The research team from the Department of Computer Science and Engineering proposes a research scheme to address security concerns in the transmission of digital images of aerial Remote sensing images over the Internet. Assistant Professor Dr Priyanka, Assistant Professor Dr Jatindra Kumar Dash, research scholar Ms K Jyothsna Devi, and BTech student Mr. M V Jayanth Krishna, published the paper A New Robust and Secure 3-Level Digital Image Watermarking Based on G-BAT Hybrid Optimization in the Mathematics Journal SCI, a Q1 Journal with an Impact Factor of 2.9. The research project combats various threats in the transmission of Remote sensing images, such as copyright protection, copy control, and unauthorized access.

Abstract

This contribution applies tools from the information theory and soft computing (SC) paradigms to the embedding and extraction of watermarks in aerial remote sensing (RS) images to protect copyright. By the time 5G came along, Internet usage had already grown exponentially. Regarding copyright protection, the most important responsibility of the digital image watermarking (DIW) approach is to provide authentication and security for digital content. The main goal of the paper is to provide authentication and security to aerial RS images transmitted over the Internet by the proposal of a hybrid approach using both the redundant discrete wavelet transform (RDWT) and the singular value decomposition (SVD) schemes for DIW. Specifically, SC is adopted in this work for the numerical optimisation of critical parameters. Moreover, 1-level RDWT and SVD are applied to digital cover images and singular matrices of LH and HL sub-bands are selected for watermark embedding. Further selected singular matrices S LH and S HL are split into 3 × 3 non-overlapping blocks, and diagonal positions are used for watermark embedding. Three-level symmetric encryption with a low computational cost is used to ensure higher watermark security. A hybrid grasshopper–BAT (G- BAT) SC-based optimization algorithm is also proposed to achieve high-quality DIW outcomes, and a broad comparison against other methods in the state-of-the-art is provided. The experimental results have demonstrated that the proposal provides high levels of imperceptibility, robustness, embedding capacity, and security when dealing with DIW of aerial RS images, even higher than the state-of-the-art methods.

The proposed scheme is easily dumped into the sender and receiver machines to work readily. MATLAB platform is the only requirement. Researchers aspire to design new image watermarking schemes using machine learning and deep learning techniques. For this project, they have collaborated with Professor José Santamaría from the Department of Computer Science, University of Jaén, and Professor Antonio Romero-Manchado from the Department of Cartographic Engineering, Geodesy, and Photogrammetry, University of Jaén.

The Department of Computer Science and Engineering is glad to announce that Assistant Professor Dr Saleti Sumalatha and her students got two of their patent applications published in a row. The patent titled “System and method for mining of constraint based high utility time interval sequential patterns” (Application number: 202241044001) was published in collaboration with the BTech students; K Rasagna, N Naga Sahithya, K Hemalatha, B Sai Charan, and Upendra Karthik.

The main intention of the proposed system is to discover the sequences that include the time period between the purchases of each item. For example, if we consider a shop which sells some groceries like Grains, Milk, Yogurt, Bread and Eggs as the set of items in the database. Now, the main aim is to find the time period between the purchases of particular items that are being sold. From this, the shop owner can easily maintain the stock of completed items according to the time period.

For example, an output sequential pattern including time intervals of the form indicates that a customer who purchased item x also bought item y after three months and visited the store again after five months to buy item z. It considers the same utility threshold for each of the items in the database, which shows that each item is assumed to have the same unit profit. This is not convincing as each item is different in real-time applications and should not be treated equally. For example, the sales of” Gold bangles” will produce more profit than the sales of” Cotton Jeans”. In view of this, the research proposes UIPrefixSpan-MMU.

The other patent titled “A system and a method for automatic essay grading” (Application number: 202241043045) was published in collaboration with M Purnima, G Haveela, K U Meghana, and S Deepthi Reddy. Essay grading systems are being adopted by different organisations to reduce the hectic workload from a teacher’s point of view. They made a model which is trained with datasets containing different essay topics and numerous essays with scores varying in a wide range.

Essay grading systems will not only save the time for evaluation but also give accurate results. The output of the system will be quick such that it could evaluate many essays and get trained. This system benefits both the student and the teacher as well. Their model predicts the scores of the essay by comparing them with the features extracted from the trained data. This model can be used to reduce the effort of teachers to grade many essays in a limited time. The work of grading will be fastened and accurate.

The Department of Computer Science and Engineering is delighted to announce that Assistant Professors, Dr Sambit Kumar Mishra and Dr Tapas Kumar Mishra, and the final year BTech students – Kotipalli Sindhu, Mogaparthi Surya Teja, Vutukuri Akhil, Ravella Hari Krishna, and Pakalapati Praveen – published a chapter titled Applications of Federated Learning in Computing Technologies in the scopus-indexed book titled Convergence of Cloud with AI for Big Data Analytics: Foundations and Innovation, a Wiley publication.

The book chapter describes the application of federated learning to various computing technologies. The federated learning concept is similar to the client-server model, where the client sends data to the server for processing and the processed data is again sent to the client. But, in federated learning, the clients are allowed separately to teach the deep neural network models with the local data combined with the deep neural network model at the central server.

Federated learning is a machine learning technique that trains the knowledge across different decentralized devices holding samples of information without exchanging them. The concept is additionally called collaborative learning. Researchers have used large frameworks for all the computations in the past years, and then they have moved to client-server frameworks.

