Vellore Institute of Technology (VIT) is a private university located in&nbsp;Tamil Nadu, India. Founded in 1984, as Vellore Engineering College, the institution offers 20 undergraduate, 34 postgraduate, four integrated and four research programs. It has campuses in Vellore, Amravati, Bhopal and Chennai.

VIT is one of the top ranked private universities in India according to NIRF, THE and QS Rankings.&nbsp;Govt. of India has recognized&nbsp;VIT, Vellore as an&nbsp;Institution of Eminence. This has allowed VIT to take independent quality initiatives and move up in world ranking.

&nbsp;

&nbsp;

VIT University

Face log detection (FLD) in the surveillance video extracts a new face image from the video sequences (VS). FLD utilizes biometric techniques for humans’ recognition. To improve the precise FLD with less complexity of our proposed method is Prewitt Logistic Deep Recurrent Neural Learning (PLDRNL) used. The input VS was received from the video database. Next, the keyframes are extracted from the VS. This proposed deep recurrent neural learning method uses four hidden layers to remove the facial features such as the face, eyes, nose, and mouth in the form of an edge. The edges of each element are derived using the Prewitt edge detector through the horizontal and vertical mask. Finally, the relevant features are fed into the output layer. The PLDRNL uses a logistic activation function at the output layer for matching the extracted related elements with the pre-stored testing feature vector. If two features are matched, then the face in the given VF is detected. The error in the FD is minimized using gradient descent function at the output layer. Based on the results, the human face effectively identified with the minimum false-positive rate (FPR). Experimental evaluation is carried out using different factors such as FLD, FPR, and time complexity. © 2021, King Fahd University of Petroleum & Minerals.

World University root node

Department of Computational Intelligence

School of Computer Science and Engineering

Vellore Campus

Arabian Journal for Science and Engineering

Prewitt Logistic Deep Recurrent Neural Learning for Face Log Detection by Extracting Features from Images

Tourism became most popular in these decades, which the intensity of the user needs is enlarged significantly. Usually the popular travel websites give wide-ranging information such as cost details of the hotel, transportation prices, etc. But this information is not enough for the customers to get satisfied with travels in this model world. They expect more challenging information for their journey, and it should be presented on the websites, which helps to make creatively and provides more special travel plans for the users. Many more existing schemes are there on the web to improve the user for tourism. But these schemes give same and repeated information, and in many cases, the websites provide a smaller amount of secured and confidential user data and present many numbers of Ads, MNC tight-up companies oriented news for making money. In this proposed system, using the concept of text mining to recommend the best tourist spot based on budget, climate, crowd, etc. and also the data of the user are to be maintained very secure. In addition, many research activities are going on to provide more effective tourist spot for the tourists. The implementation tools are MongoDB, Angular Js, and the Spring framework is used in the development of the system. In this research designing the platform for tourists in Tamil-Nadu to include additional information about the tourist locations and also access the clear view on user review that is available in our platform helps fasten the tourist’s decision-making process. © 2020, World Academy of Research in Science and Engineering. All rights reserved.

International Journal of Advanced Trends in Computer Science and Engineering

Tourist spot proposal system using text mining

For maintaining the horticultural generation, Land Selection Investigation (LSI) is essential. Though incorporates estimation of the criteria assortment from the soil, territory to financial, market, and foundation, and these components are considerably enigmatically characterized and described by their inherent ambiguity. Multi-criteria basic leadership systems like positioning, rating, and so on are utilized for reasonableness examination. Master learning and judgment by leaders at different levels is integrated into this process. In the field of farming sciences, the Fuzzy Logic (FL) strategy has been effectively used to take care of numerous issues. Fuzzy with AHP is a Hybrid Fuzzy Logic (HFL) methodology. The policies Analytic Hierarchy Process (AHP), Fuzzy Numbers, Fuzzy Degree Investigation, Alpha Cut, and Lambda capacity are associated with it. As expressed, the procedure of necessary leadership includes a scope of criteria and a considerable measure of master learning and decisions. The components result from impacts extraordinarily. The capacity of three methods to demonstrate the affectability of the necessary leadership procedure is researched. Alpha cut and lambda esteem give and encourage considerable affectability investigation. All techniques are actualized to examine the reasonableness of the crop in the Indian nation. Test results when performed on Various Datasets, demonstrate that the proposed procedure removes more highlights just as gives more exactness when contrasted with existing techniques. © 2020 SCPE.

