Iris-Controlled Wheelchair P. Mohanraj, T. Rajasekar, S. Vigneshwaran, G. Sariga, D. Vishnupriya Signals and Communication Technology, 2025
Syngo Carbon Deployment and Enhanced Cancer Analysis T. Rajasekar, S. Sanjay Pandi, P. Mohanraj, B. Kalaimathi Proceedings of the 3rd International Conference on Applied Artificial Intelligence and Computing Icaaic 2024, 2024 This research study presents two significant contributions to the healthcare domain. A hospital management system features admin and doctor logins for patient and employee management, pharmaceutical inventory and accounting. Patient data includes name, address, contact, age and ailment with employee login details are recorded. Doctors access surgery and vendor medicine details, patient additions and lab results. A cancer stage detection software analyzes the pelvis region images taken at detection, 3-months post-chemotherapy and 6-months post-detection. Using filtered back projection, it evaluates chemotherapy response and cancer cell reduction accurately. This system enhances hospital operations, patient care and treatment monitoring.
LEXIBOT: Gesture Controlled Personal Assistant Companion T Rajasekar, P Mohanraj, P Madhumitha, S Nanthitha, S Naveena Proceedings of the 5th International Conference on Smart Electronics and Communication Icosec 2024, 2024 This research work introduces an IoT-based robot control system featuring accelerometer-based gesture control or voice commands. The system comprises a transmitter section with an accelerometer or voice recognition module connected with a NodeMCU, and a receiver section consisting of a NodeMCU connected to an L298 motor driver controlling the robot’s movement. In accelerometer mode, hand gestures are detected and translated into commands for forward, backward, left, and right movements, while in voice mode, commands are recognized by a mobile app and transmitted to the NodeMCU, enabling hands-free control. This integrated approach offers intuitive, tactile control through gestures or convenient verbal commands, facilitating enhanced human-robot interaction across various applications.
Wearable Device for Visually Impaired using Deep Learning P Mohanraj, T Rajasekar, N Sivaelango, V Sri Karthickraja, N Vignesh Proceedings of the 3rd International Conference on Applied Artificial Intelligence and Computing Icaaic 2024, 2024 This research introduces a novel wearable technology specifically to assist blind and visually impaired people in navigating their surroundings safely and independently. The proposed device integrates Raspberry Pi 4, cutting-edge object detection algorithms, voice output, and sensor feedback systems to provide comprehensive ambient awareness. Real-time video feeds captured by a Raspberry Pi camera module enable robust item detection and categorization. The Raspberry Pi 4 coupled with efficient deep learning algorithms facilitates real-time inference on resourceconstrained platforms. Additionally, water and ultrasonic sensors have been seamlessly integrated to detect wet surfaces and obstacles, respectively. Upon detecting hazards, the proposed system delivers prompt haptic and audio feedback, allowing users to avoid obstacles and make informed decisions. Extensive field testing and user trials validate the efficacy and versatility of this wearable technology in diverse real-world scenarios. This innovation enables a significant advancement in assistive technology, offering enhanced freedom and safety to visually impaired individuals. Future research directions may focus on algorithm optimization, sensor integration, and user-centered design enhancements to augment user experience and functionality.
Internet of Things (IoT) based Smart Inhaler T Rajasekar, P Mohanraj, B Kalaimathi, R Saravanan, V.S Vishnu Varadhan Proceedings of the 5th International Conference on Smart Electronics and Communication Icosec 2024, 2024 The Smart Inhaler system presented in this research study integrates two NodeMCU devices to enhance the monitoring and management of respiratory health remotely. The core functionality involves the first NodeMCU unit equipped with sensors for detecting Heart rate and blood oxygen saturation (SpO2). This NodeMCU collects the vital signs data, communicates with the cloud, and displays the information on an OLED screen. On the other end, a second NodeMCU unit acts as the inhaler controller. It retrieves the SpO 2 data from the cloud and utilizes a Touch sensor interface. When triggered by the received data, the system initiates the inhaler, providing timely and targeted respiratory support to the patient. This two-node architecture enables Real-time monitoring of crucial health metrics and facilitates a responsive intervention through the inhaler subsystem. The integration of IoT technologies, cloud connectivity, and sensor data processing in this Smart Inhaler system showcases a comprehensive approach to remote patient monitoring and personalized healthcare.
