@kalasalingam.ac.in
Associate Professor
kalasalingam academy of research and education
B.E M.E Ph.D
Computer Science, Computer Science Applications, Computer Engineering, Computer Vision and Pattern Recognition
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Amaranatha Sasthry S, Shyam Sundar R, Sri Ganesh M, Anand Kumar R, and Jayalakshmi M
IEEE
In the face of mounting challenges related to traffic congestion and road safety in modern urban areas, this paper introduces an innovative real-time traffic control system that leverages the capabilities of computer vision and embedded computing technologies. To achieve this, the system harnesses the power of OpenCV, a robust computer vision library, for the purpose of live vehicle detection using pre-trained cascade classifiers. By integrating cameras with Raspberry Pi devices, the system can capture real-time traffic footage, enabling the immediate analysis of traffic density and flow in multiple directions within the urban landscape. The core functionality of the system lies in its ability to process the captured video frames in real-time. By doing so, it can accurately detect vehicles, evaluate traffic flow conditions, and identify congested routes. This critical information is then employed to dynamically adjust traffic signals in response to the detected traffic conditions. These adjustments optimize traffic control measures, leading to the alleviation of congestion, and ultimately contributing to an enhancement in road safety. This paper provides a comprehensive exploration of the system architecture, the underlying algorithms, and integration details. Through this detailed discussion, the paper demonstrates the remarkable effectiveness of this approach in the realm of urban traffic management. By seamlessly combining computer vision and embedded computing technologies, this innovative traffic control system not only provides real-time insights into traffic conditions but also actively responds to these conditions to ensure a smoother and safer flow of vehicles through urban streets.
Redhya M and M. Jayalakshmi
IEEE
Parkinson’s disease (PD) is a neuro-degenerative disease caused due to breakdown of brain cells in the central-part of the nervous system. As symptoms of PD appear only after 60% or more of these cells are destroyed, early detection is a tedious task. Several studies have proven that PD can be effectively diagnosed from MR-Images (MRI). This paper provides an ensembled machine learning model, Stacked-EG model, for MRI-image based PD classification. The dataset was obtained from PPMI (Parkinson’s Progression Markers Initiative) online repository which included 134 PD subjects (80 males and 54 females) and 126 healthy patients (60 males and 66 females) around 60 years old on average. Feature extraction is performed through the application of filters, including wavelet transformations and intensity-based enhancements. Feature selection is conducted using optimised XGBoost model’s feature significance scores. For classification, a Stacking Classifier which combines the strengths of XGBoost, SVM, and a Random Forest classifier was used. The model obtained an accuracy of 98.41% in classifying PD from healthy controls.
Santosh Kumar Henge, Gnaniyan Uma Maheswari, Rajakumar Ramalingam, Sultan S. Alshamrani, Mamoon Rashid, and Jayalakshmi Murugan
MDPI AG
This article discusses the importance of cross-platform UX/UI designs and frameworks and their effectiveness in building web applications and websites. Third-party libraries (TPL) and plug-ins are also emphasized, as they can help developers quickly build and compose applications. However, using these libraries can also pose security risks, as a vulnerability in any library can compromise an entire server and customer data. The paper proposes using multi-authentication with specific parameters to analyze third-party applications and libraries used in cross-platform development. Based on multi-authentication, the proposed model will make setting up web desensitization methods and access control parameters easier. The study also uses various end-user and client-based decision-making indicators, supporting factors, and data metrics to help make accurate decisions about avoiding and blocking unwanted libraries and plug-ins. The research is based on experimentation with five web environments using specific parameters, affecting factors, and supporting data matrices.
Shreya Patchala, Bura Vijay Kumar, Kotha Chandrakala, Abbas Hameed Abdul Hussein, and M. Jayalakshmi
IEEE
Many screening methods have been analyzed by employing the modalities of Magnetic Resonance Imaging (MRI) and Electroencephalography (EEG) to diagnose epileptic seizures. Deep Learning (DL) is a branch of artificial intelligence, which is a broad field. Conventional DL could only perform as well as people who made the features by hand. On the other hand, DL fully automates both feature extraction and categorization. EEG is an often employed and noteworthy method for assisting in the diagnosis of epilepsy and examining the movement of the human brain. Since EEG signals are non-stationary, a patient's seizure patterns will change during various recording sessions. This study uses a Bidirectional Long Short-Term Memory (Bi-LSTM) model to detect seizures and brain tumours. Preprocessing the EEG signal, extracting preictal features, hyper-optimization using Mayfly Optimization (MO). Long-term EEG recordings from the EEG and ABIDE fMRI datasets are used in the evaluation. Many returned features are used by the Bi-LSTM model, such as frequency domain and temporal information between extracted EEG channels prior to classification. The proposed Bi-LSTM accomplishes a 99.12% accuracy rate, 98.25% precision, 99.06% recall, and 98.53% f1-measure which is superior when related to existing models.
