Thangamariappan Ganesh Kumar

Thangamariappan Ganesh Kumar

Auto-generated profile, based on publicly available information

Thangamariappan Ganesh Kumar

Galgotias University

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 October 2022 in Journal of Sensors
Verified
Journal Article
Open access
PDF file
SJS. Janarthanan
S. Janarthanan
TKT. Ganesh Kumar
T. Ganesh Kumar
SJS. Janakiraman
S. Janakiraman
Rajesh Kumar DhanarajRajesh Kumar Dhanaraj
Rajesh Kumar Dhanaraj
MSMohd Asif Shah
Mohd Asif Shah
A significant amount of effort and cost is required to collect training samples for remote sensing image classifications. The study of remote sensing and how to read multispectral images is becoming more important. High-dimensional multispectral images are created by the various bands that show how materials behave. The need for more information about things and the improvement of sensor resolutions have led to the creation of multispectral data with a higher size. In recent years, it has been shown that the high dimensionality of these data makes it hard to preprocess them in multiple ways. Recent research has demonstrated that one of the most crucial methods to address this issue is by adopting a variety of learning strategies. But as the data gets more complicated, these methodologies are not adequate to support. The proposed methodology shows that the classification experiment using remote sensing images indicates the maximum likelihood classifier with different deep learning models; weight vector (WV) AdaBoost and ADAM can greatly limit overfitting, and it obtains high classification accuracy. Proposed VGG16 and Inception v3 increase classification accuracy along with optimization process produce 96.08.

25 March 2022 in Latin American Applied Research - An international journal
Verified
Journal Article
Open access
MLMuthu Lakshmi
Muthu Lakshmi
GKGanesh Kumar
Ganesh Kumar
BSBala Subramanian
Bala Subramanian
PRPriti Rishi
Priti Rishi
-- The main objective of this paper is to examine the performance of Neuro- Statistical sparse transformation function for implementation in a still image vector coding based compression system. This paper discusses the important features of low bit-rate image coding which is based on recent developments in the theory of multivariate nonlinear piecewise polynomial approximation in still images. It combines Binary Space Partition (BSP) scheme with Geometric Wavelet (GW) tree approximation so as to efficiently capture curve singularities and provide a sparse representation of the image. The quality of the reconstructed image is measured objectively using Peak Signal to Noise Ratio. Experimental results show that the proposed image compression system yields higher compression with minimal loss.

4 May 2023 in Research Square
Verified
Preprint
Open access
PDF file
SDSumathi Doraikannan
Sumathi Doraikannan
VKVaranasi Lvskb Kasyap
Varanasi Lvskb Kasyap
MRMure Sai Jaideep Reddy
Mure Sai Jaideep Reddy
VBVaranasi Srinivasa Bhaga
Varanasi Srinivasa Bhaga
TPThangamuthu Poongodi
Thangamuthu Poongodi
SSSanthosh Kumar Svn
Santhosh Kumar Svn
In recent years, considerable attention has been paid to 3D face data in many face image processing applications. Detailed 3D Face making is developing technology with multiple real-time applications. This work aims to create an exact 3D Face model with facial emotions designed based on the principle of the Face Vertex Land marking and Wulcheir distance. Convolution Neural Network (DCNN) is deployed to extract relevant facial features and those features are used for further analysis. The 3D Face models are constructed efficiently. The proposed model is a concoction of CoarseNet and FineNet through which a 3D coarse face from a bilinear face model with face landmark alignment is created. It is followed by the local corrective field which tends to refine the 3D rough face with consistent photometric constraint. This work follows the various aspects of 3D face modeling techniques: Deep Learning, Epiploic Geometry, and the One-shot learning (DEO) method. The proposed DEO Model has been evaluated using the FER2013 dataset of face images with six basic emotions via performance metrics like accuracy, precision, sensitivity, specificity, and time. The proposed model outperforms other existing methods with promising and state-of-art results. The accuracy obtained through the proposed work shows higher accuracy (more than 90%), which has been demonstrated using real-world models

