MEIVEL S
@vidwan.inflibnet.ac.in/profile
Assistant Professors and ECE Department
M.KUMARASAMY COLLEGE OF ENGINEERING, KARUR
RESEARCH, TEACHING, or OTHER INTERESTS
Electrical and Electronic Engineering, Aerospace Engineering, Industrial and Manufacturing Engineering, Engineering
Scopus Publications
Scopus Publications
K. Ambujam, A. Sridevi, S. Meivel, and T. R. Chinnusamy
Springer Science and Business Media LLC
D. Faridha Banu, N. Kumaresan, K. Geetha devi, S. Priyanka, G. Swarna Shree, A. Roshan, and S. Meivel
Springer Nature Singapore
S. Meivel, S. K. Nagaharipriya, P. Priyankadevi, and S. Sangavi
IEEE
This work introduces novel approaches to accelerate the production of deterministic test patterns for VLSI devices. These methods reduce the number of backtracking while requiring little processing effort, therefore improving the PODEM algorithm. This is accomplished by assigning additional signal lines to the relevant signals, spotting errors earlier, and eliminating extra effort during test generation. These methods have been included into ATOM, a sophisticated ATPG device for combinational circuits. The test generation results for the benchmark circuits' ATOM as well as full scan versions showed how effective these strategies were at improving performance ATOM found all tested problems in a short period of time and proved that all overlapping faults were excessive with minimum backtracking
Meivel S, Sundar G, Yaswanth K M, and Yogeshwaran S
IEEE
Safety helmets may provide sufficient protection for construction workers. But employees frequently take off their helmets due to discomfort and a lack of security knowledge, leaving them vulnerable. Workers who don't wear protective headgear are more likely to be hurt in accidents involving falling items (including humans) and vertically falling things. A fast and reliable safety helmet detector is, however, desperately needed. Observing workers to ensure they are wearing protective headgear is a crucial aspect of site management. However, the commonplace manual monitor calls for a much work, and it's tough to get people to adopt new methods for installing sensors in safety helmets. This study provides a Deep Learning (DL) based safety helmet recognition method that is both fast and accurate. Using deep learning, a system has been developed to detect hard hats in construction zones. To solve this problem, the SSD-Mobile Net technique uses convolutional neural networks. Photographs of safety helmets taken either manually from a company's video surveillance system or automatically through a web crawler may be made available to the general public. The picture set includes a training, a validation, and a test set. Our findings show that a deep learning model built using the SSD-Mobile Net method can reliably identify potentially hazardous behaviours on a construction site, such as the removal of a hard cap.
S. Meivel, S. Maheswari, and D. Faridha Banu
Springer International Publishing
S. Meivel, S. Maheswari, and D. Faridha Banu
Springer International Publishing
S. Meivel, S. Elakkiya, V. Kartheeswari, and K. V. Preethika
Springer Nature Singapore
S. Meivel, Nidhi Sindhwani, S. Valarmathi, G. Dhivya, M. Atchaya, Rohit Anand, and Sudhanshu Maurya
Springer Nature Singapore
S. Meivel and S. Maheswari
Springer Science and Business Media LLC
S. Meivel, S. Maheswari, and D. Faridha Banu
Springer International Publishing
S. Meivel, S. Maheswari, and D. Faridha Banu
Springer International Publishing
Faridha Banu D, Nidhi Sindhwani, Sasi G, Kaleel Rahuman A, and Meivel S
IEEE
Diabetic retinopathy is a vision loss and blindness disorder caused by diabetes. It affects the retina's blood vessels. The most prevalent retinal illnesses are glaucoma and diabetic retinopathy (DR) both of which are important causes of blindness. In addition to causing diabetic retinopathy (DR), longterm hyperglycemia also produces retinal lesions, exudates, microaneurysms, and haemorrhages that can result in blindness. Diabetes mellitus affects [1] an estimated 210 million people worldwide, with 10-18% having or developing the disease. Therefore, accurate and prompt DR identification is essential to avoiding DR and potential visual loss. Glaucoma is frequently associated with increased vitreous fluid pressure in the eye, but this is not always the case. Glaucoma is thought to be the world's second leading cause of blindness, after cataract. This result shows that fuzzy acceptance analysis of glaucoma and Diabetic retinopathy [2].
