Engineering and Technology Journal

Speaker recognition depends on specific predefined steps. The most important steps are feature extraction and features matching. In addition, the category of the speaker voice features has an impact on the recognition process. The proposed speaker recognition makes use of biometric (voice) attributes to recognize the identity of the speaker. The long-term features were used such that maximum frequency, pitch and zero crossing rate (ZCR). In features matching step, the fuzzy inner product was used between feature vectors to compute the matching value between a claimed speaker voice utterance and test voice utterances. The experiments implemented using (ELSDSR) data set. These experiments showed that the recognition accuracy is 100% when using text dependent speaker recognition.

A human eye is a vital organ responsible for a person's vision. So, the early detection of eye diseases is essential. The objective of this paper deals with diagnosing of seven different external eye diseases that can be recognized by a human eye. These diseases cause problems either in eye pupil, in sclera of eye or in both or in eyelid. Color histogram and texture features extraction techniques with classification technique are used to achieve the goal of diagnosing external eye diseases. Hue Min Max Diff (HMMD) color space is used to extract color histogram and texture features which were fed to Back Propagation Artificial Neural Network (BPANN) for classification. The comparative study states that the features extracted from HMMD color space is better than other features like Histogram of Oriented Gradient (HOG) features and give the same accuracy as features extracted directly from medical expert recorded symptoms. The proposed method is applied on external eye diseases data set consisting of 416 images with an accuracy rate of 85.26315%, which is the major result that was achieved in this study.

Today in the business world, significant loss can happen when the borrowers ignore paying their loans. Convenient credit-risk management represents a necessity for lending institutions. In most times, some persons prefer to late their monthly payments, otherwise, they may face difficulties in the loan payment process to the financial institution. Mainly, most fiscal organizations are considered managed and refined client classification systems, scanning a valid client from invalid ones. This paper produces the data mining idea, specifically the classification technique of data mining and builds a system of data mining process structure. The credit scoring problem will be applied using the Taiwan bank dataset. Besides that, three classification methods are adopted, Naïve Bayesian, Decision Tree (C5.0), and Artificial Neural Network. These classifiers are implemented in the WEKA machine learning application. The results show that the C5.0 algorithm is the best among them, it achieves 0.93 accuracy rates, 0.94 detection rates, 0.96 precision rates, and 0.95 F-Measure which is higher than Naïve Bayesian and Artificial Neural Network; also, the False Positive Rate in C5.0 algorithm achieves 0.1 which is less than Artificial Neural Network and Naïve Bayesian

Speaker Recognition Defined by the process of recognizing a person by his\her voice through specific features that extract from his\her voice signal. An Automatic Speaker recognition (ASP) is a biometric authentication system. In the last decade, many advances in the speaker recognition field have been attained, along with many techniques in feature extraction and modeling phases. In this paper, we present an overview of the most recent works in ASP technology. The study makes an effort to discuss several modeling ASP techniques like Gaussian Mixture Model GMM, Vector Quantization (VQ), and Clustering Algorithms. Also, several feature extraction techniques like Linear Predictive Coding (LPC) and Mel frequency cepstral coefficients (MFCC) are examined. Finally, as a result of this study, we found MFCC and GMM methods could be considered as the most successful techniques in the field of speaker recognition so far.

The concept of cloud computing has recently changed how hardware, software, and information are handled. However, security challenges and credibility requirements have never changed and may have increased. Protecting cloud computing and providing security for its resources and users is one of the critical challenges. As a result, most users are afraid to use their resources, because many security problems must be met. For example, authentication and reliability are major security constraints and must be provided in a cloud computing environment. There is a wide range of authentication protocols in use, but the researcher has recommended the Kerberos protocol to represent and test it in a complex environment such as a mixed cloud environment.
A model has been developed to implement Kerberos authentication in a hybrid cloud computing environment to securely access the cloud computing services provided. This model is represented using the OPNET Modeler 14.5 simulation system. The network efficiency was measured before and after the hacker. Findings presented in this research are supporting the ability of the Kerberos protocol to prevent illegal access to cloud computing services, whether from within the private cloud or the public cloud. While maintaining the efficient performance of the network.

