Abstract— Face detection and recognition has many applications in a variety of fields such as security system, videoconferencing and identification. Face classification is currently implemented in software. A hardware implementation allows real-time processing, but has higher cost and time to-market. The objective of this work is to implement a classifier based on neural networks MLP (Multi-layer Perceptron) for face detection. The MLP is used to classify face and non-face patterns. The systm is described using C language on a P4 (2.4 Ghz) to extract weight values. Then a Hardware implementation is achieved using VHDL based Methodology. We target Xilinx FPGA as the implementation support. Keywords—Classification, Face Detection, FPGA Hardware description, MLP. I. INTRODUCTION UMAN face detection and recognition is an active area of research spanning several disciplines such as image processing, pattern recognition and computer vision. Face detection and recognition are preliminary steps to a wide range of applications such as personal identity verification, video-surveillance, liptrocking, facial expression extraction, gender classification, advanced human and computer interaction. Most methods are based on neural network approaches, feature extraction, Markov chain, skin color, and others are based on template matching [1]. Pattern localization and classification is the step which is used to classify face and non-face patterns. Many systems dealing with object classification are based on ANN (Artificial Neural Networks). In this paper we are intersted by the design of a ANN algorithm in order to achieve image classification. This paper is organized as follows: In section II, we give an overview over classification for face detection. Description of our model is discussed in Section III. Section IV deals with the training method. Section V presents the hardware Manuscript received Mars 4, 2005. F. Smach is with the GMS Group of ENIS University, Sfax 3000, Tunisia, (corresponding author phone: +216 98 688 512; fax: +216 73 500 278 ; e- mail: smach_fethi@ yahoo.fr). M. Atri, is with the EµE Laboratory of Science Monastir University, Monastir, 5000 Tunisia (e-mail: Mohamed.atri@fsm.rnu.tn). J. Mitéran, is with the LE2I Laboratory of Université de Bourgogne Aile des Sciences de l'Ingénieur BP 47870 21078 Dijon Cedex, France, (e-mail: miteranj@u-bourgogne.fr). M. Abid is with the GMS Group of ENIS University, Sfax 3000, Tunisia, (e-mail: Mohamed.abid@enis.rnu.tn). implementation of our architecture. Section 6 provides the results. Finally, we give some concluding remarks in Section VII. II. CLASSIFICATION FOR FACE DETECTION While numerous methods have been proposed to detect face in a single image of intensity or color images. A related and important problem is how to evaluate the performance of the proposed detection methods [1]. Many recent face detection papers compare the performance of several methods, usually in terms of detection and false alarm rates. It is also worth noticing that many metrics have been adopted to evaluate algorithms, such as learning time, execution time, the number of samples required in training, and the ratio between detection rates and false alarms. In general, detectors can make two types of errors: false negatives in which faces are missed resulting in low detection rates and false positives in which an image is declared to be face. False negative = Faces Actual of Number Total Falses Missed of Number . . . . . . .. False positive = Faces Actual of Number Total Faces Detected y Incorrectl of Number . . . . . . . .. Face detection can be viewed as two-class recognition problem in which an image region is classified as being a “Face” or “nonFace”. Consequently, face detection is one of the few attempts to recognize from images a class of objects for which there is a great deal of within-class variability. Face detection also provide interesting challenges to the underlying pattern classification and learning techniques. The class of face and no face image are decidedly characterized by multimodal distribution function and effective decision boundaries are likely to be nonlinear in the image space. Pattern localization and classification are CPU time intensive being normally implemented in software, however with lower performance than custom implementations. Custom implementation in hardware allows real-time processing, having higher cost and time-to-market than software implementation. Some works [2,3,4] uses ANN for classification, and the system is implemented in software, resulting in a poor performance (10 sec for localization and classification). A similar work is presented in [5], aiming to object localization and classification, and it was also Design of a Neural Networks Classifier for Face Detection F. Smach, M. Atri, J. Mitéran and M. Abid H World Academy of Science, Engineering and Technology 11 2005 124