not classified when the distortions are small. Such problem can be solved by using the measure of variability in the classification. It will be the main topic of the article.

Słowa kluczowe

PL

informatyka; sonar sektorowy; klasyfikacja jednoklasowa; miara zmienności

EN

computer science; sector-scan sonar; one-class classification; measure of variability

• Wydawca: Wojskowa Akademia Techniczna im. Jarosława Dąbrowskiego
• Czasopismo: Biuletyn Wojskowej Akademii Technicznej
• Rocznik: 2010
• Tom: Vol. 59, nr 2
• Strony: 243--264
• Opis fizyczny: Bibliogr. 39 poz., wykr., tab.

Twórcy

autor

Borawski M.

autor

Łatuszyńska A.

• Zachodniopomorski Uniwersytet Technologiczny w Szczecinie, Wydział Informatyki, 70-210 Szczecin, ul. Żołnierska 49

Bibliografia

• [1] K. Agehed, M. L. Padgett, V. Becanovic, C. Bornich, A. Eide, P. Engman, O. Globoden, T. Lindblad, K. Lodgberg, K. Waldemark, Wreck finding and classifying with a sonar - filter, t. 3728, Workshop on virtual intelligence/dynamic neural networks, Stockholm, 1999, 362-374.
• [2] D. Alexandrou, D. Pantzartzis, Seafloor classification with neural networks, Proc. IEEE Int. OCEANS Conf., Washington, DC, 1990, 18-23.
• [3] V. Barnett, T. Lewis, Outliers in statistical data, Wiley series in probability and mathematical statistics, John Wiley & Sons Ltd., 1978.
• [4] C. Bishop, Neural Networks for Pattern Recognition, Oxford University Press, 1995.
• [5] C. Bishop, Novelty detection and neural network validation, IEEE Proceedings on Vision, Image and Signal Processing, 1994.
• [6] I. N. Bronsztejn, K. A. Siemiendiajew, G. Musiol, H. Mühlig, Nowoczesne Kompendium Matematyki, PWN, Warszawa, 2004.
• [7] G. R. Cochranea, K. D. Lafferty, Use of acoustic classification of sidescan sonar data for mapping benthic habitat in the Northern Channel Islands, California, Continental Shelf Research, 22, 2002, 683-690.
• [8] W. T. Collins, R. A. McConnaughey, Acoustic Classification of the Sea Floor to Address Essential Fish Habitat and Marine Protected Area Requirements, Canadian Hydrographic Conference CHS, Canada, 1998.
• [9] S. Daniel, S. Guillaudeux, Adaptation of a partial shape recognition approach, Proc. IEEE Conf. On Systems, Man, and Cybernetics, 1997, 2157-2162.
• [10] G. J. Dobeck, J. C. Hyland, L. Smedley, Automated detection/classification of sea mines in sonar imagery, Proc. SPIE-Int. Soc. Optics, t. 3079, 1997, 90-110.
• [11] S. Dugelay, C. Graffigne, J. M. Augustin, Deep seafloor characterization with multibeam echosonders by image segmentation using angular acoustic variations, Proc. SPIE, t. 2847, 1996.
• [12] B. Grafulla Gonzalez, Y. Petillot, C. Smith, Detection and Classification of Trawling Marks in Sidescan Sonar, Images Advances in Technology for Underwater Vehicles (ATUV), 2004, 137-146.
• [13] S. Guillaudeux, S. Daniel, Optimization of a sonar image processing chain: A fuzzy rules based expert system approach, Proc. OCEAN, Fort Lauderdale, 1996, 1319-1323.
• [14] J. Harbaugh, D. D. Sternlicht, A. Putney, M. D. Tinkle, E. M. Lavely, Acoustic Modeling for Sea-Mine CAD/CAC Development, Oceans Conference, 2005.
• [15] M. P. Hayes, T. Y. Ho, Height Estimation of a Sonar Towfish from Side-Scan Imagery, Image and Vision Computing New Zealand, 2000, 102-107.
• [16] U. Hoelscher-Hoebing, D Kraus, Unsupervised Image Segmentation and Image Fusion for Multi-Beam/Multi-Aspect Sidescan Sonar Images, OCEANS ’98 Conference Proceedings, t. 1, 1998, 571-576.
• [17] P. V. C. Hough, Machine Analysis of Bubble Chamber Pictures, Proc. Int. Conf. High Energy Accelerators and Instrumentation, 1959.