Motion Estimation (ME) is a very important operation in video coding. In order to reduce complexity of computations involved in ME and to increase quality of this process, many Block Matching Motion Estimation (BMME) Algorithms are proposed. The aim of this paper is to compare between these algorithms and find the best one. Seven BMME algorithms are used in this paper. The performance of each algorithm is evaluated for different types of motion to determine the best one of these algorithms. The evaluation is based on search points, and Peak Signal to Noise Ratio (PSNR). The simulation shows that Hexagonal Search is faster than all other Block Matching (BM) algorithms used in this paper regardless the type of video because it requires less number of search points to evaluate motion vectors for the video sequence. It requires 11.2424 average search point (SP) for small motions and 13.9708 for fast motions. It also gives a good quality that is close enough to the quality given by Full Search.

Recent advances in computer vision have allowed wide-ranging applications in every area of life. One such area of application is the classification of fresh products, but the classification of fruits and vegetables has proven to be a complex problem and needs further development. In recent years, various machine learning techniques have been exploited with many methods of describing the different features of fruit and vegetable classification in many real-life applications. Classification of fruits and vegetables presents significant challenges due to similarities between layers and irregular characteristics within the class.Hence , in this work, three feature extractor/ descriptor which are local binary pattern (LBP), gray level co-occurrence matrix (GLCM) and, histogram of oriented gradient(HoG) has been proposed to extract fruite features , the extracted features have been saved in three feature vectors , then desicion tree classifier has been proposed to classify the fruit types. fruits 360 datasets is used in this work, where 70% of the dataset were used in the training phase while 30% of it used in the testing phase. The three proposed feature extruction methods plus the tree classifier have been used to classifying fruits 360 images, results show that the the three feature extraction methods give a promising results , while the HoG method yielded a poerfull results in which the accuracy obtained is 96%.

In this work the desired aim is to study the effect of two various sorts of a foaming agents on the properties of foamed concrete to obtain high quality with a target density is nearly 1600 kg/m3. The standard samples were designed by employing two types of foam agent (FA), the first one is commercially named (EABSSOC foam agent, FA) while the second is the foam of detergent liquid (D) which known (Fairy). The results showed that the FA sample records the lower bulk density compared to the other types. The perfect mix which involved 1wt.% of (D) had higher values of the compressive strength 20.25MPa, 16.32MPa of the curing in water and air respectively and flexural strength (F.S) values were 6.89MPa,4.47MPa of the cured samples in (air, water) for various durations (7,14 and 28) days compared to the samples that contained 1and 0.8wt. % of FA. The obtained compressive strengths were 5.1MPa, 4.3MPa while the flexural strengths were 2.74MPa, 2.9MPa for the samples contained 1wt. %foam agent (FA) after the curing into water and air at the same duration. It is obvious that the addition of foam to the cement mortar paste imparts great characteristics as lightweight with flowability. These properties and others make it suitable for some applications, for example, a decrease of the dead load from the structure, thermal and acoustic insulating and use it in non-structural sections such as a wall.

Key-frame selection plays an important role in facial expression recognition systems. It helps in selecting the most representative frames that capture the different poses of the face. The effect of the number of selected keyframes has been studied in this paper to find its impact on the final accuracy of the emotion recognition system. Dynamic and static information is employed to select the most effective key-frames of the facial video with a short response time. Firstly, the absolute difference between the successive frames is used to reduce the number of frames and select the candidate ones which then contribute to the clustering process. The static-based information of the reduced sets of frames is then given to the fuzzy C-Means algorithm to select the best C-frames. The selected keyframes are then fed to a graph mining-based facial emotion recognition system to select the most effective sub-graphs in the given set of keyframes. Different experiments have been conducted using Surrey Audio-Visual Expressed Emotion (SAVEE) database and the results show that the proposed method can effectively capture the keyframes that give the best accuracy with a mean response time equals to 2.89s.

