Speeding-up normalized neural networks for face/object detection

• Autorzy: El-Bakry H. M. , Zhao Q.
• Języki publikacji: EN

Abstrakty

EN

Finding an object or a face in an input image is a search problem in the spatial domain. Neural networks have shown good results in detecting a certain face/object in a given image. In this paper, faster neural networks for face/object detection are presented. Such networks are designed based on cross correlation in the frequency domain between the input image and the input weights of neural networks. This approach is developed to reduce the computation steps required by these faster neural networks for the search process. The principles of divide and conquer strategy is applied through image decomposition. Each images is divided into small-size sub- images, and then each of them is tested separately using a single faster neural network. Furthermore, the fasted face/object detection is achieved using parallel processing techniques to test the resulting sub-images simultaneously using the same number of faster neural networks. In contrast to using faster neural networks only, the speed-up ratio is increased with the size of the input image when using faster neural networks and image decomposition. Moreover, the problem of local subimage normalization in the frequency domains is solved. The effect of image normalization on the speed-up ratio for face/object detections discussed. Simulation results show that local subimage normalization through weight normalization is faster than subimage normalization in the spatial domain. The overall speed- up ratio of the detection process is increased as the normalization of weights is carried out off line.

Słowa kluczowe

EN

fast face detection; fast object detection; neural networks; cross-correlation; image normalization; parallel processing

• Wydawca: Institute of Information Technology of the Warsaw University of Life Sciences – SGGW
• Czasopismo: Machine Graphics and Vision
• Rocznik: 2005
• Tom: Vol. 14, No. 1
• Strony: 29--59
• Opis fizyczny: Bibliogr. 43 poz., tab., wykr.

Twórcy

autor

El-Bakry H. M.

• University of Aizu, Aizu Wakamatsu, Japan 965-8580

autor

Zhao Q.

