ملخص البحث :

In this research, the well-known encryption algorithms in the encryption Systems, namely DES & RC4 and the advantages and disadvantages of each algorithm are reviewed and evaluated. These two algorithms are combined to produce of new algorithm which is more efficient unscrambling due to the increasing of the level of complexity that make it highly resistance to several attacks. The new algorithm is implemented to show its efficiency in term of time complexity,(ie Breaking the code will be much more complicated than if it would have been occurring through the use of each algorithm individually), this process can be achieved with a very small time difference (approximately neglected in the encryption process). When this algorithm is applied and tested in practice the following result has been obtained: When a block is encrypted using the DES algorithm, the time spent may be (0.000034) milliseconds, but when using the new algorithm to encrypt the same block, the time taken will be about (0.000042) milliseconds. To encrypt 1024 blocks using the DES algorithm, it will take a time of (0.0406), while by using the new algorithm the time taken for encryption is only (0.051). This gives very little increase in time compared to increasing of complexity obtained. Since the new algorithm combining from the two previous, ones allows us to encrypt each block with a key differs from the other one (ie each block is encrypted with a different key depending on the preceding block), making it very difficult to break the code leading to an increase in security and information protection against decoding.

ملخص البحث :

Face detection is a critical function of security (secure witness face in the video) who appear in a scene and are frequently captured by the camera. Recognition of people from their faces in images has recently piqued the scientific community, partly due to application concerns, but also for the difficulty this characterizes for the algorithms of artificial vision. The idea for this research stems from a broad interest in courtroom witness face detection. The goal of this work is to detect and track the face of a witness in court. In this work, a Viola-Jones method is used to extract human faces and then a particular transformation is applied to crop the image. Witness and non-witness images are classified using convolutional neural networks (CNN). The Kanade-Lucas-Tomasi (KLT) algorithm was utilized to track the witness face using trained features. In this model, the two methods were combined in one model to take the advantage of each method in terms of speed and reduce the amount of space required to implement CNN and detection accuracy. After the test, the results of the proposed model showed that it was 99.5% percent accurate when executed in real-time and with adequate lighting.

ملخص البحث :

In this study the identification of tomato diameter based on quality and size in image processing are very useful for the analysis it can be successfully done by measuring the diameter. The use of image processing for identifying the quality and size can be applied not only to tomatoes but also to other fruits such as oranges, apples, melons, etc. and also vegetables with more accuracy. Thus, this will enable the technology to be applied in many products in this case we proposed an away in to measuring the diameter of tomatoes, measuring diameter to the original image of tomatoes and the second one with the grayscale with five methods, Sobel, Perwitt, Roberts, Log and Canny methods to detect the edge of image with these methods, we get more accuracy of measuring the diameters. Farther than we use the value of threshold of each image to get clear image without noise. And the results can affect not only the use or management of the specific lot of products but also the general procedures in order to obtain better thresholds for the stages of the processing to get a good image. Four steps we applied in the program, i.e., pre-processing, segmentation, analysis and presentation of significant results (on the same device) and appling it to Sobel, Perwitt, Roberts, Log and Canny methods to obtain a good accuracy in results. Its also it is important to optimize the processing to get the exact diameter.

ملخص البحث :

Good facial recognition system is a system which can handle variation that arise when making a face image. These variation can include facial expressions, accessories that are used, the level of illumination and direction of image acquisition. Variation will be captured by the virtual lines are made of at least two prototypes in a class. The virtual line will generalize the variation that may occur on the second prototype. Face identification process will be done by finding the shortest distance between the face that will be recognized by all the Variation result of extrapolation and interpolation prototype in each class. Implementation of this method can achieve accuracy rates of more than 90% to the execution time of 0.5 seconds under optimal condition.