It is also like a traditional centralised machine learning technique. All the local datasets are uploaded to a minimum of one server, so it assumes that local data samples are identically distributed. Because of its security and privacy concerns, it’s widely utilised in many applications like IoT, cloud computing; Edge computing, Vehicular edge computing, and many more. In the chapter, different applications of federated learning, their privacy concerns, and their definition in various fields of computing technologies like IoT, Edge Computing, Cloud Computing, Vehicular edge computing, etc. are presented. It will be of advantage to graduate students, researchers, academicians, institutions, and professionals that are interested in exploring the areas of intelligent computing systems.

Intelligent transport systems embedded with latest technology is the future of automotive industry and can revolutionise the very infrastructure and architecture of transportation. The principles of intelligent transportation systems have manifested in the development of Internet of vehicles (IoV), a novel technological network that enables autonomous and connected mobile commutation facility. Dr Sriramulu Bojjagani, Assistant Professor, Department of Computer Science and Engineering, has published a paper on this seminal topic titled “Secure Authentication and Key Management Protocol for Deployment of Internet of Vehicles (IoV) Concerning Intelligent Transport Systems” in the prestigious journal, IEEE Transactions on Intelligent Transport Systems (Q1 journal) having an Impact factor of 9.551. The paper highlights the development and significance of the pioneering AKAP-IoV system in successfully tackling cyber threats and security vulnerabilities.

Abstract

Intelligent transport systems amalgamated with advanced technologies are an important element of the automotive industry, including critical infrastructure and transportation. Internet of Vehicles (IoV) is the modern technological framework designed for intelligent transportation. IoV creates a network of information relations among vehicles, thus contributing to reduced congestion, roadside infrastructure, driver/traveller safety, and traffic efficiency through wireless communication and sensing technology. However, a significant challenge in IoV applications is security, as criminals could potentially exploit these applications. Despite increasing industry awareness, the potential danger posed by security vulnerabilities and cyber threats remains high. In this study, we have designed a new system called AKAP-IoV, which supports secure communication, mutual authentication, and key management among vehicles, roadside units, and fog and cloud servers. AKAP-IoV was tested and verified using Scyther and Tamarin to ensure its resistance to cyber threats. Furthermore, we conducted a formal security analysis using the Real-or-Random (RoR) oracle model to assess security properties logically. In addition, a detailed, comprehensive comparative study was considered to evaluate the performance, functionality, efficiency, and security features supported by AKAP-IoV compared to those of recently developed schemes.

Read full article

An Image caption generator system implies the detection of the image as well as producing the caption with natural language processing by the computer. This is a tedious job. Image caption generator systems can solve various problems, such as self-driving cars, aiding the blind, etc.

The recent research at the Department of Computer Science and Engineering proposes a model to generate the captions for an image using ResNet and Long Short-Term Memory. Assistant Professors Dr Morampudi Mahesh Kumar and Dr V Dinesh Reddy have published the paper Image Description Generator using Residual Neural Network and Long-Short-Term Memory in the Computer Science Journal of Moldova with an impact factor of 0.43.

The captions or descriptions for an image are generated from an inverse dictionary formed during the model’s training. Automatic image description generation is helpful in various fields like picture cataloguing, blind persons, social media, and various natural language processing applications.

Despite the numerous enhancements in image description generators, there is always a scope for development. Taking advantage of the larger unsupervised data or weakly supervised methods is a challenge to explore in this area, and this is already there among the future plan of the researchers. Another major challenge could be generating summaries or descriptions for short videos. This research work can also be extended to other sets of natural languages apart from English.

Abstract

Human beings can describe scenarios and objects in a picture through vision easily, whereas performing the same task with a computer is a complicated one. Generating captions for the objects of an image helps everyone to understand the scenario of the image in a better way. Instinctively describing the content of an image requires the apprehension of computer vision as well as natural language processing. This task has gained huge popularity in the field of technology, and there is a lot of research work being carried out. Recent works have been successful in identifying objects in the image but are facing many challenges in generating captions to the given image accurately by understanding the scenario. To address this challenge, we propose a model to generate the caption for an image. Residual Neural Network (ResNet) is used to extract the features from an image. These features are converted into a vector of size 2048. The caption generation for the image is obtained with Long Short-Term Memory (LSTM). The proposed model was experimented with on the Flickr8K dataset and obtained an accuracy of 88.4%. The experimental results indicate that our model produces appropriate captions compared to the state of art models.

Complex IoT networks comprise multiple devices connected to the gateway. A smoother functioning is ensured through new techniques for device authentication to the gateway. On this note, the research paper titled “A Lightweight Mutual and Transitive Authentication Mechanism for IoT Network” has been published by Dr Amit Kumar Mandal, Assistant Professor, Department of Computer Science and Engineering and his research scholar Mrs Rudra Krishna Srija in the Q1 Journal Ad Hoc Networks, Elsevier having an impact factor of 4.8. The research details the use of the polynomial-based protocol in enhancing device connection for transitive communication.

Abstract of the paper

In large and complex IoT systems like the smart city or smart industry which consist of thousands of connected devices, it may not always be feasible to be directly connected to the gateway but it may be possible to be connected to another device. Therefore, already authenticated devices should facilitate the new device’s authentication by the gateway. To address this issue, the existing protocols use multiple authentication protocols based on different cryptography techniques, which are difficult to implement and manage in resource-constrained IoT devices. In this paper, we propose a Transitive device authentication protocol based on the Chebyshev polynomial.

The work is primarily aimed at improving transitive communication in machine-to-machine communication or device-to-device communication in large-scale heterogeneous IoT network scenarios. The research team targets to investigate the benefits of adopting the designed protocol in particular within low-power and lossy networks in the future.

Collaborations

Università Ca Foscari Venezia, Venice, Italy