Detekcja odcinków i łuków elips w obrazach RGB

• Autorzy: Pietruch R.

Warianty tytułu

EN

Line segments and ellipse arcs detection in RGB images

• Języki publikacji: PL

Abstrakty

PL

W ramach niniejszej pracy wykonano oprogramowanie realizujące wyznaczanie konturów w obrazach RGB. Zintegrowano kilka różnych algorytmów wygładzania obrazu i wyznaczania obrazu krawędziowego, wykorzystując głównie bibliotekę OpenCV. Napisano bibliotekę do aproksymacji wyznaczonych konturów odcinkami i łukami elips. Zweryfikowano jej działanie na przykładowych, sztucznych i naturalnych obrazach.

EN

In this paper the authors present an implementation of edge detection and contours segmentation algorithms. An extraction of pixel chain of image contour and its approximation to line segments or ellipse arcs is shown. The authors used algorithms, most of them implemented in OpenCV, Mean-Shift for image smoothing, gradient morphology filer and border following algorithm for edge extraction, the Douglas-Peucker algorithm for fitting chains into line segments and SVD based algorithm that fits ellipse to the set of 2D points. The performance of proposed methods is evaluated on computer-generated images and on natural images.

Słowa kluczowe

PL

otoczenie robota; rozpoznawanie otoczenia; system wizyjny; system sensoryczny; robotyka

EN

robot environment; ambient recognition; vision system; sensory system; robotics

• Wydawca: Oficyna Wydawnicza Politechniki Warszawskiej
• Czasopismo: Prace Naukowe Politechniki Warszawskiej. Elektronika
• Rocznik: 2014
• Tom: z. 194, t. 2
• Strony: 721--728
• Opis fizyczny: Bibliogr. 19 poz., rys.

Twórcy

autor

Pietruch R.

• Przemysłowy Instytut Automatyki i Pomiarów PIAP, Al. Jerozolimskie 202, 02-486 Warszawa

Bibliografia

• [1] OpenCV 2.4.8.0 documentation, Structural Analysis and Shape Descriptors, http://docs.opencv.Org/2.4/
• [2] G. Bradski, A. Kaehler. Learning OpenCV: Computer Vision with the OpenCV Library. O’Reilly Media 2008.
• [3] C.M. Christoudias, B. Georgescu, R. Meer. Synergism in low level vision. In: Pattern Recognition, 2002. Proceedings. 16th International Conference on. Proceedings, 2002. wolumen 4, s. 150-155.
• [4] David H. Douglas, Thomas K. Peucker. Algorithms for the reduction of the number of points required to represent a digitized line or its caricature. Cartographica: The International Journal for Geographic Information and Geovisualization, Październik, 1973, wolumen 10, numer 2, s. 112-122.
• [5] Jianping Fan et al. Automatic image segmentation by integrating color-edge extraction and seeded region growing. Image Processing, IEEE Transactions on, Oct, 2001, wolumen 10, numer 10, s. 1454-1466.
• [6] Andrew Fitzgibbon, Robert B. Fisher. A buyer’s guide to conic fitting. In: In British Machine Vision Conference. Proceedings, 1995, s. 513-522.
• [7] Walter Gander, Gene H. Golub, Rolf Strebel. Least-squares fitting of circles and ellipses. BIT Numerical Mathematics, 1994, wolumen 34, numer 4, s. 558-578.
• [8] Włodzimierz Kasprzak, Tomasz Kornuta, Cezary Zieliński. A virtual receptor in a robot control framework. In: Recent Advances in Automation, Robotics and Measuring Techniques Red. Roman Szewczyk, Cezary Zieliński, Małgorzata Kaliczyńska, wolumen 267 serii Advances in Intelligent Systems and Computing, s. 399-408. Springer International Publishing 2014.
• [9] Tomasz Kornuta, Maciej Stefańczyk, Włodzimierz Kasprzak. Basic 3d solid recognition in rgb-d images. In: Recent Advances in Automation, Robotics and Measuring Techniques. Proceedings Red. Roman Szewczyk, Cezary Zieliński, Małgorzata Kaliczyńska. Springer, 2014. wolumen 267 serii Advances in Intelligent Systems and Computing (AISC), s. 421-430.
• [10] Viorica Pătrăucean, Pierre Gurdjos, Rafael Grompone von Gioi. A parameterless line segment and elliptical arc detector with enhanced ellipse fitting. In: ECCV (2). Proceedings, 2012, s. 572-585.
• [11] Milan Sonka, Vaclav Hlavac, Roger Boyle. Image Processing, Analysis, and Machine Vision. Thomson-Engineering 2007.
• [12] Satoshi Suzuki, Keiichi Abe. Topological structural analysis of digitized binary images by border following. Computer Vision, Graphics, and Image Processing, 1985, wolumen 30, numer 1, s. 32-46.
• [13] Aarti Talreja, Ekta Udhwani, Koustubh Sinkar. Image Object Extraction using Watershed Transform and Edge REgion Information, Bachelor’s thesis, Department of Electronics and Telecommunication Engineering, Thadomal Shahani Engineering College, University of Mumbai. Praca magisterska, 2010.
• [14] Susanne Wenzel, Wolfgang Forstner. Finding poly-curves of straight line and ellipse segments in images. Raport instytutowy TR-IGG-P-2013-02, 2013.
• [15] LLC David Eberly, Geometric Tools. Distance from a Point to an Ellipse, an Ellipsoid, or a Hyperellipsoid, http://www.geometrictools.com/Documentation/Documentation.html, 2013
• [16] Kanglai Qian. What’s the difference between EDISON and cvPyrMeanShiftFiltering, http://stackoverflow.com/questions/9645713/whats-the-difference-betweenedi son-and-cvpyrmeanshiftfiltering, 2012
• [17] Yoshito Onishi et al, https://github.eom/YCAMInterlab/RAMDanceToolkit, 2013
• [18] V. Pătrăucean, P. Gurdjos, R Grompone von Gioi. ELSD (Ellipse and Line Segment Detector), http://ubee.enseeiht.fr/vision/ELSD/, 2012
• [19] Rafał Pietruch. Multidimensional Signal/lmage Processing Framework, https://github.com/rpietruc/qmediamodeler, 2014