A content-based image retrieval (CBIR) is a technique used to retrieve images from an image database. However, the CBIR process suffers from less accuracy to retrieve images from an extensive image database and ensure the privacy of images. This paper aims to address the issues of accuracy utilizing deep learning techniques as the CNN method. Also, it provides the necessary privacy for images using fully homomorphic encryption methods by Cheon, Kim, Kim, and Song (CKKS). To achieve these aims, a system has been proposed, namely RCNN_CKKS, that includes two parts. The first part (offline processing) extracts automated high-level features based on a flatting layer in a convolutional neural network (CNN) and then stores these features in a new dataset. In the second part (online processing), the client sends the encrypted image to the server, which depends on the CNN model trained to extract features of the sent image. Next, the extracted features are compared with the stored features using a Hamming distance method to retrieve all similar images. Finally, the server encrypts all retrieved images and sends them to the client. Deep-learning results on plain images were 97.94% for classification and 98.94% for retriever images. At the same time, the NIST test was used to check the security of CKKS when applied to Canadian Institute for Advanced Research (CIFAR-10) dataset. Through these results, researchers conclude that deep learning is an effective method for image retrieval and that a CKKS method is appropriate for image privacy protection.
This paper presents a new design of a nonlinear multi-input multi-output PID neural controller of the active brake steering force and the active front steering angle for a 2-DOF vehicle model based on modified Elman recurrent neural. The goal of this work is to achieve the stability and to improve the vehicle dynamic’s performance through achieving the desired yaw rate and reducing the lateral velocity of the vehicle in a minimum time period for preventing the vehicle from slipping out the road curvature by using two active control actions: the front steering angle and the brake steering force. Bacterial forging optimization algorithm is used to adjust the parameters weights of the proposed controller. Simulation resul
... Show MoreIn this study, dynamic encryption techniques are explored as an image cipher method to generate S-boxes similar to AES S-boxes with the help of a private key belonging to the user and enable images to be encrypted or decrypted using S-boxes. This study consists of two stages: the dynamic generation of the S-box method and the encryption-decryption method. S-boxes should have a non-linear structure, and for this reason, K/DSA (Knutt Durstenfeld Shuffle Algorithm), which is one of the pseudo-random techniques, is used to generate S-boxes dynamically. The biggest advantage of this approach is the production of the inverted S-box with the S-box. Compared to the methods in the literature, the need to store the S-box is eliminated. Also, the fabr
... Show MoreSecured multimedia data has grown in importance over the last few decades to safeguard multimedia content from unwanted users. Generally speaking, a number of methods have been employed to hide important visual data from eavesdroppers, one of which is chaotic encryption. This review article will examine chaotic encryption methods currently in use, highlighting their benefits and drawbacks in terms of their applicability for picture security.
In this paper, we designed a new efficient stream cipher cryptosystem that depend on a chaotic map to encrypt (decrypt) different types of digital images. The designed encryption system passed all basic efficiency criteria (like Randomness, MSE, PSNR, Histogram Analysis, and Key Space) that were applied to the key extracted from the random generator as well as to the digital images after completing the encryption process.
With the rapid development of smart devices, people's lives have become easier, especially for visually disabled or special-needs people. The new achievements in the fields of machine learning and deep learning let people identify and recognise the surrounding environment. In this study, the efficiency and high performance of deep learning architecture are used to build an image classification system in both indoor and outdoor environments. The proposed methodology starts with collecting two datasets (indoor and outdoor) from different separate datasets. In the second step, the collected dataset is split into training, validation, and test sets. The pre-trained GoogleNet and MobileNet-V2 models are trained using the indoor and outdoor se
... Show MoreNowadays, it is quite usual to transmit data through the internet, making safe online communication essential and transmitting data over internet channels requires maintaining its confidentiality and ensuring the integrity of the transmitted data from unauthorized individuals. The two most common techniques for supplying security are cryptography and steganography. Data is converted from a readable format into an unreadable one using cryptography. Steganography is the technique of hiding sensitive information in digital media including image, audio, and video. In our proposed system, both encryption and hiding techniques will be utilized. This study presents encryption using the S-DES algorithm, which generates a new key in each cyc
... Show MoreWhenever, the Internet of Things (IoT) applications and devices increased, the capability of the its access frequently stressed. That can lead a significant bottleneck problem for network performance in different layers of an end point to end point (P2P) communication route. So, an appropriate characteristic (i.e., classification) of the time changing traffic prediction has been used to solve this issue. Nevertheless, stills remain at great an open defy. Due to of the most of the presenting solutions depend on machine learning (ML) methods, that though give high calculation cost, where they are not taking into account the fine-accurately flow classification of the IoT devices is needed. Therefore, this paper presents a new model bas
... Show MoreThe matter of handwritten text recognition is as yet a major challenge to mainstream researchers. A few ways deal with this challenge have been endeavored in the most recent years, for the most part concentrating on the English pre-printed or handwritten characters space. Consequently, the need to effort a research concerning to Arabic texts handwritten recognition. The Arabic handwriting presents unique technical difficulties because it is cursive, right to left in writing and the letters convert its shapes and structures when it is putted at initial, middle, isolation or at the end of words. In this study, the Arabic text recognition is developed and designed to recognize image of Arabic text/characters. The proposed model gets a single l
... Show More