Kumaran Pazhamalai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Kumaran Pazhamalai has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Chitrakala Sakthivel
Department of Computer Science and Engineering, College of Engineering, Guindy, Anna University, Chennai, India
6 shared publications R. Ashoka Rajan
Computer Science and Engineering, School of Computing, SASTRA Deemed to be University, Thanjavur, India
2 shared publications- Rajeswari Sridhar
Computer science & Engineering, National Institute Technology, Tiruchirappalli, India
2 shared publications Ashoka Rajan R
1 shared publicationChitrakala S.
Department of Computer Science and Engineering, College of Engineering, Guindy Anna University Chennai India
1 shared publicationG Gracia Michelle
Department of Computer Science and Engineering, College of Engineering, Guindy, Anna University, Chennai, Tamil Nadu, India
1 shared publication
Latest Publications
- 5 December 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (17) , 1411
Enhancing Underwater Imagery: A Deep Attention-Based deWater Network (DAW-Net) Approach
- 5 December 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (17) , 1411
Enhancing Underwater Imagery: A Deep Attention-Based deWater Network (DAW-Net) Approach
- 1 March 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (02) , 023030-20
ELDA: enhanced linear discriminant analysis for cashew crop disease detection using precision agriculture
This work provides a method for reducing dimensionality using a median-based discriminatory analysis, which considers the adverse impact on the class mean caused by outliers, referred to as an enhanced linear discriminant analysis (ELDA). ELDA is based on a point-to-median distance, used as a measure metric to describe the within-class and between-class median scatter. The standard LDA algorithm may not be effective when the class distribution is uneven, and the limited sample size for inequality datasets and outliers persists. To address all issues, the proposed ELDA method is applied for cashew leaf disease identification on the cashew crop disease database (CCDDB), and it is more accurate at identifying and classifying cashew leaf illnesses in different sets of training and testing samples on three datasets. Research exhibits greater accuracy on CCDDB with 97.7%, on image database of plant disease symptoms dataset with 99.8%, and on dataset for crop pest and disease detection with 81.7%.
- 1 March 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (02) , 023030-20
ELDA: enhanced linear discriminant analysis for cashew crop disease detection using precision agriculture
This work provides a method for reducing dimensionality using a median-based discriminatory analysis, which considers the adverse impact on the class mean caused by outliers, referred to as an enhanced linear discriminant analysis (ELDA). ELDA is based on a point-to-median distance, used as a measure metric to describe the within-class and between-class median scatter. The standard LDA algorithm may not be effective when the class distribution is uneven, and the limited sample size for inequality datasets and outliers persists. To address all issues, the proposed ELDA method is applied for cashew leaf disease identification on the cashew crop disease database (CCDDB), and it is more accurate at identifying and classifying cashew leaf illnesses in different sets of training and testing samples on three datasets. Research exhibits greater accuracy on CCDDB with 97.7%, on image database of plant disease symptoms dataset with 99.8%, and on dataset for crop pest and disease detection with 81.7%.
- 23 May 2023
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 17 (7) , 3323-3330
Early detection and control of anthracnose disease in cashew leaves to improve crop yield using image processing and machine learning techniques
- 23 May 2023
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 17 (7) , 3323-3330
Early detection and control of anthracnose disease in cashew leaves to improve crop yield using image processing and machine learning techniques
- 25 February 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (6) , 8943-8982
Topic adaptive sentiment classification based community detection for social influential gauging in online social networks
- 25 February 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (6) , 8943-8982
Topic adaptive sentiment classification based community detection for social influential gauging in online social networks
- 7 September 2021
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 34 (3) , e6597
A novel mathematical modeling in shift in emotion for gauging the social influential in big data streams with hybrid sarcasm detection
Summary: Online social network is a platform that plays an essential role in identifying the emotional values of user‐generated content such as blogs, posts, and comments along with their influential factors. Especially on Twitter, network users are growing worldwide day by day and creating a massive amount of data that is not analyzed effectively in a quick way. Identifying the most influential persons on the social network is also a challenging task over the wide range of real‐time applications like recommendation systems. Now, to handle these situations, this article proposes a novel approach for prediction of information diffusion that includes emotion recognition with sarcasm detection based influence spreader identification (PID‐ERSDISI). The proposed method uses the user‐generated posts for emotion recognition in tandem with sarcasm detection both implicitly and explicitly. This approach helps to gauge the leverage that influences spreaders and also enhances the prediction accuracy of information diffusion in a better way. The implementation of the proposed work executed their task one after another in the following way, namely, sarcasm detection, emotional‐level computation, breakpoint computation, breakpoint validation, influential user generation, and information diffusion. After the successful implementation of this proposed PID‐ERSDIS, it produced prominent results against other state‐of‐art methods.
