Nowadays, people's expression on the Internet is no longer limited to text, especially with the rise of the short video boom, leading to the emergence of a large number of modal data such as text, pictures, audio, and video. Compared to single mode data ,the multi-modal data always contains massive information. The mining process of multi-modal information can help computers to better understand human emotional characteristics. However, because the multi-modal data show obvious dynamic time series features, it is necessary to solve the dynamic correlation problem within a single mode and between different modes in the same application scene during the fusion process. To solve this problem, in this paper, a feature extraction framework of

    Developing a solid e-voting system that offers fairness and privacy for users is a challenging objective. This paper is trying to address whether blockchain can be used to build an efficient e-voting system, also, this research has specified four blockchain technologies with their features and limitations. Many papers have been reviewed in a study covered ten years from 2011 to 2020. As a result of the study, the blockchain platform can be a successful public ledger to implement an e-voting system. Four blockchain technologies have been noticed from this study. These are blockchain using smart contracts, blockchain relying on Zcash platform, blockchain programmed from scratch, and blockchain depending on digital signature. Each bl

In this work, we are obviously interested in a general solution for the calculation of the image of a single bar in partially coherent illumination. The solution is based on the theory of Hopkins for the formation of images in optical instruments in which it was shown that for all practical cases, the illumination of the object may be considered as due to a self – luminous source placed at the exit pupil of the condenser , and the diffraction integral describing the intensity distribution in the image of a single bar – as an object with half – width (U0 = 8 ) and circular aperture geometry is viewed , which by suitable choice of the coherence parameters (S=0.25,1.0.4.0) can be fitted to the observed distribution in various types of mi

In modern era, which requires the use of networks in the transmission of data across distances, the transport or storage of such data is required to be safe. The protection methods are developed to ensure data security. New schemes are proposed that merge crypto graphical principles with other systems to enhance information security. Chaos maps are one of interesting systems which are merged with cryptography for better encryption performance. Biometrics is considered an effective element in many access security systems. In this paper, two systems which are fingerprint biometrics and chaos logistic map are combined in the encryption of a text message to produce strong cipher that can withstand many types of attacks. The histogram analysis o

Data steganography is a technique used to hide data, secret message, within another data, cover carrier. It is considered as a part of information security. Audio steganography is a type of data steganography, where the secret message is hidden in audio carrier. This paper proposes an efficient audio steganography method that uses LSB technique. The proposed method enhances steganography performance by exploiting all carrier samples and balancing between hiding capacity and distortion ratio. It suggests an adaptive number of hiding bits for each audio sample depending on the secret message size, the cover carrier size, and the signal to noise ratio (SNR). Comparison results show that the proposed method outperforms state of the art methods

Finger vein recognition and user identification is a relatively recent biometric recognition technology with a broad variety of applications, and biometric authentication is extensively employed in the information age. As one of the most essential authentication technologies available today, finger vein recognition captures our attention owing to its high level of security, dependability, and track record of performance. Embedded convolutional neural networks are based on the early or intermediate fusing of input. In early fusion, pictures are categorized according to their location in the input space. In this study, we employ a highly optimized network and late fusion rather than early fusion to create a Fusion convolutional neural network

An Auto Crop method is used for detection and extraction signature, logo and stamp from the document image. This method improves the performance of security system based on signature, logo and stamp images as well as it is extracted images from the original document image and keeping the content information of cropped images. An Auto Crop method reduces the time cost associated with document contents recognition. This method consists of preprocessing, feature extraction and classification. The HSL color space is used to extract color features from cropped image. The k-Nearest Neighbors (KNN) classifier is used for classification. 

Data steganography is a technique used to hide data, secret message, within another data, cover carrier. It is considered as a part of information security. Audio steganography is a type of data steganography, where the secret message is hidden in audio carrier. This paper proposes an efficient audio steganography method that uses LSB technique. The proposed method enhances steganography performance by exploiting all carrier samples and balancing between hiding capacity and distortion ratio. It suggests an adaptive number of hiding bits for each audio sample depending on the secret message size, the cover carrier size, and the signal to noise ratio (SNR). Comparison results show that the proposed method outperforms state of the art methods

In the present work, an image compression method have been modified by combining The Absolute Moment Block Truncation Coding algorithm (AMBTC) with a VQ-based image coding. At the beginning, the AMBTC algorithm based on Weber's law condition have been used to distinguish low and high detail blocks in the original image. The coder will transmit only mean of low detailed block (i.e. uniform blocks like background) on the channel instate of transmit the two reconstruction mean values and bit map for this block. While the high detail block is coded by the proposed fast encoding algorithm for vector quantized method based on the Triangular Inequality Theorem (TIE), then the coder will transmit the two reconstruction mean values (i.e. H&L)

There are large numbers of weakness in the generated keys of security algorithms. This paper includes a new algorithm to generate key of 5120 bits for a new proposed cryptography algorithm for 10 rounds that combine neural networks and chaos theory (1D logistic map). Two methods of neural networks (NN) are employed as Adaline and Hopfield and the results are combined through several sequential operation. Carefully integrating high quality random number generators from neural networks and chaos theory to obtain suitable key for randomness and complexity.