25 July 2023 in Research Square
Verified
Preprint
Open access
PDF file
VPVijitha Devi. P
Vijitha Devi. P
KKK. Kavitha
K. Kavitha
DUD. Usha
D. Usha
TKT. Ganesh Kumar
T. Ganesh Kumar
The Wireless Sensor Networks (WSN) include several sensors which collect data from their surroundings and transmit it to the destination node. In WSN, sensor nodes operate individually and build the ad-hoc network infrastructure. The observing data perceives the sensor nodes and the data have been sent to the Base Station (BS) with the utilization of gateway and Cluster Head (CH). The WSNs have limited battery power which diminishes the lifetime of the network. The clustering algorithm is one of the efficient solutions that help in improving the WSNs network. Many algorithms were applied in the clustering process in rectifying the issue of NP-hard optimization. But, these algorithms lead to disadvantages namely delay, slow conveyance rate and weak exploitation stage. Therefore, in this paper, a novel Fuzzy Chaotic Whale Wolf (FCWW) optimization is proposed for selecting the optimal CH. The primary goal of our article is to lessen computational complexity with increased network lifetime. The main operation of the CH is gathering information from the sensor nodes and directly communicating with the BS. As compared to other techniques, the proposed FCWW algorithm provides 712 kbps of throughput, 98.8% of packet delivery ratio, network lifetime of 1652 seconds, 35% of energy consumption and 82 ms of end-to-end latency respectively.

1 January 2024 in AIP Conference Proceedings
Verified
Conference Paper
Open access
PDF file
CDCharul Dewan
Charul Dewan
TKT. Ganesh Kumar
T. Ganesh Kumar
SGSunil Gupta
Sunil Gupta
The Tele care medicine information system (TMIS) enable patients to gain admission to healthcare services over long distances. Patients and physicians seek The Tele care medicine information system (TMIS) enable patients to gain admission to healthcare services over long distances. Patients and physicians seek

30 June 2026 in Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications
Verified
Journal Article
Closed access
SKS. Karthikeyan
S. Karthikeyan
HKHarish Kumar
Harish Kumar
.K.T. Ganesh Kumar
.T. Ganesh Kumar
.P.T. Poongodi
.T. Poongodi
The present work offers an architecture of a real-time intrusion detection system for IoT smart home gateway aiming at filling the research gap related to the contradiction between the high accuracy of deep-learning architectures and their application restrictions. The first task for this paper is to create a CNN-LSTM-Attention-based architecture ensuring not only high accuracy but also low-latency of operation in a streaming environment. The proposed methodology combines the use of convolutional layers for the local feature extraction, LSTM for sequence learning, and multi-head attention for the adaptive weight distribution in a four-thread pipeline processing the packet capturing, inference, alert generation, and flow management. Evaluation of the architecture is conducted on the NSL-KDD, CICIDS-2017 and TON_IoT datasets with the help of a set of preprocessing procedures, including the use of Min-Max normalization, RFECV feature selection and DeepSMOTE balancing for the class imbalance. The experimental results demonstrate the high accuracy of 99.61% on NSL-KDD, 99.21% on CICIDS-2017, and 98.84% on TONIoT. In the case of CICIDS-2017, the precision, recall, F1-score, and significantly decreased rate of false positives equal to 99.14%, 99.22%, 99.18%, and 0.43%, respectively, where the last one is 77% smaller than the baseline CNN, which makes the system extremely appropriate to be used in reducing alert fatigue in real-world applications. Also, the developed approach reaches an end-to-end latency of 0.37 ms per flow and an acceptable throughput of around 18,000 flows per second, meeting the criteria of real-time gateways under heavy traffic conditions. It has been shown via statistical significance tests that all obtained performance gains are statistically significant, where p<0.01. Overall, it can be stated that the study proves the importance of system-level optimizations and proper pipeline design over model complexity for building efficient, real-life, and high-performing IoT Intrusion Detection Systems.