Tytuł artykułu
Autorzy
Wybrane pełne teksty z tego czasopisma
Identyfikatory
Warianty tytułu
A method for the automatic image recognition of motor vehicles`number plates
Języki publikacji
Abstrakty
Zaprezentowano metodę rozpoznawania obrazów dyfrakcyjnych na podstawie ekstrakcji cech charakterystycznych. Zastosowano detektory prostokątne autorskiej konstrukcji. Ich wprowadzenie pozwoliło uprościć hybrydowy system automatycznego rozpoznawania obrazów. Ekstrakcję cech charakterystycznych tablic rejestracyjnych uzyskano z fourierowskiej przestrzeni obrazowej. Moduł kwadratu dwuwymiarowego widma obrazu jest próbkowany przez zestaw detektorów i wykorzystywany do trenowania oraz testowania sztucznej sieci neuronowej. Klasyfikację cech zrealizowano w ukrytych warstwach tej sieci. Odpowiedzią sieci jest aktywacja jednego z wyjść, decydująca o przyporządkowaniu obrazu wejściowego do właściwego wzorca. Przedstawiona metoda umożliwia pracę w systemie obliczeń równoległych.
A method of diffraction pattern recognition, based on the of the extraction of characteristic features was described. The conception of original rectangular detectors developed by authors is presented. The application of these aIlowed for a simplification of a hybrid system of automatic image recognition. The extraction of characteristic features of number plates was received on the image Fourier space. The absolute value of the square of the two-dimensional image spectrum is sampled by a set of detectors and is used for training and testing the artificial neural network. The features are cIassified in hidden layers of the network. An activation of one of the outputs gives the response of the network (in an initial layer) and it assigns the input image to the appropriate pattern. The presented method enables us to work in a system of parallel calculations.
Rocznik
Tom
Strony
67--86
Opis fizyczny
Bibliogr. 29 poz., schem., wykr., rys.
Twórcy
Bibliografia
- [1] George N., Wang S., Venable D.L.: Pattern recognition using the ring-wedge detector and neural network software. Proc. of SPIE, 1989, vol. 1134 s. 96-106.
- [2] George N.: Automatic pattern recognition. Automatic Recognition and Control, New York 1991.
- [3] George N., Wang S.: Neural networks applied to diffraction - pattern sampling. Applied Optics, 1994, vol.33, s. 3127-3134.
- [4] Berfanger D.M., George N.: All-digital ring-wedge detector applied to fingerprint recognition. Applied Optics, 1999, 38 (2), s. 357-369.
- [5] Fitz A.P., Green R.J.: Fingerprint classification using a hexagonal fast Fourier transform. Pattern Recognition, 1996, 29, s. 1587-1597.
- [6] Chen Z., Sun Y., Zhan Y., Mu G.: Hybrid optical/digital control using fingerprint identification. Optical Engineering, 1995, 34, s. 834-839.
- [7] Willis A.J., Myers I.A.: A cost-effective fingerprint recognition system for use with low-quality prints and damaged fingerprints. Pattern Recognition, 2001, 34, s. 255-270.
- [8] Soifer V., Kotlyar V., Khonina S., Skidanov R.: Optical-digital methods of fingerprint identification. Optics and Lasers in Engineering, 1998, 29, s. 351-359.
- [9] Ganorota D., Joseph J., Singh K.: Modified geometry of ring-wedge detectors for sampling Fourier transform of fingerprints for classification using neural networks. Optics and Lasers in Engineering, 2004, 42, s. 167-177.
- [10] Berfanger D.M., George N.: All-digital ring-wedge detector applied to image quality assessment. Applied Optics, 2000, 39 (23), s.4080-4097.
- [11] Ganorota D., Joseph J., Singh K.: Neural network based face recognition by using diffraction pattern sampling with a digital ring-wedge detector. Optics Communications, 2002, 202, s. 61-68.