- 7 September 2021
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 34 (3) , e6597
A novel mathematical modeling in shift in emotion for gauging the social influential in big data streams with hybrid sarcasm detection
Summary: Online social network is a platform that plays an essential role in identifying the emotional values of user‐generated content such as blogs, posts, and comments along with their influential factors. Especially on Twitter, network users are growing worldwide day by day and creating a massive amount of data that is not analyzed effectively in a quick way. Identifying the most influential persons on the social network is also a challenging task over the wide range of real‐time applications like recommendation systems. Now, to handle these situations, this article proposes a novel approach for prediction of information diffusion that includes emotion recognition with sarcasm detection based influence spreader identification (PID‐ERSDISI). The proposed method uses the user‐generated posts for emotion recognition in tandem with sarcasm detection both implicitly and explicitly. This approach helps to gauge the leverage that influences spreaders and also enhances the prediction accuracy of information diffusion in a better way. The implementation of the proposed work executed their task one after another in the following way, namely, sarcasm detection, emotional‐level computation, breakpoint computation, breakpoint validation, influential user generation, and information diffusion. After the successful implementation of this proposed PID‐ERSDIS, it produced prominent results against other state‐of‐art methods.
- 22 October 2020
- Published by Oxford University Press (OUP) in The Computer Journal
- Vol. 64 (3) , 391-407
Prediction of Social Influence for Provenance of Misinformation in Online Social Network Using Big Data Approach
Online social networks (OSNs) is a platform that plays an essential role in identifying misinformation like false rumors, insults, pranks, hoaxes, spear phishing and computational propaganda in a better way. Detection of misinformation finds its applications in areas such as law enforcement to pinpoint culprits who spread rumors to harm the society, targeted marketing in e-commerce to identify the user who originates dissatisfaction messages about products or services that harm an organizations reputation. The process of identifying and detecting misinformation is very crucial in complex social networks. As misinformation in social network is identified by designing and placing the monitors, computing the minimum number of monitors for detecting misinformation is a very trivial work in the complex social network. The proposed approach determines the top suspected sources of misinformation using a tweet polarity-based ranking system in tandem with sarcasm detection (both implicit and explicit sarcasm) with optimization approaches on large-scale incomplete network. The algorithm subsequently uses this determined feature to place the minimum set of monitors in the network for detecting misinformation. The proposed work focuses on the timely detection of misinformation by limiting the distance between the suspected sources and the monitors. The proposed work also determines the root cause of misinformation (provenance) by using a combination of network-based and content-based approaches. The proposed work is compared with the state-of-art work and has observed that the proposed algorithm produces better results than existing methods.
- 22 October 2020
- Published by Oxford University Press (OUP) in The Computer Journal
- Vol. 64 (3) , 391-407
Prediction of Social Influence for Provenance of Misinformation in Online Social Network Using Big Data Approach
Online social networks (OSNs) is a platform that plays an essential role in identifying misinformation like false rumors, insults, pranks, hoaxes, spear phishing and computational propaganda in a better way. Detection of misinformation finds its applications in areas such as law enforcement to pinpoint culprits who spread rumors to harm the society, targeted marketing in e-commerce to identify the user who originates dissatisfaction messages about products or services that harm an organizations reputation. The process of identifying and detecting misinformation is very crucial in complex social networks. As misinformation in social network is identified by designing and placing the monitors, computing the minimum number of monitors for detecting misinformation is a very trivial work in the complex social network. The proposed approach determines the top suspected sources of misinformation using a tweet polarity-based ranking system in tandem with sarcasm detection (both implicit and explicit sarcasm) with optimization approaches on large-scale incomplete network. The algorithm subsequently uses this determined feature to place the minimum set of monitors in the network for detecting misinformation. The proposed work focuses on the timely detection of misinformation by limiting the distance between the suspected sources and the monitors. The proposed work also determines the root cause of misinformation (provenance) by using a combination of network-based and content-based approaches. The proposed work is compared with the state-of-art work and has observed that the proposed algorithm produces better results than existing methods.
