Description

ABSTRACT

Convolutional Neural Networks (CNNs) have shown a great success within the field of face recognition. In this paper, we propose a robust face recognition method, which is based on Principal Component Analysis (PCA) and CNN. In our method, PCA is employed to reduce the size of data. Afterwards, we use a CNN as a classifier for face recognition. We relatively decrease the number of layers used in the CNN architecture, utilizing the dropout regulation technique. Most importantly, we implement the classification step on the GPU component. Several experiments have been implemented using available well-known databases. The experimental results verify the effectiveness of our approach, which keeps good recognition accuracy. Thus, it expands an important acceleration of the classification face compared to the standard CNN implementation without data reduction. It achieves also lower memory consumption due to the smaller amounts of data processing used through the PCA method. Moreover, our model is intentionally designed such that both its running time and memory requirements are decreased

INTRODUCTION

Face recognition is among the most successful applications of pattern recognition. It is an application capable of identifying people by a digital image or video frame. Face recognition is introduced across different applications like visual surveillance and collecting demographic. Although, researchers have made efforts on this field for many years. However, in the real world, face recognition still remains a complicated task. This is due to several factors such as ; different facial expressions, changes in lighting, masked parts of faces, etc. Thus, more efforts have been made to solve these problems and to implement recognition systems in good conditions. Deep convolutional neural networks have become the first solution for computer vision and object recognition, especially for face recognition. Hence, researchers have conducted the powerful analysis ability for this type of applications. In some cases, the implementation and practical use of CNN becomes difficult by the interaction of some problems. The first problem is linked to the large size of training data. Collecting wide datasets of faces can increase some protection issues of privacy. Therefore, successful face-based CNN applications are often formed by large private datasets. Accordingly, GPUs are used to implement different applications which require a lot of computation in different areas. Among these applications, we have face recognition models which require very high performance. As a result, it is important to effectively implement these methods by GPUs, to speed up the computation and processing process. This paper presents a fast hybrid implementation of a face recognition system, which contains feature extraction and classification blocks. The implementation is capable to obtain a good performance by using the ConvNet classifier based on a parallel computing unit GPU. In addition, the idea of using PCA as a features extractor provides to the optimizing memory consumption and the decrease of elapsed time for the running program. Furthermore, PCA has the characteristic of declining dimension and correlation elimination .

PROPOSED METHOD

A  real time face detection and recognition system is proposed in this project which has the capability to recognize human faces in single as well as multiple face images in a database in real time. Feature extraction process includes both PCA and SURF features. Face detection is performed by using viola jones algorithm. Detected faces are cropped out of the input image to make computation fast. SURF features are extracted from the cropped image. For face matching, putative feature matching is carried out and outliers are removed using M-estimator Sample Consensus (MSAC) algorithm. Single as well as multiple person color images from class persons dataset are used to evaluate the system. Here The proposed method is composed of two main parts: In the first one, we use the PCA method to reduce the dimensions of the input data. In the second one, the implementation of the classifier utilizing ConvNet is done using a GPU processor. The adaptive classifier with the PCA technique presents more effective results in terms of processing time and consumption memory. All of these factors enable obtaining a fast and complete recognition model with a significant recognition.

DEMO VIDEO

BLOCK DIAGRAM

CONCLUSION

In this paper, we have suggested a hybrid model of face recognition, which is based on the PCA technique and the CNN classifier. Therefore, we have used the advantages of PCA to reduce dimensions of processed data as well as parallel implementation. In addition, we have minimized the number of layers in order to decrease the use of calculation blocks. Our experiments have been carried out employing well-known database ORL. An important precision performance has been defined using the PCA-CNN set. Furthermore, the proposed optimization methodology is simple but efficient in terms of execution time and memory usage on the GPU implementation for the model set PCACNN. As future work, a larger dataset will be used and treated with a more complex CNN classifier, which can achieve a reasonable balance between accuracy, time efficiency and memory usage.      

For more Image Processing projects ,Click here