Tilt correction is an essential step in the license plate recognition system (LPR). The main goal of this article is to provide a review of the various methods that are presented in the literature and used to correct different types of tilt that appear in the digital image of the license plates (LP). This theoretical survey will enable the researchers to have an overview of the available implemented tilt detection and correction algorithms. That’s how this review will simplify for the researchers the choice to determine which of the available rotation correction and detection algorithms to implement while designing their LPR system. This review also simplifies the decision for the researchers to choose whether to combine two or more of the existing algorithms or simply create a new efficient one. This review doesn’t recite the described models in the literature in a hard-narrative tale, but instead, it clarifies how the tilt correction stage is divided based on its initial steps. The steps include: locating the plate corners, finding the tilting angle of the plate, then, correcting its horizontal, vertical, and sheared inclination. For the tilt correction stage, this review clarifies how state-of-the-art literature handled each step individually. As a result, it has been noticed that line fitting, Hough transform, and Randon transform are the most used methods to correct the tilt of a LP.

Robust facial feature extraction is an effective and important process for face recognition and identification system. The facial features should be invariant to scaling, translation, illumination and rotation, several feature extraction techniques may be used to increase the recognition accuracy. This paper inspects three-moment invariants techniques and then determines how is influenced by the variation which may happen to the various shapes of the face (globally and locally) Globally means the whole face shapes and locally means face part's shape (right eye, left eye, mouth, and nose). The proposed technique is tested using CARL database images. The proposal method of the new method that collects the robust features of each method is trained by a feed-forward neural network. The result has been improved and achieved an accuracy of 99.29%.

Recently, age estimates from speech have received growing interest as they are important for many applications like custom call routing, targeted marketing, or user-profiling. In this work, an automatic system to estimate age in short speech utterances without depending on the text is proposed. From each utterance frame, four groups of features are extracted and then 10 statistical functionals are measured for each extracted dimension of the features, to be followed by dimensionality reduction using Linear Discriminant Analysis (LDA). Finally, bidirectional Gated-Recurrent Neural Networks (G- RNNs) are used to predict speaker age. Experiments are conducted on the VoxCeleb1 dataset to show the performance of the proposed system, which is the first attempt to do so for such a system. In gender-dependent system, the Mean Absolute Error (MAE) of the proposed system is 9.25 years, and 10.33 years, the Root Mean Square Error (RMSE) is 13.17 and 13.26, respectively, for female and male speakers. In gender_ independent system, the MAE of the proposed system is 10.96 years, and the RMSE is 15.47. The results show that the proposed system has a good performance on short-duration utterances, taking into consideration the high noise ratio in the VoxCeleb1 dataset.

Five different 3 mol.% yttria-tetragonal zirconia polycrystals/alumina (3Y-TZP/Al2O3) -glass specimens have been fabricated by the glass infiltration method. graphite additives were added to the (3Y-TZP/20Wt.%Al2O3) to obtain porous structure. The ATZ (Alumina Toughened Zirconia) skeleton have prepared by the normal compaction method. The specimen had been infiltrated by the glass (18Wt.% lithium hydroxide, 72 wt. % feldspar, and 10Wt.% nano Titanium oxide) at 1185ᵒC. Results showed that the porosity of (3Y-TZP/Al2O3) composite was increased while bulk density decreased with increasing graphite additives. The elastic modulus of (3Y-TZP/Al2O3)-glass was decreased (159-109GPa) with increasing the amount of glass infiltrated. Hardness was constant for all the specimens in value of (5.66 GPa), while increment of materials fracture toughening was due to the increasing the glass weight fraction, the kIc was lied between (0.53-1. 54MPa.m1/2). The glass infiltrated (3Y-TZP/Al2O3) process was successfully increased the fracture toughness by penetrated different amounts of glass into the (3Y-TZP/Al2O3) structure.