Smart Public Transport Monitoring and Demand Analysis with Machine Learning Mohanraj. P, Rajasekar. T, Jeeva. V, Jerrin. J, Nithinkrishna. P 5th International Conference on Electronics and Sustainable Communication Systems Icesc 2024 Proceedings, 2024 In India, the passengers, who travel through public transport are unable to find out the available buses on the route. It is important to provide consumers with live location and departure information for buses to ease their travel. It might be challenging for those waiting for a bus at the bus stop to know the status of the particular bus on that route when there is a delay in the vehicle due to traffic or other issues. There are some systems which is costly and does not predict number bus required. The aim of this research study is to develop an embedded system that can display the current location of the selected bus and passenger count in real time. This will make it much simpler for individuals who need to plan their travels by providing them with real information about the bus's status through an application, and it will also use simple machine-learning algorithms to calculate the number of buses needed for that particular route with the account of the demand of each hour in a day and also to have a less system power consumption by the ML modal trained.
Adaptive Traffic Congestion Control Approach with Emergency Vehicle Protocol Rajasekar T, Mohanraj P, Abishek R K, Haries M, Dhivya Dharshini P Proceedings of the 8th International Conference on Communication and Electronics Systems Icces 2023, 2023 Traffic Congestion and an overabundance of cars on the streets are common sights in most of the major Indian cities. Smart road dividers can be expensive to install and maintain, especially compared to traditional concrete barriers or other passive safety measures. This cost may be prohibitive for some cities or regions. Smart road dividers use advanced technologies such as sensors, cameras, and communication systems to detect and respond to changing traffic conditions. This technical complexity can pose challenges for maintenance and repair. This initiative serves as an example of a method for managing traffic signals based on the volume of traffic and the presence of pedestrians. Python Language is used to count the quantity of cars using image processing. Data from both of these devices are sent to the Raspberry Pi. Raspberry Pi 3 used in this system because the execution of output is much accurate and faster compared to other microcontroller. The system received a signal from emergency vehicles based on Radio frequency (RF) transmission, and the Raspberry Pi microprocessor was used to change the sequence back to the regular sequence before the emergency mode was triggered by the system. This strategy will reduce collisions that frequently happen at traffic signal intersections where other cars gathered to give an emergency vehicle a special route. The traffic light management system for emergency vehicles in this project effectively analyzed and applied wireless communication, or radio frequency (RF) transmission. The project's prototype operates on a frequency of 434 MHz, switching to emergency mode when a traffic light is passed by an emergency vehicle and back to regular mode when the emergency mode was not activated. The vehicle counts and the detection of the ambulance of the proposed system is monitored in cloud. In the future, this prototype system can be enhanced by implementing in highways by extending the moving length of the divider.