Scalable Computing

Investigation on agricultural land selection using hybrid fuzzy logic system ∗

The key formation is a crucial primitive for constructing secure networks: in a multi-party setting, it permits two parties using public authenticated communication to create a secret session key, which used to encrypt messages. QKD is an innovative technology that exploits the laws of quantum mechanics to generate and distribute unconditionally secure shared key for use in cryptographic applications. Currently, many symmetric key cryptographic tools, which are known to be quantum-safe. They all share secret symmetric keys, through untrusted communication, which finished with public-key systems, which are prone to quantum attacks. One of the recommended solutions to the key distribution problem is Quantum Key Distribution (QKD). QKD can be used to generate secret keys that are secure against any future algorithmic /computational improvements. QKD utilizes an authenticated communication channel along with a quantum communication channel so that a secret key generated. Whatever may be the protocol to implement proposed QKD, they all need both a quantum channel and an authenticated classical. OTP (One Time Pad) is a randomly generated key. Therefore, this key length is equal to the message. The proposed QKD result is 35.67% smaller keys for quantum security. The performance of a QKD system against a modeled photon number splitting attack analysed in simulation results and standard key agreement protocols can be secure. © School of Engineering, Taylor’s University

Journal of Engineering Science and Technology

An efficient algorithm for quantum computing with quantum key distribution with secure communication

The primary sources for ecological degradation currently are the Forest Fires (FF). The present observation frameworks for FF absence need supporting in constant checking of each purpose of the location at all time and prime location of the fire dangers. This approach gives works on preparing UAV (Unmanned Aerial Vehicle) aeronautical picture information as indicated by the prerequisites of ranger service territory application on a UAV stage. It provides a continuous and remote watch on a flame in forests and mountains, all the while the UAV is flying and getting the elevated information, helping clients maintain the number and area of flame focuses. Observing programming spreads capacities, including Fire: source identification, area, choice estimation, and LCD module. This paper proposed includes (1) Color Code Identification, (2) Smoke Motion Recognition, and (3) Fire Classification algorithms. Strikingly, the use of a helicopter with visual cameras portrayed. The paper introduces the strategies utilized for flame division invisible cameras, and the systems to meld the information acquired the following: Correctly, the current FF location stays testing, given profoundly convoluted and non-organized conditions of the forest, smoke hindering the flame, the movement of cameras mounted on UAVs, and analogs of fire attributes. These unfavorable impacts can truly purpose either false alert. This work focuses on the improvement of trustworthy and exact FF recognition algorithms which apply to UAVs. To effectively execute missions and meet their relating execution criteria examinations on the best way to diminish false caution rates, increment the possibility of profitable recognition, and upgrade versatile abilities to different conditions are firmly requested to improve the unwavering quality and precision of FF location framework. © 2019 Elsevier B.V.