The present work aims to study physical tests such as the water absorption tests of PVA; PVA/corn starch blends at different mass percent (25, 30, 35, 40, and 50%) of corn starch after immersion in distilled water for ten minutes. The blends were also characterized by FTIR analysis, and an optical microscope. Casting method used to prepare samples. The results of water absorption exhibited that the weight losses of the sample increases the starch content rises in the PVA matrix. Also, it was found the highest value of the swelling ratio % at (50% PVA /50% Starch) blend, while minimum values of the swelling ratio % at (75%PVA /25%Starch) blend and pure PVA film.
It was observed from optical microscope that the starch granules disperse well at 25 and 30 wt.% of the starch in the blend films. Nevertheless, clustering could be observed in the polymer blend at 35, 40, and 50 wt.% of starch component. It was shown that porous and spherical voids after the samples were immersed in distilled water.

Recently, face recognition system (FRS) is implemented in different applications including a range of vital services like airports and banking systems for security purposes. Therefore, deployed surveillance systems have been established which led to the urgent need to develop a vital face recognition system. In this work, a new algorithm was proposed for recognition of the face, personal and color images by training the convolutional neural network using the MATLAB program to build a new program for detection of the face, then building a separate program to discover the lips, nose, and eyes, New methods were explored to analyze the main and independent components to improve face detection, which is considered one of the important techniques in this work using neural networks and implementation through the MATLAB program.

With the development of the Internet coupled with expanding the accessibility of multimedia, various copyright issues have resulted. Several researchers have been working on watermarks to provide the security, durability, and a perception of multimedia. This paper a review of some recent works is presented related to video watermarking techniques, this study focuses on the pros and cons of recent video watermarking, areas of application, and the different types and attacks that are standing against these watermarking techniques. The results obtained from this study showed that watermark techniques based on the transfer domain, are more popular than those of the spatial domain.

Consider feature selection is the main in intelligent algorithms and machine learning to select the subset of data to help acquire the optimal solution. Feature selection used an extract the relevance of the data and discarding the irrelevance of the data with increment fast to select it and to reduce the dimensional of dataset. In the past, it used traditional methods, but these methods are slow of fast and accuracy. In modern times, however, it uses the intelligent methods, Genetic algorithm and swarm optimization methods Ant colony, Bees colony, Cuckoo search, Particle optimization, fish algorithm, cat algorithm, Genetic algorithm ...etc. In feature selection because to increment fast, high accuracy and easy to use of user. In this paper survey it used the Some the swarm intelligent method: Ant colony, Bees colony, Cuckoo search, Particle optimization and Genetic algorithm (GA). It done take the related work for each algorithms the swarm intelligent the ideas, dataset and accuracy of the results after that was done to compare the result in the table among the algorithms and learning the better algorithm is discuses in the discussion and why it is better. Finally, it learning who is the advantage and disadvantage for each algorithms of swarm intelligent in feature selection.

Video encrypting is one technique to protect digital videos, it used to avoid unwanted interference and viewing of the transmitted videos. In this paper, a new selective video cryptography algorithm is suggested using light stream algorithm. As it known video size is large in size and it consume time in the encryption process, ChaCha a light encryption algorithm has been used to reduce the encryption time, encryption is done by Xoring frames of video with the key generated from ChaCha algorithm, it produced an acceptable results from robustness point view, but still encryption process consumed time, thus to speed up this process, feature detection operator (FAST) is used to encrypt key points result from FAST operator, in addition key points from this is increased to optimized between speed and robustness of proposed algorithm. In evaluation process, some of measuring quality factors MSE, PSNR, Correlation, NPCR, UACI and entropy are specified for evaluating and comparing between two suggested encryption algorithms which gave good result in encryption process (ChaCha and ChaCha with FAST Enhancement). Experimental results have discovered that the current projected has less encrypting time and better encrypting influence

