Majlesi Journal of Multimedia Processing
Volume:3 Issue: 1, Mar 2014

Illumination variation is one of the main factors degrading the performance of face verification systems. Near infrared (NIR) and thermal infrared (TIR) face imaging systems are promising ways towards illumination invariant face verification. In this paper, it is shown that when IR face images cannot be observed, learning the relationship between IR information and the corresponding visible images can provide useful source of complementary information about visible light image data. In particular, Canonical Correlation Analysis (CCA) is used to synthesise the NIR and TIR eigenfaces and fisherfaces from their corresponding visible ones. In this paper, the verification performance of a CCA-based synthesising algorithm is developed first. It is then shown that by fusing the visible and synthesised NIR and TIR information at the score level, the performance of the face authentication system considerably improves. Due to the properties of IR imaging systems, more improvement is seen as the visible images are captured under poor illumination conditions.

This paper presents a new system to simulate a Persian talker robot which receives a phoneme type of Persian text as input via GUI and convert it to the speech and simultaneously an animated face that tries to move its lips like a human. The research has been divided into two parts. In the first part, the designed software receives the input text and creates an audio/video file. In the second part, the audio/video file which created in the first part will be used and the lips on the animated face will be detected by image processing. Then, some major points will be assigned on the lips. These points will be tracked by KLT algorithm and the software will send the changed position to robot’s mouth simulator. The “V-Realm Builder” software in MATLAB supposed as the mouth simulator with six points instead of lips actuators. The MOS assessment has been used for assess the outputs from these two parts. These outputs are 1-moving the lips in the animated face and 2-moving points in the “V-Realm Builder” software. The test results confirm the performance of this system.

This paper briefly reviews the presents notions and ideas associated with the biometric techniques for recognition of users of system. The recent advances of information technology and the increasing requirement for security have led to a rapid development of intelligent personal identification systems based on biometrics. As humans, we all use our natural abilities to recognize people through their voices, faces and other characteristics. Technology advances, particularly in biometrics, are helping to close the gap between human perception and machine recognition. A priority goal of the use of biometrics is to provide identity assurance or the capability to accurately recognize individuals with greater reliability. Biometric recognition or, simply, biometrics refers to an automatic recognition of individuals based on their physiological and/or behavioral characteristics. This paper gives a brief overview of biometric, biometric recognition system, and various biometric recognition methods and their advantages, and disadvantages.

Today, The mortality induced by cardiac diseases has become one of the most important factors of death in the world. Among these, the cardiac arrhythmias are one group of the most important heart diseases. Atrial fibrilation is one of cardiac arrhythmias that in addition to strongly decreasing the quality of patient life,it causes very serious dangers and effects such as cardiac stroke. If the paroxysmal atrial fibrillation is recognised on time, the heart rhythm can be restored by performing medical actions and it’s later effects and also disease sustainability can be prerented. The variation in each system is performed gradually.One of the characteristics of electrocardiogram signal that is changed when the condition of proxysmal auricle fibrilation becomes apparent is regulation. There are much indices to study the regulation of a physiologic time series that one of them is approximate entropy. But because of prolems with approximate entropy, the sample entropy was introduced.the index used here to study the regulation is sample entropy.In this work, at first each of signals is cut into 10 second segments. The noises deviated from base line,high frequency and interfrence of power line are removed from signal. Then the atrial electrical activity was obtained by rejecting ventricle activity. To obtain the pattern of ventricle activity, the method of analysis of main components was used. The main atrial wave was obtained by applying a FIR filter of linear phase.The value of sample entropy of mentioned wave was calculated. Finally, we predicted the paroxysmal atrial fibrillation by using sample entropy.

in this paper, we recommend a method of the signal processing for analyzing EEG. To this end,, the signal using the continuous wavelet transform (CWT) is decomposed into dominant scales and a set of statistical features is extracted from these scales, which shows the distribution of wavelet coefficients. Then, the feature selection sequential forward search (SFS) and sequential backward search (SBS) is used to reduce the dimension of the data. Finally, these features give as input to the Bayes and Naïve Bayes classifier with three kinds of discrete outputs: normal, inter-ictal, and ictal. The results of this study show that the highest performance is related to the Bayes classifier, so that the classification accuracy of this classifier using all the features is %99 and using the selected features by SFS and SBS is %100.

There are several methods such as CT SCAN, BIA, and DEXA for measurement and diagnosis of subcutaneous adipose tissue which most of them are problematic processes and require intricate equipment or because of exposing to infra-red are harmful. Xerography method represents a noninvasive evaluation method for diagnosis of subcutaneous adipose tissue. Using Monte-Carlo method and thermal traits in infrared images, 50 people were studied for diagnosis of subcutaneous adipose tissue. Diagnosing of the subcutaneous adipose tissue for all people with different size is possible by using the above method. Adipose tissue with less volume is observable.