Department of Energy and Power Electronics

School of Electrical Engineering

Department of Information Technology

School of Information Technology and Engineering

School of Electronics Engineering

Computer Communications

Unmanned Aerial Vehicle (UAV) based Forest Fire Detection and monitoring for reducing false alarms in forest-fires

Agriculture has always been the backbone of human civilization and something which has helped us flourish as a species since the beginning. Along with this, agriculture and the associated sectors have always had a major contribution towards India's GDP (Gross Domestic Product). As of 2017, they serve about 17.3 percent of the total GDP of India. Agriculture has also been the largest source of livelihood, for decades, in over 58 percent of the rural households of the country. Over the years, agriculture has seen many technological advancements. Despite this, agricultural practices in India are still largely traditional and are dependent on human involvement in a major way. India has a net sown area of around 14 crore hectares out of which only 47 percent of the land is under irrigation and still majorly depends majorly on good monsoon to fulfil the water requirements. Irrigation aside, farmers also need to know a lot of other problems like not getting enough value for their crops, producing more crops than the demand for that crop and spending excess amount of money in transporting the crops to the market. All this can be avoided if we introduce the technology that has been theoretically present for quite some time but has not been implemented at the ground level, yet. We are talking about IoT (Internet of Things) which is basically a network of objects enabled by internet coupled with the web services that helps in interact with these objects. The IoT network would include three subsystems which would each aim to solve the three major problems stated above. The first would solve the supply-demand anomaly, the next would solve the irrigation problem but would also perform data acquisition using the different sensor nodes. The last subsystem would be aimed at designing such a model that would reduce the transportation costs of the farmers considerably by enabling a pool system wherein every farmer would benefit and reduce their expense. © 2019 IEEE.

VIT Unitversity

2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI)

IoT based model of automated agricultural system in India

Cloud computing uses scheduling and load balancing for virtualized file sharing in cloud infrastructure. These two have to be performed in an optimized manner in cloud computing environment to achieve optimal file sharing. Recently, Scalable traffic management has been developed in cloud data centers for traffic load balancing and quality of service provisioning. However, latency reducing during multidimensional resource allocation still remains a challenge. Hence, there necessitates efficient resource scheduling for ensuring load optimization in cloud. The objective of this work is to introduce an integrated resource scheduling and load balancing algorithm for efficient cloud service provisioning. The method constructs a Fuzzy-based Multidimensional Resource Scheduling model to obtain resource scheduling efficiency in cloud infrastructure. Increasing utilization of Virtual Machines through effective and fair load balancing is then achieved by dynamically selecting a request from a class using Multidimensional Queuing Load Optimization algorithm. A load balancing algorithm is then implemented to avoid underutilization and overutilization of resources, improving latency time for each class of request. Simulations were conducted to evaluate the effectiveness using Cloudsim simulator in cloud data centers and results shows that the proposed method achieves better performance in terms of average success rate, resource scheduling efficiency and response time. Simulation analysis shows that the method improves the resource scheduling efficiency by 7% and also reduces the response time by 35.5 % when compared to the state-of-the-art works. © 2018 Elsevier B.V.

Applied Soft Computing

Resource scheduling algorithm with load balancing for cloud service provisioning

By description, a mobile ad hoc network does not stay on any permanent infrastructure; in its place, all networking utilities (e.g., routing, mobility management, etc.) are accessed by the nodes themselves in a self-organizing manner. On the other hand, it is rigid to support cooperativeness among the nodes for their own restricted resources that require to be conserved. These scrupulous nodes which are also termed as selfish nodes decline to help other nodes in forwarding packets owing to the anxiety of having resource. Several researches design a new method that aims at attaining confidentiality of the location for an efficient communication. Thus, privacy appears from the mobile network and users gain control over the disclosure of their locations. In this work, we propose closeness mechanism accepted from the assumption of small-world event or also termed as degrees of separation to persuade cooperativeness among nodes in a trusted node's community. This paper also provides some general idea on how to develop security on the trusted MANET community by adapting security features of trust. The simulation of the proposed Efficient Node Cooperation and Security [ENCS] in MANET work is done for varying topology, node size, attack type and intensity with different pause time settings and the performance evaluations are carried over in terms of node cooperativeness, clustering efficiency, communication overhead and compared with an existing secured key model.