The effects of varying laser pulse numbers on the fabricated of ZnONPs by pulsed laser ablation in deionized water of Zn-metal are investigated. The Nd: YAG laser at energy 600mJ prepared three samples by change the laser pulse number (100, 150, and 200). The results were collected and examined using an electron scanning microscope, XRD – diffraction, and transmission electron microscope. The result revealed the colloidal spherical shape and the homogeneous composition of the ZnO NPs. The nanoparticles resulted in different concentrations and sized distributions by changing the pulse number of a laser. The average particle size and the mass concentration of particle size increase with an increasing number of laser pulses by fixed the laser energy.

Data centers are the main nerve of the Internet because of its hosting, storage, cloud computing and other services. All these services require a lot of work and resources, such as energy and cooling. The main problem is how to improve the work of data centers through increased resource utilization by using virtual host simulations and exploiting all server resources. In this paper, we have considered memory resources, where Virtual machines were distributed to hosts after comparing the virtual machines with the host from where the memory and putting the virtual machine on the appropriate host, this will reduce the host machines in the data centers and this will improve the performance of the data centers, in terms of power consumption and the number of servers used and cost

Voice over Internet Protocol (VoIP) calls are susceptible to interfere at many points by many attackers, thus encryption considered an important part in keeping VoIP.
In this paper, Encryption VoIP based on Generated Biometric Key for RC4 Algorithm is proposed to encrypt the voice data before transmitting it over the network. The system uses a stream algorithm based on RC4 encryption with the new method of biometrics based Key generation technique. This system has generated complex keys in offline phase which is formed depend on features extracted using Linear Discernment Analysis (LDA) from face images.
The experimental work shows that the proposed system offers secrecy to speech data with voice cipher is unintelligible and the recovered voice has perfect quality with MSR equal to zero and PSNR equal to infinity.

Smart home indicates an application for different technological implementations, it could indicate any system which controls the door lock and several other devices. Facial identification which is an important section to achieve surveillance and safety, especially for handicapped people, can be considered as one of the ways that deal with biometrics and performed to identify facial images via utilizing fundamental features of the face. A Raspberry Pi-based face recognition system using conventional face detection and recognition techniques is going to be supplied, so the method in which image-built biometrics uses a Raspberry Pi is described. The aim of the paper here can be considered as transferring face recognition to a level in which the system can replace the utilizing of RF I-Cards and a password to access any system of security and making the system alive and protect the door from being open by hackers, especially by using the picture of an authorized person, we make the raspberry pi turn off and cannot turn on only by a command from the authorized person's mobile. The result of the presented proposal is a system that uses face recognition by utilizing OpenCV, Raspberry Pi, and it functions on an application of Android, and this system percentage becomes 99.63%. It should be cost-effective, of high performance, secured, and easy to use, which can be used in any smart home application.

Image steganography is the art of hiding data into an image by using the secret key. This paper presents two techniques that combine the most significant bit (MSB) as well as the least significant bit (LSB) based on a color image (24bit for RGB). The presented study proposes a novel method to combine (LSB and MSB) bits based on check MSB values and replace bits from LSB with a secret message. The result of this proposed method that made not affect quality stego -image based on the resulting histogram that shows a match between the cover image and stego- image and more secure because not hidden in all image. The factors were used Mean Square Error (MSE), Compute Payload, in addition to Peak Signal to Noise Ratio (PSNR). The PSNR’s rate is high and MSE is less. The result of this paper when applying on the different image gives high PSNR of 87.141 and less MSE of 0.00012 when inserting message 80 bits and reduction value PSNR of 72.023 and MSE of 0.0040 when inserting message 1200 bits and measure entropy is the same value for cover image and stego –image then this method is more security for the attacker.