D. Santhi, M. Carmel Sobia, and M. Jayalakshmi
CRC Press
R. Raja Selvi, M. Shruthi, G. Nithya, S. Kalaiselvi, M. Jayalakshmi, and V. Gomathi
AIP Publishing
Ksn Sushma, M. Jayalakshmi, and Tapas Guha
IEEE
Data security has become a major issue as digitalization across the world is rapidly increasing. Personal information of the user is extracted and exploited using one of the cyber-attack known as phishing. To detect phishing, and to avoid the user from visiting a non-legitimate website, a method is proposed in this paper. This method uses unique features of Uniform Resource Locator which can differentiate between legitimate and non-legitimate websites. To classify websites, Random forest and Support Vector Machines are two machine learning methods that are employed in this work. The internet offers a wealth of information that may be accessed. Because of the rapid development of technology, an unavoidable dependency on the Internet has emerged in all aspects of life. As the number of apps that run on the Internet continues to rise, there is also a rising worry of threats and a need to address problems that are connected to security. There are numerous web application threats in the Internet domain. These threats aim to either steal sensitive information from users, change the database of web servers, or undermine the credibility of a particular web application. There are many web application threats in the Internet domain. These threats include: One of the most significant dangers to information security is the character. Attacks that Deny Service to Others Session Hijacking attempts The CrossSite Scripting Language XSS Phishing and Buffer Overflow are Two Common Scams Phishing is one of these methods, and it involves tricking people into giving sensitive information such as usernames and passwords, credit card details, and sensitive bank information by way of email spoofing, instant messaging, or fake web sites whose look and feel give the appearance of a legitimate one. Examples of such information include usernames and passwords for online accounts, credit card details, and sensitive bank information. Important forms of phishing include deceptive phishing, malware-based phishing, host file poisoned content, and others. injection phishing through man-in-the-middle attacks, phishing via search engines, and phishing via social engineering Due to the fact that phishing may cause significant losses, it is necessary to implement new line processes in order to identify and stop phishing attacks. At this time, there are a number of techniques for detecting phishing. Some of these methods include blacklisting and whitelisting, visual resemblance, content-based approaches, and detecting phishing.
R. Muthukkumar, Lalit Garg, K. Maharajan, M. Jayalakshmi, Nz Jhanjhi, S. Parthiban, and G. Saritha
PeerJ
Background The energy-constrained heterogeneous nodes are the most challenging wireless sensor networks (WSNs) for developing energy-aware clustering schemes. Although various clustering approaches are proven to minimise energy consumption and delay and extend the network lifetime by selecting optimum cluster heads (CHs), it is still a crucial challenge. Methods This article proposes a genetic algorithm-based energy-aware multi-hop clustering (GA-EMC) scheme for heterogeneous WSNs (HWSNs). In HWSNs, all the nodes have varying initial energy and typically have an energy consumption restriction. A genetic algorithm determines the optimal CHs and their positions in the network. The fitness of chromosomes is calculated in terms of distance, optimal CHs, and the node's residual energy. Multi-hop communication improves energy efficiency in HWSNs. The areas near the sink are deployed with more supernodes far away from the sink to solve the hot spot problem in WSNs near the sink node. Results Simulation results proclaim that the GA-EMC scheme achieves a more extended network lifetime network stability and minimises delay than existing approaches in heterogeneous nature.
N.B. Prakash, M. Murugappan, G.R. Hemalakshmi, M. Jayalakshmi, and Mufti Mahmud
Elsevier BV
M. Jayalakshmi, K. Maharajan, K. Jayakumar, and G. Visalaxi
Wiley
M. Jayalakshmi, Lalit Garg, K. Maharajan, K. Jayakumar, Kathiravan Srinivasan, Ali Kashif Bashir, and K. Ramesh
Computers, Materials and Continua (Tech Science Press)
M. Jayalakshmi and V. Gomathi
Springer Science and Business Media LLC
K. Pavithra and M. Jayalakshmi
IEEE
India is the largest country where most of the human lives depends upon agriculture. Due to the shortage of water facilities, the yield of irrigation is affected. To reduce this shortage of water, different types of methods are invented to supply the correct level of water to the agricultural lands. Here precision irrigation system is used to maintain the perfect water supply from this method wastage of water can be reduced. By using this technique, we can achieve better harvesting results. The total amount of water for lands and plants can be calculated by using the PH value. To detect the PH level, a random forest algorithm is used. There is a direct relationship between the number of plants in the land and the results it can get. It improves the prediction accuracy.
M. Jayalakshmi and V. Gomathi
Springer Science and Business Media LLC
A. Vasukidevi, M. Jayalakshmi, and V. Gomathi
IEEE
A wireless sensor network is used to sense the infected person's body condition and transmits the collected data to the database. The details including temperature, heartbeat, and blood pressure are sensed using appropriate sensors and stored in the servers. The details are encrypted using Paillier and ElGamal key cryptosystems for providing high security. The server is helpful in providing querying service to users. The encrypted data is fetched from the server and decrypted before authentication. To prevent the patient data from the attackers, new protocol is proposed, where the patient details are converted into numerical form. This numerical form is difficult to understand by the attackers. Then the numerical values are sent to servers through secure channels. The patient data are securely distributed by employing the Paillier and ElGamal cryptosystems. This project focuses on secure storage and communication. The secure data will be implemented based on cryptographic principle.
M. JayaLakshmi and V. Gomathi
IEEE
This paper describes about the design and implementation of a water leakage monitoring and detection system to monitor and detect leak with the help of wireless networked sensors. The objective of this enhanced system is to detect possible underground water leakage for residential water pipes that are monitored from a PC. Therefore, a robust and reliable Wireless sensor network which composes small Printed Circuit Boards (PCB), data from remote sensors of different types (acoustic, pressure, temperature, flow rate, etc.) are collected and monitored on a PC to detect the exact leakage position. Once a leak is detected, the water utility must take corrective action to minimize water losses in the water distribution system. Thus the proposed system will be used to save water and reduces the replacing cost.