International Journal of Engineering and Technology

Efficient node cooperation and security in MANET using closeness technique

A Mobile Ad-Hoc Network (MANET) is termed as a set of wireless nodes which could be built with infrastructure less environment where network services are afforded by the nodes themselves. In such a situation, if a node refuses to co-operate with other nodes, then it will lead to a considerable diminution in throughput and the network operation decreases to low optimum value. Mobile Ad hoc Networks (MANETs) rely on the collaboration of nodes for packet routing ahead. Nevertheless, much of the existing work in MANETs imagines that mobile nodes (probably possessed by selfish users) will pursue prearranged protocols without variation. Therefore, implementing the co-operation between the nodes turn out to be an significant issue. The previous work described a secured key model for ad hoc network with efficient node clustering based on reputation and ranking model. But the downside is that the co-operation with the nodes is less results in a communication error. To enhance the security in MANET, in this work, we present a hybrid approach, build a node co-operation among the nodes in MANET by evaluating the weightage of cooperativeness of each node in MANET. With the estimation of normal co-operative nodes, nodes are restructured on its own (self). Then clustering is made with the reorganized nodes to form a secured communication among the nodes in the MANET environment. The Simulation of the proposed Hybrid Approach for Node Cooperation based Clustering (HANCC) work is done for varying topology, node size, attack type and intensity with different pause time settings and the performance evaluations are carried over in terms of node cooperativeness, clustering efficiency, communication overhead and compared with an existing secured key model. Compared to an existing secured key model, the proposed HANCC performance is 80-90% high. © 2013 Science Publications.

Fulltext

Journal of Computer Science

A hybrid approach for node co-operation based clustering in mobile Ad Hoc networks

An ad-hoc network is referred as wireless network in which the nodes in the network are independent of each other and free to move. Since nodes are independent, there is a great extent of loss of packet data while transferring the packet from source to destination. There is a chance of some nodes to be misbehaved. This unsecured ad-hoc network environment is caused due to the dynamic nature of networks and node mobility. Key management is also difficult in ad-hoc network. Several researches have proposed techniques using clusters for an efficient communication from one network to other network. The previous work presented a fully-distributed ID-based Multiple secrets Key Management scheme (IMKM) for a secure communication. Using cryptography and ID based secret key, group key management is done. It provides an authentication about the nodes in the networks efficiently. But the drawback is that it consumes more bandwidth usage for a secure communication and the clustering with IMKM is also unreliable and it takes several rounds to identify the secure channel. In this work, we are going to present a Self Re-Organizing Node Clusters for Secured Ad Hoc Communication. The proposed work present a secured key model for ad hoc network with efficient node clustering based on reputation and ranking model. Misbehavior and selfish nodes in ad-hoc network are detected efficiently with reputation ranking. With the evaluation of normal co-operative nodes, clustering of node is reorganized on its own (self). It does not allow any unauthorized node to involve in the communication between the nodes in ad-hoc network. The Simulation of our work is done for varying topology, node size, attack type and intensity with different pause time settings. The performance metrics used in the evaluation are node reputation count, node density, and attack intensity, response time for malicious node detection, cluster size, and cluster purity index. © 2012 EuroJournals Publishing, Inc.

European Journal of Scientific Research

A secured key model for efficient node clustering based on reputation and ranking model

Wireless sensor networks are catching up as the primary mode for monitoring and collecting data in physically challenging environments. They And applications in various fields varying from environment monitoring, military applications to monitoring patients in hospitals. The constraints due to their inherent features make routing in wireless sensor networks a big challenge. This paper covers Genetic algorithm (GA) based simple straight forward, address based shortest path routing in Wireless ad hoc sensor networks (WASN). The routing in packet switched multi-hop networks can be described as a classical combinatorial optimization problem i.e. a shortest path routing problem in graphs. The stress is on energy efficient routing where the network optimizes between efficient routing and maximizing life of the network. The results show that the GA is efficient for energy constrained WASNs due to their fastness in computation comparing to all other algorithms. © 2010 IEEE.