- [12] Kaye P.H., Barton J.E., Hirst E, Clark J.M.: Simultaneous light scattering and intrinsic fluorescence measurement for the classification of airbone particles. Applied Optics, 2000, 39, s. 3738-3745.
- [13] Coston S.D., George N.: Recovery of particle size distribution by inversion of optical transform intensity. Opt. Lett., 1991, 16, s.1918-1920.
- [14] Nebeker B.M., Hirleman E.D.: Light scattering by particles and defects on surfaces: semiconductor wafer inspector. Lecture Notes in Physics, 2000, 534, s. 237-257.
- [15] O'Toole R.K., Stark H.: Comparative study of optical-digital vs. all-digital techniques in textural pattern recognition. Applied Optics, 1980, 19, s. 2496-2506.
- [16] Fares A., Bouzid A., Hamdi M.: Rotation invariance using diffraction pattern sampling in optical pattern recognition. J. of Microwaves and Optoelectronics, 2000, 2 (2), s. 33-39.
- [17] Mait J.N., Athale R., Gracht J.: Evolutionary paths in imaging and recent trends. Optics Express, 2003, 11 (18), s. 2093-2101.
- [18] Casasent D., Song J.: A computer generated hologram for diffraction-pattern sampling. Application of Holography, SPIE, 1985, vol. 532, s. 227-236.
- [19] Cyran K.: Rozpoznawanie obrazów z wykorzystaniem komputerowo generowanych hologramów pełniących funkcję detektora pierścieniowo-klinowego. Rozprawa doktorska, Politechnika Śląska w Gliwicach, Wydział Automatyki, Elektroniki i Informatyki, Gliwice 2000.
- [20] Cyran K.A., Jaroszewicz L.R., Niedziela T.: Neural network based automatic diffraction pattern recognition. Opto-Electronics Reviev, 2001, 9, 3, s. 301-307.
- [21] Cyran K.A. Jaroszewicz L.R., Niedziela T., Merta I.: Concurrent signal processing in optimized hybrid CGH-ANN systems. Optica Applicata, 2001, XXXI, 4, s. 675-689.
- [22] Cyran K.A., Niedziela T., Jaroszewicz L.R.: Grating-based DOVDs in high-speed semantic pattern recognition. SPIE International Technical Group Newsletter, Holography, 2001, 12, 2, s. 10-13.
- [23] Cyran K.A., Niedziela T., Jaroszewicz L.R., Podeszwa T.: Neural classifiers in diffraction image processing. Int. Conf. on Computer Vision and Graphics, Zakopane 2000.
- [24] Cyran K.: Image recognition with a diffractive optical variable device and probabilistic neural network: improvements of feature space. WSEAS Trans. on Information Science & Applications, December 2005, vol. 2, issue 12, s. 2212-2219.
- [25] Cyran K.A.: Integration of classifiers working in discrete and real valued feature space applied in two-way optoelectronic image recognition system. Proc. of Fifth IASTED Int. Conf. Visualization, imaging, and image processing, September 7-9, 2005, Benidorm, Spain, s. 592-597.
- [26] Cyran K.A.: Combining rule based and connectionist approaches in diffraction pattern recognition. Proc. of Artificial Intelligence, vol.2, 2005, 25, s.149-157.
- [27] Cyran K.A.: Comparison of artificial neural networks and rough set based classifiers applied to a hybrid pattern recognition system. Proc. of 5th WSEAS Int. Conf. on Signal Processing, Computational Geometry & Artificial Vision, Malta, September 15-17, 2005, s. 210-215.
- [28] Cyran K.A., Niedziela T.: Automatic recognition of the type of road vehicles with use of optimized ring-wedge detector and neural network (Submitted to Archives of Transport, 2006).
- [29] Cyran K.A., Niedziela T.: Infrared images in automatic recognition of the type of road obstacle in a fog (Submitted to Archives of Transport, 2006).
Typ dokumentu
Bibliografia
Identyfikator YADDA
bwmeta1.element.baztech-article-PWA6-0035-0004