- 2 April 2020
- Published by SAGE Publications in Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
- Vol. 38 (5) , 6437-6444
Multi-biometric cryptosystem using graph for secure cloud authentication
A recent trend of the information technology is cloud computing technology where many complex tasks are simplified with increased speed and low cost. However, cloud authentication plays a crucial role once all the data’s are uploaded in the cloud. In this paper, multi-biometric template security based on generation of unique graph is proposed to ensure a safe and secured cloud authentication mechanism. To overcome the vulnerability issues of traditional password and token based authentication methods, in this work, a multi-biometric system is proposed. The left, right fingerprint and palm print multi modal traits are given as input to the system. After preprocessing, all the features are combined to generate a weighted graph called as branching factor graph. In the end, the node and edge values of the branching factor graph will be stored in the cloud database. Experimental study shows that the proposed method achieved a very low equal error rate than the other existing works.
- 2 April 2020
- Published by SAGE Publications in Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
- Vol. 38 (5) , 6437-6444
Multi-biometric cryptosystem using graph for secure cloud authentication
A recent trend of the information technology is cloud computing technology where many complex tasks are simplified with increased speed and low cost. However, cloud authentication plays a crucial role once all the data’s are uploaded in the cloud. In this paper, multi-biometric template security based on generation of unique graph is proposed to ensure a safe and secured cloud authentication mechanism. To overcome the vulnerability issues of traditional password and token based authentication methods, in this work, a multi-biometric system is proposed. The left, right fingerprint and palm print multi modal traits are given as input to the system. After preprocessing, all the features are combined to generate a weighted graph called as branching factor graph. In the end, the node and edge values of the branching factor graph will be stored in the cloud database. Experimental study shows that the proposed method achieved a very low equal error rate than the other existing works.
- 1 November 2019
- Published by American Scientific Publishers in Journal of Computational and Theoretical Nanoscience
- Vol. 16 (11) , 4883-4888
Template Protection Using Multi Biometric Web Modulo Graph
To develop a complete biometric authentication system, security is highly needed. Even though there are several methods for storing fingerprint templates, they are compromised by the attacker leaving it as an unprotected system. In this paper, a novel method is proposed for protecting biometrics through an user defined graph named Web Modulo Graph. Feature vectors are extracted from the Left Fingerprint, Right Fingerprint and Palm Print during the enrollment process. The captured information from the biometrics are combined and stored in Web Modulo Graph where the insertion and traversal of feature vectors are unknown to the attacker. So even if the database or the graph structure is stolen by the attacker the correct sequence cannot be obtained. In this case, guessing the correct sequence is not almost possible as user defined graph is used and the system can achieve this with an Equal Error Rate (EER) of 4.8%. After various analyses, the proposed system is found to have high computational hardness.
- 1 November 2019
- Published by American Scientific Publishers in Journal of Computational and Theoretical Nanoscience
- Vol. 16 (11) , 4883-4888
Template Protection Using Multi Biometric Web Modulo Graph
To develop a complete biometric authentication system, security is highly needed. Even though there are several methods for storing fingerprint templates, they are compromised by the attacker leaving it as an unprotected system. In this paper, a novel method is proposed for protecting biometrics through an user defined graph named Web Modulo Graph. Feature vectors are extracted from the Left Fingerprint, Right Fingerprint and Palm Print during the enrollment process. The captured information from the biometrics are combined and stored in Web Modulo Graph where the insertion and traversal of feature vectors are unknown to the attacker. So even if the database or the graph structure is stolen by the attacker the correct sequence cannot be obtained. In this case, guessing the correct sequence is not almost possible as user defined graph is used and the system can achieve this with an Equal Error Rate (EER) of 4.8%. After various analyses, the proposed system is found to have high computational hardness.