• University of Aizu, Aizu Wakamatsu, Japan 965-8580

Bibliografia

• [1] Cooley J. W., Tukey J. W.: An algorithm for the machine calculation of Complex Fourier series. Math. Comput., 19, 297-301, 1965.
• [2] Klette R., Zamperon : Handbook of Image Processing Operators. John Wiley & Sons ltd, 1996.
• [3] Ben-Yacoub S.: Fast Object Detection using M LP and FFT. Research report IDIA P-RR 11, at IDIAP Research Institute, http://www.idiapch/ ; 1997
• [4] Fasel B.: Fast Multi-Scale Face Detection. Research report IDIAP-Com 98-04, at IDIAP Research Institute, http://www.idiap.ch/ ; 1998.
• [5] Rowley H. A., Baluja S., Kanade T.: Neural network - based face detection. IEEE Trans. on PAMI, 20(1), 23-38, 1998.
• [6] Schneiderman H., Kanade T.: Probabilistic modeling of local appearance and spatial relationships for object recognition. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR ), 45-51, SantaBarbara, CA, 1998.
• [7] Ben-Yacoub S., Fasel B., Luettin J.: Fast face detection using MLP and FFT. Proc. of the Second Int. Conf. on Audio and Video-based Biometric Person Authentication (AVBPA’99), 1999.
• [8] El-Bakry H. M., Abo-elsoud M. A., Kamel M. S.: Fast modular neural networks for human face detection. Proc. of IEEE-INNS-ENNS Int. Joint Conf. on Neural Networks, Como, Italy, 24-27 July 2000, Vol. Ill, 320-324.
• [9] El-Bakry H. M.: Fast iris detection using cooperative modular neural nets. Proc. of the 6th sterowanie zaawansowanych Int. Conf. on Soft Computing, 1-4 Oct. 2000, Japan.
• [10] Feraud R., Bernier O., Viallet J. E., Collobert M.: A fast and accurate face detector for indexation of face images. Proc. of the Fourth IEEE Int. Conf. on Automatic Face and Gesture Recognition, Grenoble, France, 28-30 March, 2000.
• [11] Zhu Y., Schwartz S., Orchard M.: Fast face detection using subspace discriminate wavelet features. Proc. of IEEE Computer Society Int. Conf. on Computer Vision and Pattern Recognition (CVPR’00), South Carolina, June 13-15 2000, vol. 1, 1636-1643.
• [12] El-Bakry H. M.: Automatic human face recognition using modular neural networks. MG&V, vol. 10(1), 47-73, 2001.
• [13] El-Bakry H. M.: Automatic human face recognition using modular neural networks. MG&V, vol. 10(1), 47-73, 2001.
• [14] El-Bakry H. M.: Fast iris detection using cooperative modular neural networks. Proc. of the 5th Int. Conf. on Artificial Neural Nets and Genetic Algorithms, 22-25 April 2001, Sydney, Czech Republic, pp. 201-204.
• [15] El-Bakry H. M.: Fast iris detection using neural nets. Proc. of the 14 th Canadian Conf. on Electrical and Computer Engineering, 13-16 May 2001, Canada, pp.1409-1415.
• [16] El-Bakry H. M.: Human iris detection using fast cooperative modular neural nets. Proc. of INNSIEEE Int. Joint Conf. on Neural Networks, 14-19 July 2001, Washington, DC, USA, 577-582.
• [17] El-Bakry H. M.: Human iris detection for information security using fast neural nets. Proc. of the 5 th World Multi-Conf. on Systemics, Cybernetics and Informatics, 22-25 July 2001, Orlando, Florida, USA.
• [18] El-Bakry H. M.: Human iris detection for personal identification using fast modular neural nets. Proc. of the 2001 Int. Conf. on Mathematics and Engineering Techniques in Medicine and Biological Sciences, 25-28 July 2001, Monte Carlo Resort, Las Vegas, Nevada, USA, 112-118.
• [19] El-Bakry H. M.: Human face detection using fast neural networks and image decomposition. Proc. the fifth Int. Conf. on Knowledge-Based Intelligent Information & Engineering Systems, 6-8 September 2001, Osaka-kyoiku University, Kashiwara City, Japan, 1330-1334.
• [20] El-Bakry H. M.: Fast iris detection for personal verification using modular neural networks. Proc. of the Int. Conf. on Computational Intelligence, 1-3 Oct. 2001, Dortmund, Germany, 269-283.
• [21] El-Bakry H. M.: Fast cooperative modular neural nets for human face detection. Proc. of IEEE Int. Conf. on Image Processing, 7-10 Oct. 2001, Thessaloniki, Greece.
• [22] El-Bakry H. M.: Fast face detection using neural networks and image decomposition. Proc. of the 6th Int. Computer Science Conf., Active Media Technology, Dec. 18-20. 2001, Hong Kong - China, 205-215.
• [23] El-Bakry H. M.: Face detection using fast neural networks and image decomposition. Neurocomputing Journal, 48, 1039-1046, 2002.
• [24] El-Bakry H. M.: Human iris detection using fast cooperative modular neural networks and image decomposition. MG&V, 11(4), 498-512, 2002.
• [25] El-Bakry H. M.: Face detection using fast neural networks and image decomposition. Neurocomputing Journal, 48, 1039-1046, 2002.
• [26] El-Bakry H. M.: Face detection using fast neural networks and image decomposition. Proc. Of INNS-IEEE Int. Joint Conf. on Neural Networks, 14-19 May, 2002, Honolulu, Hawaii, USA.
• [27] Srisuk S., Kurutach W.: A new robust face detection in color images. Proc. of IEEE Computer Society International Conference on Automatic Face and Gesture Recognition, Washington D.C., USA, May 20-21, 2002, 306-311.
• [28] El-Bakry H. M.: Comments on using M LP and FFT for fast object/face detection. Proc. of IEEE IJCNN’03, Portland, Oregon, July, 20-24, 2003, pp. 1284-1288.
• [29] El-Bakry H. M., Zhao Q.: Fast object/detection using neural networks and fast Fourier transform. Int. Journal of Signal Processing, 1(3), 182-187, 2004.
• [30] El-Bakry H. M., Zhao Q.: A modified cross correlation in the frequency domain for fast pattern detection using neural networks. Int. Journal of Signal Processing, 1(3), 188-194, 2004.
• [31] El-Bakry H. M., Zhao Q.: Face detection using fast neural processors and image decomposition. Int. Journal of Computational Intelligence, 1(4), 313-316, 2004.
• [32] El-Bakry H. M., Stoyan H.: Fast neural networks for object/face detection, Proc. of the 30th Anniversary SOFSEM Conference on Current Trends in Theory and Practice of Computer Science, 24-30 January, 2004, Czech Republic.
• [33] El-Bakry H. M., Stoyan H.: Fast neural networks for sub-matrix (object/face) detection. Proc. Of IEEE Int. Symp. on Circuits and Systems, Vancouver, Canada, 23-26 May, 2004.
• [34] El-Bakry H. M.: Fast sub-image detection using neural networks and cross correlation in frequency domain. Proc. of IS 2004: 14th h Annual Canadian Conf. on Intelligent Systems, Ottawa, Ontario, 6-8 June, 2004.
• [35] El-Bakry H. M., Stoyan H.: Fast neural networks for code detection in a stream of sequential data. Proc. of CIC 2004 Int. Conf. on Communications in Computing, Las Vegas, Nevada, USA, 21-24 June, 2004.
• [36] El-Bakry H. M.: Fast neural networks for object/face detection. Proc. of 5 th Int. Symp. on Soft Computing for Industry with Applications of Financial Engineering, June 28 - July 4, 2004, Sevilla, Andalucía, Spain.
• [37] El-Bakry H. M., Stoyan H.: A fast searching algorithm for sub-image (object/face) detection using neural networks. Proc. of the 8 th World Multi-Conference on Systemics, Cybernetics and Informatics, 18-21 July, 2004, Orlando, USA.
• [38] El-Bakry H. M., Stoyan H.: Fast neural networks for code detection in sequential data using neural networks for communication applications. Proc. of the First Int. Conf. on Cybernetics and Information Technologies, Systems and Applications: CITSA 2004, 21-25 July, 2004, Orlando, Florida, USA, Vol. IV, 150-153.
• [39] El-Bakry H. M., Zhao Q.: Fast Complex valued time delay neural networks. Int. Journal of Computational Intelligence, 2(1), 16-26, 2005.
• [40] El-Bakry H. M., Zhao Q.: A fast neural algorithm for serial code detection in a stream of sequential data. Int. Journal of Information Technology, 2(1), 71-90, 2005.
• [41] El-Bakry H. M., Zhao Q.: A new symmetric form for fast sub-matrix (object/face) detection using neural networks and FFT. Int. Journal of Signal Processing, to be published.
• [42] El-Bakry H. M., Zhao Q.: Fast pattern detection using normalized neural networks and cross correlation in the frequency domain. EU RASIP Journal on Applied Signal Processing, to be published, 2005.
• [43] Lewis J .P.: Fast Normalized Cross Correlation. Available from : http://www.idiom.com/~zilla/Papers/nvisionInterface/nip.html ; 2005