Real Time Embedded Face Recognition using ARM7

Transcription

1 Real Time Embedded Face Recognition using ARM7 B.koteswar rao 1, P. Rama Krishna 2, MA.Wajeed 3 *(Department of ECE, SBCE, Khammam ** (Department of ECE, LITS, Khammam ramakrishnaporandla@gmail.com *** (Department of ECE, SITS, Khammam ( wajeed.mohamad@gmail.com) ABSTRACT Face recognition is a widely used biological recognition technology. In comparison with other identification methods, this type of recognition has direct, friendly and convenient features. The embedded face recognition system is based on LPC ARM 2148 development board, using Windows operating system, detecting face by using HAAR feature, and then recognizing face by using LBP features. Matlab is used as a front end process and keil code runs at back end in ARM 7 Processor. I. INTRODUCTION Face Recognition technology developed in the 80's and it has rapidly developed, especially in the 90's. Automatic face recognition is a widely used biological recognition technology. One of the key challenges of face recognition is finding efficient and discriminative facial appearance descriptors that can counteract large variations in illumination, pose, facial expression, ageing, partial occlusions and other changes [19].In comparison with other identification methods, face recognition has more convenient features. Human can visually identify people by human face. People can be fairly identified even in the very serious visual stimulated situation. The embedded face recognition in this approach is based on LPC ARM 2148 on windows operating systems. Our approach is similar to [14] in that his approach is on ARM 9 using Linux operating system. For Linux based operating systems ARM 9 supports. In our approach we have used windows as operating system. For windows operating systems Microsoft has made several advancements and changes that have made it a much easier to use Operating System, and although arguably it may not be the easiest Operating System, it is still easier than Linux. Because of the large amount of Microsoft Windows users, there is a much larger selection of available software programs, utilities, Windows.Microsoft Windows includes its own help section, has vast amount of available online documentation and help, as well as books on each of the versions of Windows. The main reason of selecting ARM7 is that it supports windows operating systems and our application supports windows. Now the market share of face recognition is just less than the fingerprint recognition and the proportion is increasing, which have broken the situation that fingerprint recognition monopolized the market in the international Page 273

2 biological recognition market. Therefore, it is very important for the research of face recognition technology. This project, which is based on ARM7 and windows operating system, is a portable device and meets the Requirements. In the last years, different problems in face image analysis, such as face detection, burr recognition [1] and facial expression recognition [5] have received very much attention in computer vision research. The problem is more interesting from the viewpoints of basic research aiming to efficient descriptors for facial images and of applications such as surveillance and human-computer interaction [2].A key issue in face analysis are finding efficient de-scriptors for face appearance. Given the low inter-person variation in\ face images, ideal descriptors should be very discriminative. Still, they should be robust to different perturbations and changes such as illumination and pose changes, aging of the subjects, etc. for displaying the image of output for convenient and for more user options purpose code is written in matlab as front end. Because Matlab is a high performance language for technical computing It integrates computation visualization and programming in an easy to use environment. And processing of image and recognition of image is done by using ARM processor that runs at backend. This paper is organized as follows. In section II we overview about some of the prior face recognition related research that led to our research. In section III we present the details of our face recognition system and in section IV about the results and followed by experiment analysis and conclusion. II.RELATED WORK In this section, we briefly describe some other techniques for face recognition. Fei zuo, with Eindhoven [26] proposed Real-time embedded face recognition for smart home in which he proposed a near real-time face recognition system for embedding in consumer applications. Their aim is to design and build a face recognition system that is robust for natural consumer environments and can be executed on low-cost hardware.fa007 is an embedded facial recognition system with the latest version of leading Dual Sensor Algorithm Ver 3.0, which applied for high level access control application. Its classic slope and industrial design is good for market like Government, Civil ID. It is also good for commercial market like Enterprise, Bank, and building automation and so on. Chin-Shu Chang; Teh-Lu Liao; Po-Yun Hsu; Kuo-Kuang Chen proposed human face recognition system using modified PCA algorithm and ARM platform [27]. They used the human face as the main subject for authentication. Because the ARM system provides strong computation ability, various interfaces such as camera, mass storage, etc. Their study includes recognizer algorithm-principal Component Analysis (PCA) is completely accomplished on the ARM system. Finally, an ARM-based human face recognition platform is implemented in this study.vikram kulkarni, k.laxminarashima rao proposed Embedded car security system on face detection [28]. In this proposal embedded car security system, FDS (Face Detection System) is used to detect the face of the driver and compare it with the predefined face. In case any theft to car information is sent to the owner through MMS. So now owner can obtain the image of the thief in his mobile as well as he can trace the location through GPS. The location of the car as well as its speed can be displayed to the owner through SMS. So by using this system, owner can identify the thief image as well as the location of the car. Page 274

