Journal of Cybersecurity and Information Management

Journal DOI

Submit Your Paper

2690-6775ISSN (Online) 2769-7851ISSN (Print)

A Novel Image Encryption with Deep Learning Model for Secure Content based Image Retrieval

Mohamed Elsharkawy , Ahmed N. Al Masri

From the last decades, a massive quantity of images gets generated and continues to rise to a maximum extent in the forthcoming data. The process of retrieving images based on a query image (QI) is a proficient method of accessing the visual properties from large datasets. Content-based image retrieval (CBIR) provides a way of effectively retrieving images from large databases. At the same time, image encryption techniques can be integrated into the CBIR model to retrieve the images securely. Therefore, this paper presents new image encryption with a deep learning-based secure CBIR model called IEDL-SCBIR. The proposed IEDL-SCBIR technique intends to encrypt the images as well as securely retrieve them. The proposed IEDL-SCBIR technique follows a two-stage process: optimal elliptic curve cryptography (ECC) based encryption and DL based image retrieval. The proposed model derives a cuckoo search optimization (CSO) with the ECC technique for the image encryption process in which the CSO algorithm is applied for optimal key generation. In addition, VGG based feature extraction with Euclidean distance-based similarity measurement is applied for the retrieval process. To validate the enhanced performance of the IEDL-SCBIR technique, a comprehensive results analysis takes place, and the obtained results demonstrate the betterment over the other methods.

Read More


Vol. 0 Issue. 2 PP. 54-64, (2019)

Mitigating DDoS Attacks in Wireless Sensor Networks using Heuristic Feature Selection with Deep Learning Model

Abdul Rahaman Wahab Sait , Irina Pustokhina , M. Ilayaraja

A wireless sensor network (WSN) encompasses a massive set of sensors with limited abilities for gathering sensitive data. Since security is a significant issue in WSN, there is a possibility of different types of attacks. In Distributed Denial of Service (DDOS) attack, the malicious node can adapt to several attacks, namely flooding, black hole, warm hole, etc., to interrupt the working of the WSN. The recently developed deep learning (DL) models can effectively detect DDoS attacks in the network. Therefore, this article proposes a heuristic feature selection with a Deep Learning-based DDoS (HFSDL-DDoS) attack detection model in WSN. The proposed HFSDL-DDoS technique intends to identify and categorize the occurrence of DDoS attacks in WSN. In addition, the HFSDL-DDoS technique involves the immune clonal genetic algorithm (ICGA) based feature selection (FS) approach to improve the detection performance. Moreover, a fruit fly algorithm (FFA) with bidirectional long, short-term memory (BiLSTM) based classification model is employed. The experimental analysis of the HFSDL-DDoS technique is performed, and the results are examined interms of several performance measures. The resultant experimental results pointed out the betterment of the HFSDL-DDoS technique over the other techniques.

Read More


Vol. 0 Issue. 2 PP. 65-74, (2019)

Dragonfly Algorithm with Gated Recurrent Unit for Cybersecurity in Social Networking

Yutao Han , Ibrahim M. EL-Hasnony , Wenbo Cai

The advancements of information technologies and wireless networks have created open online communication channels. Inappropriately, trolls have abused the technologies to impose cyberattacks and threats. Automated cybersecurity solutions are essential to avoid the threats and security issues in social media. This paper presents an efficient dragonfly algorithm (DFA) with gated recurrent unit (GRU) for cybersecurity in social networking. The proposed DFA-GRU model aims to determine the social networking data into neural statements or insult (cyberbullying) statements. Besides, the DFA-GRU model primarily undergoes preprocessing to get rid of unwanted data and TF-IDF vectorizer is used. In addition, the GRU model is employed for the classification process in which the hyperparameters are optimally adjusted by the use of DFA, and thereby the overall classification results get improved. The performance validation of the DFA-GRU model is carried out using benchmark dataset and the results are examined under varying aspects. The experimental outcome highlighted the enhanced performance of the DFA-GRU model interms of distinct measures.

Read More


Vol. 0 Issue. 2 PP. 75-88, (2019)