Reducing Driver's Range Anxiety for Electric Vehicle using Machine Learning Rajasekar T, Anu Varshini R P, Mohanraj P, Hemmasri K, Adel Mariam A Proceedings of the 8th International Conference on Communication and Electronics Systems Icces 2023, 2023 Vehicles are an indispensable part of peoples’ living be it for commercial or personal uses. The Vehicle that is powered by an Electric Motor that can be recharged when power drained and produces zero emission while driving which makes it an environmentally friendly option. The adoption of Electric Vehicles is still hampered by a number of problems, range anxiety being one of them. Range anxiety is the fear that an Electric Vehicle (EV) driver could have if their battery runs out and it is unable to reach a charging station. The proposed system overcomes the driver’s anxiety of battery’s performance by effective machine learning prediction and displays the amount of energy consumed earlier, before the start of the journey. The driver can compare and conclude whether the place can be reached before battery runoff. The system is compared with three different algorithms - multilinear regression algorithm, polynomial regression algorithm, random forest algorithm .The reason for choosing these three algorithms is ,it can process one dependent variables with more than one independent variables respectively .The innovative method used here scrutinised the consumption of battery into three mode sport mode ,eco mode and city mode.each mode gives Thus the system provides the potential of accuracy as 81.11% and to be a useful tool for electric vehicle owners, as it provides personalised energy consumption estimates and route planning capabilities based on the specific journey details
Pest Identification and Control in Paddy Plants using Ml with an Optimised Activation Function Mohanraj. P, Rajasekar. T, Aksha Varshini. B, Charumathi. V. S, Dhaarani. S Proceedings of the 2023 2nd International Conference on Augmented Intelligence and Sustainable Systems Icaiss 2023, 2023 The rice plant, one of the most extensively cultivated crops in India, suffers from a number of diseases at various stages of its growth. Farmers find it incredibly difficult to manually diagnose these diseases properly because of their lack awareness. Caused by many types, Such as; Fungi, Bacteria and Viruses and Pest. In Image processing technology many algorithms are used to classify rice leaf diseases. In the southern regions of India, paddy is the primary source of food. It has an impact on the nation's infrastructure. Its main illnesses were found in paddy field crops, which also significantly impair its profitability. In this research study, the automatic diagnosis and classification of different paddy leaf diseases using an image processing system was demonstrated. Deep Learning can be used for pest identification in paddy plants. Using a collection of pictures of paddy plants with and without pests, a Convolutional Neural Network (CNN) can be trained in this application. The network can learn to identify the characteristic features of pests in the images, such as shape, color, size, and texture. Once the network is trained, it can be used to classify new images of paddy plants as containing pests or not. By using deep learning, it is possible to automate the process of pest identification in paddy plants, reducing the time and effort required for manual inspection and increasing the accuracy of the results. So this paper, Classify the normal leaf and disease affected leaf using CNN algorithm. Software has been developed as part of the proposed system to automatically reduce the paddy leaf disease. The proposed system is based on the Deep Learning method. All the process is done by using python Programming language. In the proposed system tkinter is the front end and Python is back end.
Design of a Dense Layered Network Model for Epileptic Seizures Prediction with Feature Representation Summia Parveen, S. A. Siva Kumar, P. MohanRaj, Kingsly Jabakumar, R. Senthil Ganesh International Journal of Advanced Computer Science and Applications, 2022 —Epilepsy is a neurological disorder that influences about 60 million people all over the world. With this, about 30% of the people cannot be cured with surgery or medications. The seizure prediction in the earlier stage helps in disease prevention using therapeutic interventions. Certain studies have sensed that abnormal brain activity is observed before the initiation of seizure which is medically termed as a pre-ictal state. Various investigators intend to predict the baseline for curing the preictal seizure stage; however, an effectual prediction model with higher specificity and sensitivity is still a challenging task. This work concentrates on modelling an efficient dense layered network model (DLNM) for seizure prediction using deep learning (DL) approach. The anticipated framework is composed of pre-processing, feature representation and classification with support vector based layered model (dense layered model). The anticipated model is tested for roughly about 24 subjects from CHBMIT dataset which outcomes in attaining an average accuracy of 96% respectively. The purpose of the research is to make earlier seizure prediction to reduce the mortality rate and the severity of the disease to help the human community suffering from the disease.
Review of Vedic Multiplier Using Various Full Adders A Abdul Hayum, S Chinnapparaj, G Sujatha, G Thamarai Selvi, Mohd Naved, P Mohanraj Proceedings 5th International Conference on Computing Methodologies and Communication Iccmc 2021, 2021
Adaptive filter using multiprecision multiplier International Journal of Applied Engineering Research, 2015
