Building a system to identify individuals through their speech recording can find its application in diverse areas, such as telephone shopping, voice mail and security control. However, building such systems is a tricky task because of the vast range of differences in the human voice. Thus, selecting strong features becomes very crucial for the recognition system. Therefore, a speaker recognition system based on new spin-image descriptors (SISR) is proposed in this paper. In the proposed system, circular windows (spins) are extracted from the frequency domain of the spectrogram image of the sound, and then a run length matrix is built for each spin, to work as a base for feature extraction tasks. Five different descriptors are generated from the run length matrix within each spin and the final feature vector is then used to populate a deep belief network for classification purpose. The proposed SISR system is evaluated using the English language Speech Database for Speaker Recognition (ELSDSR) database. The experimental results were achieved with 96.46 accuracy; showing that the proposed SISR system outperforms those reported in the related current research work in terms of recognition accuracy.
To achieve safe security to transfer data from the sender to receiver, cryptography is one way that is used for such purposes. However, to increase the level of data security, DNA as a new term was introduced to cryptography. The DNA can be easily used to store and transfer the data, and it becomes an effective procedure for such aims and used to implement the computation. A new cryptography system is proposed, consisting of two phases: the encryption phase and the decryption phase. The encryption phase includes six steps, starting by converting plaintext to their equivalent ASCII values and converting them to binary values. After that, the binary values are converted to DNA characters and then converted to their equivalent complementary DN
... Show MoreEstimating the semantic similarity between short texts plays an increasingly prominent role in many fields related to text mining and natural language processing applications, especially with the large increase in the volume of textual data that is produced daily. Traditional approaches for calculating the degree of similarity between two texts, based on the words they share, do not perform well with short texts because two similar texts may be written in different terms by employing synonyms. As a result, short texts should be semantically compared. In this paper, a semantic similarity measurement method between texts is presented which combines knowledge-based and corpus-based semantic information to build a semantic network that repre
... Show MoreThe widespread of internet allover the world, in addition to the increasing of the huge number of users that they exchanged important information over it highlights the need for a new methods to protect these important information from intruders' corruption or modification. This paper suggests a new method that ensures that the texts of a given document cannot be modified by the intruders. This method mainly consists of mixture of three steps. The first step which barrows some concepts of "Quran" security system to detect some type of change(s) occur in a given text. Where a key of each paragraph in the text is extracted from a group of letters in that paragraph which occur as multiply of a given prime number. This step cannot detect the ch
... Show MoreThe text has many connotations in the Arabic language, such as vowel points, designation, completion, etc., and the original meaning of the text is to show. The Western text has its owen independent semantic unit .The biblical texts are a mixture of what was reported by the Prophet Moses (peace be upon him) and what the authors described in terms of texts over many centuries.The meaning of the text is guidance and payment, and it is a natural connotation. The religious text for Muslims is divided into peremptory texts that are national proof. The evidence for the meaning of the text is proven by language, and it is not required that the researcher be a jurist. The approach is a factual questionnaire by the researcher according to a speci
... Show MoreIn this paper, the goal of proposed method is to protect data against different types of attacks by unauthorized parties. The basic idea of proposed method is generating a private key from a specific features of digital color image such as color (Red, Green and Blue); the generating process of private key from colors of digital color image performed via the computing process of color frequencies for blue color of an image then computing the maximum frequency of blue color, multiplying it by its number and adding process will performed to produce a generated key. After that the private key is generated, must be converting it into the binary representation form. The generated key is extracted from blue color of keyed image then we selects a c
... Show MoreMany approaches of different complexity already exist to edge detection in
color images. Nevertheless, the question remains of how different are the results
when employing computational costly techniques instead of simple ones. This
paper presents a comparative study on two approaches to color edge detection to
reduce noise in image. The approaches are based on the Sobel operator and the
Laplace operator. Furthermore, an efficient algorithm for implementing the two
operators is presented. The operators have been applied to real images. The results
are presented in this paper. It is shown that the quality of the results increases by
using second derivative operator (Laplace operator). And noise reduced in a good
Image classification is the process of finding common features in images from various classes and applying them to categorize and label them. The main problem of the image classification process is the abundance of images, the high complexity of the data, and the shortage of labeled data, presenting the key obstacles in image classification. The cornerstone of image classification is evaluating the convolutional features retrieved from deep learning models and training them with machine learning classifiers. This study proposes a new approach of “hybrid learning” by combining deep learning with machine learning for image classification based on convolutional feature extraction using the VGG-16 deep learning model and seven class
... Show More