GKGunasekaran Kulandaivelu

Gunasekaran Kulandaivelu

*This is an algorithmically generated author record based on publicly available information.

Gunasekaran Kulandaivelu has a total of 20 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Trust‐Aware Routing Protocol Using Hierarchical Manta Ray Foraging Optimization Algorithm With Selfish Node Detection in MANET

    by Suresh Jeganathan, Gunasekaran Kulandaivelu, Dhurgadevi Muthusamy, Kishore Verma Samraj

    Mobile ad hoc networks (MANETs) are fast, self‐organizing, infrastructure‐less wireless networks; it is highly suitable for use in emergencies, natural catastrophes, and between places without infrastructure. However, this approach fails to focus on the selfish nodes. To address this, a trust‐aware routing protocol using the Hierarchical Manta Ray Foraging Optimization Algorithm (HMRFOA) with selfish node detection in MANET (SND‐MSWGCN‐TRP‐MANET) is proposed in this manuscript. Initially, clustering is processed using Localized Sparse Incomplete Multiview Clustering (LSIMVC). The node is clustered by merging or combining two maximum comparable cluster nodes; then, the clustered node is assigned to the Mud Ring Optimization Algorithm (MROA), where the cluster head is selected using the parameters, such as energy (E), distance (Dist), delay (D), and quality of service (QoS). The selected cluster head is fed to the Multiscale Superpixel‐Guided Weighted Graph Convolutional Network (MSWGCN) for effectively detecting the selfish node to ensure secure and efficient data routing in MANET. Finally, trust‐aware routing is performed using the HMRFOA, where it selects the optimal path. Then, the proposed SND‐MSWGCN‐TRP‐MANET is implemented, and the performance metrics like energy consumption, time consumption, end‐to‐end delay, throughput, average delay, packet delivery ratio (PDR), and packet loss ratio (PLR) are examined. The performance of the SND‐MSWGCN‐TRP‐MANET approach attains 14.82%, 21.63%, and 31.57% lower energy consumption and 16.23%, 24.19%, and 31.82% higher throughput when analyzed to the existing techniques, like selfish node trust aware along optimized clustering for dependable routing protocol (SNTA‐CRRP‐NANET), trust‐aware safe energy‐efficient hybrid protocol in MANET (TASEP‐MANET), and cluster with angular base energy efficient trusted routing protocol in MANET (EPTRP‐MANET).


  • Detection of Video Anomaly in Public With Deep Learning Algorithm

    by M. Dhurgadevi, D. Vimal Kumar, R. Senthilkumar, K. Gunasekaran

    For traffic control and public safety, predicting the movement of people is crucial. The presented scheme entails the development of a wider network that can better satisfy created synthetic images by connecting spatial representations to temporal ones. The authors exclusively use the frames from those occurrences to create the dense optical flow for their corresponding normal events. In order to eliminate false-positive detection findings, they determine the local pixel reconstruction error. This particle prediction model and a likelihood model for giving these particles weights are both suggested. These models effectively use the variable-sized cell structure to produce sceneries with variable-sized sub-regions. It also successfully extracts and utilizes the video frame's size, motion, and position information. On the UCSD and LIVE datasets, the proposed framework is evaluated with the most recent algorithms reported in the literature. With a significantly shorter processing time, the suggested technique surpasses state-of-the-art techniques in relation to decreased equal error rate . Request access from your librarian to read this chapter's full text.

  • Retracted
    Open Access

    [Retracted] A Deep Learning Framework for Earlier Prediction of Diabetic Retinopathy from Fundus Photographs

    by K. Gunasekaran, R. Pitchai, Gogineni Krishna Chaitanya, D. Selvaraj, S. Annie Sheryl, Hesham S. Almoallim, Sulaiman Ali Alharbi, S. S. Raghavan, Belachew Girma Tesemma

    Diabetic patients can also be identified immediately utilizing retinopathy photos, but it is a challenging task. The blood veins visible in fundus photographs are used in several disease diagnosis approaches. We sought to replicate the findings published in implementation and verification of a deep learning approach for diabetic retinopathy identification in retinal fundus pictures. To address this issue, the suggested investigative study uses recurrent neural networks (RNN) to retrieve characteristics from deep networks. As a result, using computational approaches to identify certain disorders automatically might be a fantastic solution. We developed and tested several iterations of a deep learning framework to forecast the progression of diabetic retinopathy in diabetic individuals who have undergone teleretinal diabetic retinopathy assessment in a basic healthcare environment. A collection of one‐field or three‐field colour fundus pictures served as the input for both iterations. Utilizing the proposed DRNN methodology, advanced identification of the diabetic state was performed utilizing HE detected in an eye’s blood vessel. This research demonstrates the difficulties in duplicating deep learning approach findings, as well as the necessity for more reproduction and replication research to verify deep learning techniques, particularly in the field of healthcare picture processing. This development investigates the utilization of several other Deep Neural Network Frameworks on photographs from the dataset after they have been treated to suitable image computation methods such as local average colour subtraction to assist in highlighting the germane characteristics from a fundoscopy, thus, also enhancing the identification and assessment procedure of diabetic retinopathy and serving as a skilled guidelines framework for practitioners all over the globe.


