Image Retrieval Based on Wavelet Transform and Neural Network Classification

Recuperación de Imágenes sobre la Base de la Transformada Ondeleta y su Clasificación Mediante Redes Neuronales

A. C. Gonzalez–Garcia^1,2, J. H. Sossa–Azuela ², E. M. Felipe–Riveron² and O. Pogrebnyak²

¹ Technologic Institute of Toluca, Electronics and Electrical Engineering Department Av. Instituto Tecnologico w/n Ex Rancho La Virgen, Metepec, Mexico, P.O. 52140
e–mail: alaing@ittoluca.edu.mx

² Computing Research Center, National Polytechnic Institute Av. Juan de Dios Batiz and Miguel Othon de Mendizabal, P.O. 07738, México, D.F.
e–mail: hsossa@cic.ipn.mx ; edgardo@cic.ipn.mx ; olek@cic.ipn.mx

Article received on December 06, 2005; accepted on April 20, 2007

Abstract

The problem of retrieving images from a database is considered. In particular, we retrieve images belonging to one of the following six categories: 1) commercial planes in land, 2) commercial planes in air, 3) war planes in land, 4) war planes in air, 5) small aircraft in land, and 6) small aircraft in the air. During training, a wavelet–based description of each image is first calculated using Daubechies 4–wavelet transformation. The resulting coefficients are used to train a neural network (NN). During classification, test images are treated by the already trained NN. Three different ways to obtain the coefficients of the Daubechies transform were proposed and tested: from the entire image color channels, from the histogram of the biggest circular window inside the image color channels, and from the histograms of the square sub–images in the image color channels of the original image. 120 images were used for training and 240 for testing. The best efficiency of 88% was obtained with the third method.

Key Words: Image Retrieval, Wavelet Transform, Neural Classification.

Resumen.

Se considera el problema de la recuperación de imágenes desde una base de datos de imágenes. En particular, se recuperan las imágenes que pertenecen a una de las siguientes seis categorías: 1) aviones comerciales en la tierra, 2) aviones comerciales en el aire, 3) aviones de guerra en la tierra, 4) aviones de guerra en el aire, 5) avionetas en la tierra, y 6) avionetas en el aire. Primeramente se calcula una descripción con base en la transformada ondeleta de cada imagen mediante la ondeleta Daubechies–4. Los coeficientes resultantes se usan para entrenar una red neuronal. Durante la clasificación, se prueba el sistema con imágenes ya tratadas por la ya entrenada red neuronal. Se propusieron y probaron tres métodos diferentes para obtener los coeficientes de la ondeleta Daubechies–4: desde los canales de color RGB de la imagen completa, desde el histograma de la mayor ventana circular inscrita en los canales de color RGB de la imagen, y desde los histogramas de sub–imágenes cuadradas insertadas en los canales de color RGB de la imagen. Se usaron 120 imágenes para el entrenamiento de la red neuronal y 240 para probar el sistema. La mejor eficiencia de 88% se obtuvo con el tercer método.

Palabras Clave: Recuperación de Imágenes, Transformada Ondeleta, Clasificación Neuronal.

DESCARGAR ARTÍCULO EN FORMATO PDF

Acknowledgements

This work was economically supported by SIP–IPN under grants 20050156, 20060517 and 20071438 and by CONACYT under grant 46805.

References

1. A. Del Bimbo (1999). Visual Information Retrieval, Morgan Kaufmann Publishers.        [ Links ]

2. S. M. Lew (2000). Next–Generation Web Searches for Visual Content, Computer, Information Retrieval, Volume 33, Number 11, Computer Society IEEE. Pp. 46–53.        [ Links ]

3. C. Leung (1997). Visual Information Systems, Springer.        [ Links ]

4. M. Vetterli (2000). Wavelets, Approximation, and Compression, IEEE Signal Processing Magazine. 1:59–73.        [ Links ]

5. I. Daubechies (1988). Orthonormal bases of compactly supported wavelets, Comm. Pure and Applied Mathematics. 4: 909–996.        [ Links ]

6. I. Daubechies (1992). Ten Lectures on Wavelets, CBMS–NSF Lecture Notes 61, SIAM.        [ Links ]

7. Z. Xiong, K. Ramchandran and M. T. Orchard (1997). Space–Frequency Quantization for Wavelet Image Coding, IEEE Transactions on Image Processing, 6(5):        [ Links ]

8. N. Papamarcos, A. E. Atsalakis, and Ch. P. Strouthopoulos (2002). Adaptive Color Reduction, IEEE Transactions on Systems, Man and Cybernetics – Part B: Cybernetics, 32(1):        [ Links ]

9. R. C. González, R. E. Woods and S. L. Eddins (2004). Digital Image Processing Using Matlab, Pearson Prentice Hall.        [ Links ]

10. F. D. Jou, K. Ch. Fan and Y. L. Chang (2004). Efficient matching of large–size histograms, Pattern Recognition Letters, 25:277–286.        [ Links ]

11. S. B. Park, J. W. Lee and S. K. Kim (2004). Content–based image classification using a neural network, Pattern Recognition Letters, 25:287–300.        [ Links ]

12. P. McGuire and G. M. T. D'Eleuterio (2001). Eigenpaxels and a Neural–Network Approach to Image Classification, IEEE Transactions on Neural Networks, 12(3).        [ Links ]

13. A. E. Gasca and A. R. Barandela (1999). Algoritmos de aprendizaje y técnicas estadísticas para el entrenamiento del Perceptrón Multicapa, IV Simposio Iberoamericano de Reconocimiento de Patrones, Cuba. Pp. 456–464.        [ Links ]

14. S. Zhang and E. Salari (2005). Image denoising using a neural network based on non–linear filter in wavelet domain, Acoustics, Speech, and Signal Processing, Proceedings IEEE International Conference ICASSP '05, 18–23 March. 2:989–992.        [ Links ]

15. N. S. Manish, M. Bodruzzaman and M. J. Malkani (1998). Feature Extraction using Wavelet Transform for Neural Network based Image Classification, IEEE Proceedings of the Thirtieth Southeastern Symposium on System Theory, 1:412–416. 1998.        [ Links ]

16. J. Puzicha, Th. Hofmann and J. M. Buhmann (1999). Histogram Clustering for Unsupervised Segmentation and Image Retrieval, Pattern Recognition Letters, 20: 899–909.        [ Links ]

17. M. K. Mandal and T. Aboulnasr (1999). Fast Wavelets Histogram Techniques for Image Indexing, Computer Vision and Understanding, 75(1–2):99–110.        [ Links ]

18. S. Liapis and G. Tziritas (2004). Color and texture image retrieval using chromaticity histograms and wavelet frames, Multimedia, IEEE Transactions on Multimedia (5):676 – 686.        [ Links ]

19. Everest Mathias and Aura Conci (1998). Comparing the Influence of Color Spaces and Metrics in Content–Based Image Retrieval, Anais do X SIBGRAPI (10): 1–8.        [ Links ]

20. Zhiyong Zeng and Lihua Zhou (2006). A Novel Image Retrieval Algorithm Using Wavelet Packet Histogram Techniques, Systems and Control in Aerospace and Astronautics, 1st International Symposium: 1194–1197.        [ Links ]

21. De Bianchi, M.F., Guido, R.C., Nogueira, A.L. and Padovan, P. (2006). A Wavelet–PCA Approach for Content–Based Image Retrieval, System Theory, Proceeding of the Thrity–Eighth Southeastern Symposium (38):425–428.        [ Links ]