ملخص البحث :

Digital video contains potentially very important data that needs high protection, So researchers began competing in how to find ways to preserve data from illegal access, including the encryption that was used to protect the video content (fully or selectively). Where all methods aim to provide the highest levels of protection for the individuals. So the challenges began in designing strong encryption algorithms that are hard to break and also be fast. The goal of this study is to show some of the techniques used in video encryption, with mentioning the pros and cons of each technique to identify these cons and avoid them in future work. From our study and analysis, it was found that most technologies achieve one aspect, either increased accuracy rate or speedup implementation time, especially when implemented in real time, because both negatively affect the other. For example, methods that use lightweight algorithms such as RC4, chacha and others focus on speed, while other algorithms such as RSA, DES... etc focus on the level of security. Therefore we suggest combining the advantages for algorithms by hybridization to building a fast and robust algorithm that can be implemented in real-time.

ملخص البحث :

Over the previous three decades, the area of computer networks has progressed significantly, from traditional static networks to dynamically designed architecture. The primary purpose of software-defined networking (SDN) is to create an open, programmable network. Conventional network devices, such as routers and switches, may make routing decisions and forward packets; however, SDN divides these components into the Data plane and the Control plane by splitting distinct features away. As a result, switches can only forward packets and cannot make routing decisions; the controller makes routing decisions. OpenFlow is the communication interface between the switches and the controller. It is a protocol that allows the controller to identify the network packet’s path across the switches. This project uses the SDN environment to implement the firefly optimization algorithm to determine the shortest path between two nodes in a network. The firefly optimization algorithm was implemented using Ryu control. The results reveal that using the firefly optimization algorithm improves the selected short path between the source and destination.

ملخص البحث :

This Face detection is considering one of the important topics for recognizing human, it is the first step before the face recognition process, it is considered one of the biggest challenges in the field of vision computer. In recent years Many algorithms for detection have appeared, which depend on extracting the features of the human face, and works continue to develop them to this day. This paper aims to make a comparison between two of the most commonly face detection methods, Viola Jones (V_J) and YOLO v3. This comparison is made to determine which of the two algorithms is being most useful when used to detect faces in digital video. These algorithms are used in many applications, including image classification, medical analysis of image, and objects detection in real time (especially in surveillance cameras). Both algorithms are applied to detect faces in the real time video. The experimental results of a sample consists of 20 video frames show that V_J algorithm consumes less time in comparison with YOLO v3 algorithm, but its results are less accurate, unlike the YOLO v3 algorithm, which is slower in detect face with high accurate rate.

ملخص البحث :

In this paper we enhance an algorithm to hide a text into a digital image as steganography technique. This algorithm consists of two major parts, the first, is procedure to hide a text into a digital image, and the second, is procedure to get a hide text from it, further than this algorithm consists of many parts or procedures to enhance this algorithm to complete this process. We use the Least Significant Bit method to hide the message or text, in order to keep the features and characteristics of the original image. By experiment and implement of this algorithm by visual basic 6.0, the algorithm seems work in a butter case by hiding and getting text from a digital image which is used as a carrier of this text, it means that we use two stages in this paper to accomplish it: 1) detecting that steganography has been used, and then 2) by reading the embedded message.

ملخص البحث :

The encryption of digital images plays a very important role in maintaining privacy and security, in order to protect the people, like witnesses. So, an individual image must be encrypted before it is sent over the Internet to the intended recipient. In this paper, images are encrypted using the rabbit algorithm and the chaotic equation represented by Lévy flight for several reasons, including employing beginning seed number, variance of parameters, and unpredictable random walk direction by making the encryption key more robust through the randomization of the initial vector values. After applying the original and improved rabbit encryption algorithms to the images, the results showed that the proposed Rabbit cipher has shown higher robustness against brute force assaults and larger capability of defending against the statistical cracking. Which has been verified and validated through the use of security metrics are PSNR, MSE, and entropy criteria., The encryption of a 256-size image was achieved in just 3 seconds when the improved Rabbit algorithm has been used, so it will be possible to encrypt video frames in real time because it requires a lesser time and having higher security.