  • Open Access

    Deep multimodal biometric recognition using contourlet derivative weighted rank fusion with human face, fingerprint and iris images

    by K. Gunasekaran, J. Raja, R. Pitchai

    The goal of multimodal biometric recognition system is to make a decision by identifying their physiological behavioural traits. Nevertheless, the decision-making process by biometric recognition system can be extremely complex due to high dimension unimodal features in temporal domain. This paper explains a deep multimodal biometric system for human recognition using three traits, face, fingerprint and iris. With the objective of reducing the feature vector dimension in the temporal domain, first pre-processing is performed using Contourlet Transform Model. Next, Local Derivative Ternary Pattern model is applied to the pre-processed features where the feature discrimination power is improved by obtaining the coefficients that has maximum variation across pre-processed multimodality features, therefore improving recognition accuracy. Weighted Rank Level Fusion is applied to the extracted multimodal features, that efficiently combine the biometric matching scores from several modalities (i.e. face, fingerprint and iris). Finally, a deep learning framework is presented for improving the recognition rate of the multimodal biometric system in temporal domain. The results of the proposed multimodal biometric recognition framework were compared with other multimodal methods. Out of these comparisons, the multimodal face, fingerprint and iris fusion offers significant improvements in the recognition rate of the suggested multimodal biometric system.


  • Performance analysis of integrated multimodal biometrics by means of soft computing techniques

    by K. Gunasekaran, D. Saravanan, P. Akilan

    Providing security by means of using traditional methods such as PINs, Passwords, access cards are being vulnerable to various attacks and are easily hacked by intruders or unauthorized users. The aim is to develop an efficient personal identification system to provide secure access for legitimate users which is a challenging task. In order to avoid these problems, the use of biométrie traits reliably provides efficient and effective solutions. Though unimodal biometrics faces various issues such as non-universality, distinctiveness, spoof attacks, noise in sensed data, intra-class variations. To overcome these issues, Multimodal biometrics based personal identification system is developed using Soft Computing Techniques. The biométrie traits such as fingerprint, palmprint and finger knuckle print are used for recognition system. Artificial Neural Networks (ANN) and Genetic Algorithm are used for this purpose. The Haralick features are used for the recognition system. The proposed system achieves excellent identification rate and higher user acceptability. Thus the proposed system reduces the FRR rate.

  • Enforcing secure and privacy in information brokering system

    by K. Gunasekaran, M. Indumathi, D. Saravanan

    An Information Brokering System is one that is used for information sharing based on the client-server request. It consists of data servers and brokering components which helps client queries to locate the data servers. In traditional system, the brokers had the data confidentiality for accessing data in server side control in IBS system. At the same time the privacy is maintained in the data sharing system in data location and the data consumer queries within the IBS system. In addition with this, the end applicant will have problems such as attribute correlation and inference attacks and does not provide a system-wide coordination while information is sharing. The proposed system named Privacy Preserving Information Brokering (PPIB) using the automaton segmentation scheme and query segment encryption scheme to overcome the above discussed attacks. In this Privacy Preserving Information Brokering system the privacy and secure transmissions of data is maintained and secured trust is evaluated.

  • Implementation of multimodal biometric authentication using soft computing techniques

    by K. Gunasekaran, P. Mahalakshmi

    Identity Management refers to a challenge of providing authorized users with secure and easy access to information and services for personal identification system. The objective is to implement a secure identification system for determining an individual's identity. Conventional methods for personal identification include keys, tokens, access cards, PINs and passwords which leads to drawbacks such as stealing, duplicating, cracking and sharing. In order to overcome the drawbacks, biometrics based identification is implemented. Nevertheless, unimodal biometrics is suffered due to noise, intra class variations, spoof attacks, non-universality. To avoid these attacks, multimodal biometrics which is the combination of more modalities is adapted. Fingerprint and Palmprint identification are the emerging and leading technologies used as a source of authentication. Neural Networks one of the Soft Computing techniques has been proposed for feature extraction and minutiae point detection. Euclidean distance values are used for minutiae matching. The proposed method achieves excellent identification rate and are more secure in noisy environment.


  • Comparison of Artificial Neural Network Models and Multiple Linear Regression Models in Cargo Port Performance Prediction

    by P.Oliver Jayaprakash, K. Gunasekaran, S. Muralidharan

    Cargo ports operational performance was specified typically through revenue earned, quantum of cargo handled and number of ships serviced. It was predisposed by infrastructural facilities and cargo handling rate; it had an effect over pre-berth waiting time of vessels waiting and berthing time of ships at a port. An Indian port’s ship movement and port operational characteristics had been studied for five years (2005-2009). Ship’s service time was the crucial parameter used to quantify the port performance. This paper focused on building an artificial neural network technique based model to illustrate the relationship between service time and port operational characteristics. Validations of ANN model, comparing multiple linear regression model outputs were reported.