3 III. IMPLEMENTATION Figure 2: A facial image divided into 3x3 size. Fig 1: flow chart of embedded face recognition system The first step in this approach is to detect face. Face detection is one of the major part in automatic face detection. To best recognize the face in the given image the system uses face class Harr [3] feature to detect human face. In this thesis the face recognition is based on the LBP algorithm and PCA with nearest neighbor classifier as the core algorithms of the system identification. Face samples are preprocessed by using LBP and PCA to reduce the vector dimension and we call it as reduced image feature. The input image is taken and processed and compared with the reduced image feature. The distance from reduced image face to image face is a measure of face similarity. Convert the local regions and texture descriptors to binary values. Name it as LBP labels. The LBP label contains information about the patterns on a pixellevel, the labels are summed over a small region to produce information on a regional level and the regional histograms are concatenated to build a global description of the face. The regions can be weighted based on the importance of the information they contain. For example, the weighted Chi square distance can be defined as X 2 (A,B) = I (Ai-Bi)2 Ai+Bi Where A, B are two image descriptors. After converting image into the binary format calculate the distance between the given input image and the database image. The distance or similarity metric from image A to image B is D (A, B) = pixels (i, j) of B w (d A ky (i,j) (i; j)) III.1Methodology of main Algorithm In this process first the facial image is divided into local regions and texture descriptors are extracted from each region independently. The descriptors are then concatenated to form a global description of the face. See Fig 2 for an example of a facial image divided into rectangular regions of size 3x3. Here, ky (i,j) is the code value of pixel (i; j) of image B and w( ) is a user -defined function giving the penalty to include for a pixel at the given spatial distance from the nearest matching code in A. Next process is to check for the similarity between images according to the distance among the images. The process of calculating the distance is called kernel Page 275

4 principle component analysis after calculation of distance all the images are sorted according to the distance with the input image and displays an output image which has very less distance. IV.SYSTEM TEST Fig 5: The selected image as shown in figure Fig 3: figure showing uploading the data base image and reading the image as shown in below. Fig 6: output of the matched image shown in figure V. EXPERIMENT ANALYSIS Fig 4: figure showing the selecting image from the database Here in this process front end is written in matlab code in windows operating system and back end is written in keil code. Matlab is used for convenient display of results, and it has more options to select, optimize. The image that to be matched is given in front end and that display is shown in Fig 2.The data that is to be matched is given as another input for comparison. For comparison the images are already stored in data base.select any image for comparison that is shown in Fig 3. And selected image display is shown in Fig 4. Then the comparison is done in hardware where the program is written in keil in Page 276

5 ARM 7 processor the required output whether match found or not found is shown in Fig 5. VI.CONCLUSION Embedded face recognition system based on ARM 7 on windows operating system is proposed and delivered good recognition performances under controlled conditions. Face recognition security system is designed with the combination of embedded and Matlab. By this approach we can provide complete security for safeguarding to offices, houses, banks, etc. In this thesis we proposed our work based on windows operating systems.futher research can be implemented on RTOS and other sophisticated OS. ACKNOWLEDGEMENTS Thanks to the SARADA INSTITUTE OF TECHNOLOGY AND SCIENCE, Khammam for technical support for the proposed system. REFERENCES [1] Recognition of Blurred Faces Using Local Phase Quantization,Timo Ahonen, Esa Rahtu, Ville Ojansivu, Janne Heikkil a Machine Vision Group, University of Oulu, PL 4500, FI Oulun yliopisto, FINLAND, 2008 IEEE. [2] Chellappa R,Wilson C L,Sirohey S.Human and machine recognition of faces:a survey[j]. Digital Object Identifier, 1995, 83(5): Feature and Adaboost [J]. Journal of Optoelectronics y Laser, 2006, 17(8): [6] Cao Lin, Wang Dong-feng, Liu Xiao-jun, Zou Mou-yan. Face Recognition Based on Two- Dimensional Gabor Wavelets [J]. Journal of Electronics & Information Technology, 2006, 28(3): [7] Liu Jing, Zhou Ji-liu. Face recognition based on Gabor features and kernel discriminant analysis AUTHOR S PROFILE: Mr. B.Koteswarao, B.Tech(EC),M.Tech(ES) is currently working as an Assistant Professor, in Department of Electronics & Communication, in Swarna Bharathi College Of Engineering Khammam, A.P, India. His research interests on digital circuits, embedded systems, image processing and signals & systems. He had 7 years of Teaching Experience. [3] Chen Jian-zhou, Li Li, Hong Gang, Su Da-wei. An Approach of Face Detection In Color Images Based on Haar Wavelet [J]. Microcomputer Information, 2005, 21(10-1): [4] P. Viola, M. Jones. Rapid object detection using a Boosted cascade of simple features. In: IEEE Conference on Computer Vision and Pattern Recognition, pp , [5] Zhu Jian-xiang, Su Guang-da, Li Ying-chun. Facial Expression Recognition Based on Gabor Page 277

6 Mr. Rama Krishna P. received his B.Tech Degree in Electronics and communication engineering in 2008 and M.Tech degree in DIGITAL ELECTRONICS AND COMMUNICATION SYSTEMS in 2011 from JNTU, Hyderabad, and A.P-India. He is currently working as an Assistant Professor in the Department of ECE in LAQSHYA INSTITUTE OF TECHNOLOGY AND SCIENCE, Khammam, and A.P-India. His research interests on digital circuits, IMAGE PROCESSING and Communication Systems.He had 3 years of Teaching experience and also A LIFE MEMBER OF INDIAN SOCIETY FOR TECHNICAL EDUCATION Mr.MA.Wajeed, B.Tech(EC),M.T ech(decs) Gold medalist, ISTE MEMBER is currently working as an Assistant Professor, in Department of Electronics & Communication, in SARADA INSTITUTE OF TECHNOLOGY & SCIENCES, Khammam, A.P, India. He was received a GOLD MEDAL in B.Tech (ECE). And Pursuing M.Tech belongs to Digital Electronics &Communication systems.his research interests on digital circuits, embedded systems &Electro magnetic and transmission lines and signals and systems. He had 2 years of Teaching Experience in academic as well as competitive exams (GATE). Page 278