• [18] N. Japkowicz, Concept-Learning in the absence of counterexamples: an autoassociation-based approach to classification, praca doktorska, New Brunswick Rutgers, The State University of New Jersey, 1999.
• [19] D. MacKay, Bayesian methods for adaptive models, California Institute of Technology, 1992.
• [20] M. Mignotte, C. Collet, P. Pérez, P. Bouthemy, Three-class Markovian segmentation of high-resolution sonar images, Computer Vision and Image Understanding, t. 76, 3, 1999, 191-204.
• [21] M. Moya, M. Koch, L. Hostetler, One-class classifier networks for target recognition applications, International Neural Network Society, 1993.
• [22] J. Paramo, S. Bertrand, H. Villalobos, F. Gerlotto, A three-dimensional approach to school typology using vertical scanning multibeam sonar, Fisheries Research, 84, 2007, 171-179.
• [23] S. W. Perry, Applications of Image Processing to Mine Warfare Sonar, Aeronautical and Maritime Research Laboratory, 2000.
• [24] Y. R. Petillot, S. R. Reed, J. M. Bell, Real time AUV pipeline detection and tracking using side scan sonar and multi-beam echo-sounder, Oceans apos, MTS/IEEE, t. 1, 217-222.
• [25] J. Pociecha, B. Podolec, A. Sokołowski, K. Zając, Metody taksonomiczne w badaniach społeczno-ekonomicznych, PWN, Warszawa, 1988.
• [26] S. Reed, Y. Petillot, J. Bell, An Automatic Approach to the Detection and Extraction of Mine Features in Sidescan Sonar, IEEE Journal of Oceanic Engineering, t. 28, 1, 2003.
• [27] G. Ritter, M. Gallegos, Outliers in statistical pattern recognition and an application to automatic chromosome classification, Pattern Recognition Letters, 1997.
• [28] S. Roberts, W. Penny, Novelty, confidence and errors in connectionist systems, Technical report, Imperial College, London, TR-96-1, 1996.
• [29] S. Roberts, L. Tarassenko, J. Pardey, D. Siegwart, A validation index for artificial neural networks, Proceedings of Int. Conference on Neural Networks and Expert Systems in Medicine and Healthcare, 1994.
• [30] J. Siwabessy, J. Penrose, R. Kloser, D. Fox, Seabed habitat classification, Shallow Survey ’99, International Conference on High Resolution Surveys in Shallow Water, Sydney, 1999.
• [31] D. D. Sternlicht, R. D. Dikeman, D. Lemonds, M. T. Korporaal, A. M. Teranishi, Target Confirmation Architecture for a Buried Object Scanning Sonar, IEEE OCEANS 2003 Conference Proceedings, 2003.
• [32] W. K. Stewart, M. Jiang, M. Marra, A neural network approach to classification of sidescan sonar imagery from a midocean ridge area, IEEE J. Ocean. Eng., t. 19, 2, 1994, 214-224.
• [33] W. K. Stewart, M. Jiang, M. Marra, A hierarchical approach to seafloor classification using neural networks, Proc. IEEE Int. OCEANS Conf., Oct. 1992, 109-113.
• [34] D. M. J. Tax, One-class classification: Concept-learning in the absence of counter-examples, praca doktorska, Technical University of Delft , 2001.
• [35] B. Zerr, E. Maillard, D. Gueriot, Seafloor classification by neural hybrid system, Proc. IEEE Int. OCEANS Conf., 1994, 239-243.
• [36] A. Stateczny, T. Praczyk, Sztuczne sieci neuronowe w rozpoznawaniu obiektów morskich, Gdańskie Towarzystwo Naukowe, Gdańsk, 2002.
• [37] A. Stateczny, Pozyskiwanie geodanych metodami hydroakustycznymi, V Ogólnopolskie Sympozjum Geoinformacyjne „Geoinformatyka dla środowiska i społeczeństwa - badania i zastosowania”, Kraków, 17-19 września 2009.
• [38] A. Stateczny, Problemy komputerowego wyznaczania pozycji okrętu metodami porównawczymi, Zeszyty Naukowe AMW, 107A, 1990.
• [39] A. Stateczny, Problems of comparative plotting of the ship’s position, Reports on Geodesy, 1, 61, 2002.