Abstarct

Face recognition is an important and fundamental problem in computer vision, and there have been many attempts to address it. Correlation and high information redundancy in face images result in inefficiencies when such images are used directly for recognition. In this paper, an efficient hybrid approach to face recognition is presented, which combines image compression and neural network (NN) techniques together. The compression is achieved by applying fast discrete cosine transforms (DCTs) to reduce image information redundancy because only a subset of the transform coefficients are necessary to preserve the most important facial features such as hair outline, eyes and mouth. The compressed transform coefficients are used for back propagation NN classification. A high recognition rate can be achieved by using a very small proportion of transform coefficients. This makes DCT-based face recognition much faster than other approaches.
The proposed system is implemented and tested using gray-scale images contains 40
distinct persons, each person having 10 different images and it gives good performance at
high speed.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

This paper describes hybrid access control system based on a new approach of face recognition and PIN (Personal Identification Number) code identification, which is done by finding the closest match between features implemented on the same databases. This approach involves two identification and matching subsystems. The outputs from these subsystems are combined to decide the final output that authorizes the access to the secure building for certain persons while denying the access for the others.
In verification, a biometric identifier is compared to another template through a PIN number of database. Identification involves comparing an existing image against a large image database in order to identify the individual depicted. For this purpose, a camera is set in the door looking at the enter and the captured frames are analyzed. The implementation includes neural networks for PIN code identifications, and analysis starts by selecting the regions of interest in the face images using NN.
A neural network based upright frontal face recognition approach has been presented in this paper. The algorithm of face recognition is implemented using preprocessing of face image and two-stage neural network. The first net finds the eyes region of a person and improving image by histogram equalization. The second uses an image of the area around eyes region to identify the person. Using arbitration between multiple networks with rarely overlap in region images to clean up the results significantly improves the accuracy of the recognition. A feasibility investigation and evaluation for face recognition based solely on new concept is conducted which covers all conditions of human face recognition under varying lighting condition, varying facial expression, and varying pose.
The proposed system structure is implemented with newly proposed methods that allow fast processing and accurate recognition as a software package using C++ and Visual Basic.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

The objective of this paper is to design and construct a solar tracking systembased on a microcontroller. The system design depends on some mathematicalequations to send three signals to drive circuit to change the position of the solarcell by changing the polarity of two motors. These mathematical equations are usedto compute the solar height angle (elevation) and the solar horizon angle (azimuth),whereas the usage of the fixed solar cells does not accomplish the desired object,that means the use of fixed solar cells does not grant a suitable output during a dayand a season, where the sun position differs at the morning to the noon and at thesetting of the sun, this is due to the spherical shape of Earth and to its rotationaround the sun.The solar cell is controlled vertically and horizontally at period equal to onehour, whereas the stored data that denote of the sun position is computed each hourfrom the sunrise to the sunset, the amount of the stored data is different from a dayto another, this variation is produced by the difference of day length during theyear, whereas the amount of computed data at the summer is more than the amountof computed data at the winter. The microcontrollers vouch for processing the dataand issue the commands to actuators to change the orientation of the solar cell. Allobtained results are very acceptable, when the system has tested in certain days. Byusing the microcontroller, the project efficiency is improved, and the cost ofhardware is reduced.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

In this paper, parallel Concatenated Convolutional Codes (PCCCs) is modeled as a
special case of Serial Concatenated Convolutional Code (SCCCs). Consequently, resulting
in Adaptive (parallel/serial) concatenated convolutional code in which the same encoder
and decoder can be used for both types of concatenated convolutional codes. To achieve
this goal some interleaver restrictions are made to modify the classical structure of
Semi-Random interleaver. The decoding stage is based on classical SCCC iterative decoder,
with added interleaver restrictions. The core of decoder structure is a soft-input soft-output
(SISO) a posteriori probability (APP) module. Log-Map is used as (APP) for its superior
performance.
This work also presents some classical structured interleavers such as Block and Circular interleavers, adding to them the presented restrictions for sake of comparison. The resulted performance curves from computer simulation, shows that degradation signal per bit energy to noise ratio (Eb/No) for the proposed decoder as compared to the classical PCCC is no more than 0.45 dB at less than 10-5 bit error rates (BER),over Fading channel in worst case.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