ملخص البحث :

With ever-increasing network connectivity, message integrity and authenticity are critical. The primary building component of message integrity is cryptographic hash the functions. Hash functions are utilized and developed in a variety of ways. The purpose of this study is to a propose and discuss the new keyed hash function. The Hash function and the Henon map are used in this suggested technique. For whatever length of input, this technique generates a 128-bit hash code. The function hashes a message with a key so that an intruder who doesn't know the key can't fabricate the hash code, and so it meets a security, authentication, and integrity requirements for the communication in a network. The paper explains the function design an algorithm, as well as its security and implementation details. The simulation findings suggest that a text authentication and forensics algorithm with strong tampering localization ability may be utilized to authenticate and forensics text authenticity and integrity.

ملخص البحث :

Sentiment analysis is used in practically every aspect of human life and has a substantial impact on our actions. There is a vast amount of data that presents people'opinions in numerous domains, such as business and politics, thanks to the expansion and use of online technology. Naïve Bayes (NB) is utilized in this paper to examine opinions by text classification and categorizing them into the relevant category (business, entertainment, politics, sprot, and tech). It examines the impact of combining NB classifiers (Multinomial, Gaussian, Bernoulli, and Complement) in conjunction with feature extraction methods in such of frequency-inverse document frequency (TF-IDF) on an accuracy of classifying BBC news data. Some techniques were used to measure the performance of classifiers such as recall, precision, and F1-score. The outcomes show that the Complement scored the highest in accuracy, with a score of 97.604 percent. The Complement was found to be the best fit.

ملخص البحث :

The edge detection process continued by the edge linking which is presented in this paper can be clear by the following: 1) it appears from the results of the edge linking that the part with bright colour is indicative of the big change in the intensity value of the original image, and 2) the edge detection in the geometric shapes or Morphological shapes are more significant or more contrast rather than ordinary image. The edge linking, the gradient components Gx and Gy of the image are present here to detect an edge linking of geometric figure (morphology shape) and digital image. There are three kinds of discontinuity in digital image; they are point, line, and edge. In practice, the most common method to find the discontinuity of an image is to use the mask that is operated in all of the pixels in the image, we use of the masks of the size of (3* 3). To implement the contrast stretching process in a computer, we present an algorithm to Sobel method to write a program able to 1) open the file containing digital images, 2) display the digital image in the screen, 3) process the edge linking, and 4) display the digital image resulting from the process, and we used image file in BMP format and of the size of (256* 256) pixel with 256 grey level.

ملخص البحث :

A large and comprehensive data base has been built in this system contains all relevant and required information of employees of any constitution in detail. This data base includes vast information such as (thankfullnesses, Penalties, deputing, personal, functional, promotions, absence, vacations and permissions, such that a particular pop-window or menu can be displayed. Each popped up detail will be manipulated or edited independently. The manipulation processes represent procedures of addition, cancelling or updating any records of interest. This system differs to the others by the possibility of preparing of open reports by extraction certain fields from main ones available in the data base and displaying them in the required report, such as a report of employees names, birth dates, ID No., in addition to the typical reports of the system.

ملخص البحث :

Voice recognition is an interesting topic, as many researchers have sought to use different applications and devices in the field of voice recognition. But more advanced solutions use artificial intelligence techniques for their great ability to simulate the human mind. Voice recognition is useful in many aspects, especially in security fields, for example, witness protection. In this paper, a new model is designed to protect the witness and ensure the Safety and security of witnesses in order to ensure their cooperation and testimony. By recognizing his/her voice in real-time, then changing it by applying by applying some filters and manipulating the sampling frequency and time to amplify the sound. The method relied on the salient features of sound, and the Mel-Frequency Cepstral Coefficients is one of the important and effective techniques on feature extraction (MFCC) approach, it was used as a first step in the proposed model. The model then used a convolutional neural network (CNN), due to its high classification and prediction accuracy, to recognize the witness's voice. Then his/her voice was changed as a last step. The proposed model achieved a classification accuracy of about 99% in distinguishing the witness's voice.