Imperfect transmission of block-coded images often causes lost blocks.These blocks may contain very important information of image. In thispaper, an efficient method of error concealment scheme is proposed forrestoring lost blocks and lines in medical images that are transmitted overerror prone network such as the internet and wireless networks. It adoptedthe idea of data hiding that can be implemented in the DCT domain. Thealgorithm determines the most important information in each block oforiginal image and rotated one by 900 clockwise and embeds this informationinto another block that is not adjacent to it, since adjacent blocks have highprobability to be lost at the same time. Simulation results show that thevisual quality and the PSNR evaluation of a reconstructed image aresignificantly improved using the proposed scheme with respect to othertechniques.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

A nonlinear filtering is an important partof processing and restoring image sequence.The computation complexity of these filteringalgorithms makes them difficult for real-timeprocessing. This paper presents the designand implementation of a pure fuzzy filter,which is effective to remove the impulse noise,and a new method for hardwareimplementation into single chip FPGA isproposed. The construction of the fuzzy filteris simple and depends on a set of fuzzy rulesto well detect and remove noise pulses.Simulation results show that the fuzzy filterexhibits better filtering properties thanstandard median filters. The basic FPGAimplementation of the filter cannot reach thelimit for real time processing with goodefficiency. The proposed fuzzy filter isimplemented by modifying the input datalocations and increase the speed of processingseven times without any increase in the cost.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

With the great development in information technology, the commonly used biometrics have becomevery easy to hack such as face, voice and fingerprint etc. This has led to the discovery of new horizons in thisfield. One of the safest biometrics of today is finger vein- But this technique faces some unique difficulties, the mostcommonly used being that the vein pattern is difficult to remove because finger vein images are always low in quality,significantly hampered the feature extraction and classification stages. For this purpose, professional algorithmsmust be considered with the conventional hardware for capturing finger- vein images using Red Surface MountedDiode ( SMD) led. For capturing images, Canon 750D camera with micro lens is used. For high quality imagesthe integrated micro lens is used, and with some adjustments it can also obtain finger print. Features extractionwas used by a combination of Hierarchical Centroid and Histogram of Gradients. Results were evaluated with KNearest Neighbor and Deep Neural Networks using 6 fold stratified cross validation. Results displayed improvementas compared to three latest benchmarks in this field that used 6- fold validation ( 5- samples for training and onesample for test) for SDUMLA- HMT. The work novelty is owing to the hardware design of the sensor within thefinger- vein recognition system to obtain, simultaneously, highly secured recognition with low computation time,finger vein and finger print at low cost, unlimited users for one device and open source.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

There are two methods to implement LUT up to 7-bit depends on the type of Xilinx chip hardware and the software that can use in design and the code generation. The first method implements LUT as a RAM. This method gives high speed and requires a very high cost.The second method implements LUT as logic gates. This method requires special software and gives a low speed implies. This paper proposed a modification to the second method that will save the speed of the first method and low cost of the second method. It depends on the design of the LUT. Therefore it will not require special software.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

The main contribution of this paper to describes and implements the Castella tracking system (CTS) in high volume Field Programmable Gate Arrays (FPGA) devices, which presents the complete design of an adaptive two-state Kalman tracking filter that is suggested by Castella to track the maneuvering and nonmaneuvering targets using FPGA. The basic design for this system required a very high cost lie out of range of FPGA capacity. This paper will present a novel approach to reduce the cost of this system. The new method depends on the reduction of the width of data bus of the system without reduction the accuracy of the system. However the novel approach will reduce the cost to about 10% from the original cost to implement the system in a single chip FPGA. Finally, two simulation scenarios are also given to illustrate the efficiency of this adaptive filter comparing with the conventional Kalman filter.

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex

Abstarct

This work investigates the effectiveness of Turbo Codes and Orthogonal Frequency Division Multiplexing (OFDM) system, for communication over the flat Rayleigh fading additive white Gaussian noise (AWGN) channel. The Fast Fourier Transform (FFT) is used in the transmitter and receiver to satisfy the orthogonality between sub-carriers and reduce the implementation complexity. The system performance is investigated for various Doppler frequencies for channel. To reduce the interference at receiver, a guard time with cyclic extension is used. The simulation results show that the turbo code orthogonal frequency division multiplexing (TCOFDM) system achieves large coding gain with lower BER and reduced decoding iterations, where the Turbo code improves the performance of OFDM system by a factor of 5-8 dB, therefore offering higher data rate in wireless mobile communications. الخلاصة

Keywords

Export bibliographic record

Choose an application

• Reference Manager
• EndNote
• BibTex