S. Meivel, C. Praghadeesh, A. Ravinder, and D. Sibisaran
IEEE
Every school, college, and university maintains a record of each student's attendance. Faculty are required to retain accurate and current attendance records. Because it takes a long time to organize records and determine each student's average attendance, the manual attendance record system is inefficient. As a result, a system for organizing student records and calculating average attendance is required The proposed system should be able to save student attendance records in digital format, making attendance management easier. Even in the twenty-firs t century, students' attendance is recorded on attendance forms presented in the classroom by staff members, which takes time and is completely manual. Despite the fact that RFID-based and face recognition-based systems have been shown, they are implemented separately. A RFID reader is integrated with a Face recognition system in the suggested system for student attendance. RFID readers, as well as facial recognition cameras, would be placed throughout campus and in classes. When a student walks onto campus, the reader communicates their id to the server, allowing them to be easily tracked After that, using a deep learning approach known as HAAR Cascade and Neural network algorithm, recognize the face from a real-time camera and match it to a database. This strategy ensures that the attendance records of the pupils are stored correctly and efficiently. As a result, the system will generate a list of kids who have been assigned to detention. It's a small-scale automated programme that's easy to operate, redeemable in time, and dependable.
K. Indira Devi, S. Meivel, K. Ranjit Kumar, and J. Vijayamenaka
Elsevier BV
S. Meivel, K. Indira Devi, S. Uma Maheswari, and J. Vijaya Menaka
Elsevier BV
S. Meivel, K. Indira Devi, T. Muthamil Selvam, and S. Uma Maheswari
Elsevier BV
S. Meivel, Nidhi Sindhwani, Rohit Anand, Digvijay Pandey, Abeer Ali Alnuaim, Alaa S. Altheneyan, Mohamed Yaseen Jabarulla, and Mesfin Esayas Lelisho
Hindawi Limited
The drones can be used to detect a group of people who are unmasked and do not maintain social distance. In this paper, a deep learning-enabled drone is designed for mask detection and social distance monitoring. A drone is one of the unmanned systems that can be automated. This system mainly focuses on Industrial Internet of Things (IIoT) monitoring using Raspberry Pi 4. This drone automation system sends alerts to the people via speaker for maintaining the social distance. This system captures images and detects unmasked persons using faster regions with convolutional neural network (faster R-CNN) model. When the system detects unmasked persons, it sends their details to respective authorities and the nearest police station. The built model covers the majority of face detection using different benchmark datasets. OpenCV camera utilizes 24/7 service reports on a daily basis using Raspberry Pi 4 and a faster R-CNN algorithm.
S. Meivel and S. Maheswari
Springer Science and Business Media LLC
Nidhi Sindhwani, Rohit Anand, Meivel S., Rati Shukla, Mahendra Yadav, and Vikash Yadav
European Alliance for Innovation n.o.
INTRODUCTION: In recent years, deep learning techniques have been made to outperform the earlier state-of-the-art machine learning techniques in many areas, with one of the most notable cases being computer vision. Deep learning is also employed to train the neural networks with the images and to perform the various tasks such as classification and segmentation using several different models. The size and depth of current deep learning models have increased to solve certain tasks as these models provide better accuracy. As pre-trained weights may be used for further training and prevent costly computing, transfer learning is therefore of vital importance. A brief account is given of their history, structure, benefits, and drawbacks, followed by a description of their applications in the different tasks of computer vision, such as object detection, face recognition etc. OBJECTIVE:. The purpose of this paper is to train a deep neural network to properly classify the images that it has never seen before, define techniques to enhance the efficiency of deep learning and deploy deep neural networks in various applications. METHOD: The proposed approach represents that after the reading of images, 256x256 pixel image’s random parts are extracted and noise, distortion, flip, or rotation transforms are applied. Multiple convolution and pooling steps are applied by controlling the stride lengths. RESULT: Data analysis and research findings showed that DNN models have been implemented in three main configurations of deep learning: CNTK, MXNet and TensorFlow. The proposed work outperforms the previous techniques in predicting the dependent variables, learning rate, image count, image mean, performance analysis of loss rate and learning rate during training, performance Analysis of Loss with respect to Epoch for Training, Validation and Accuracy. CONCLUSION: This research encompasses a large variety of computer applications, from image recognition and machine translation to enhanced learning. DNN models have been implemented in three main configurations of deep learning: CNTK, MXNet and TensorFlow. Extensive research has been conducted using the various deep architectures such as AlexNet, InceptionNet, etc. To the best of authors’ knowledge, this is the first work that presents a quantitative analysis of the deep architectures mentioned above.
S. Meivel and S. Maheswari
IEEE
Production in agriculture is not sufficient in today’s world. Therefore, we need to increase production to equalize the needs. However, due to the development in various fields, the human source for working and maintaining the cultivation land with proper consistency is insufficient. When it comes to Indian Agriculture System, the climatic environment is isotropic and there is a lack in the usage of agriculture assets. The irrigation system, which is controlled manually, is not an inefficient manner. There are several problems such as additional water consumption, bad quality of fertilizer preparation, Additional or insufficient fertilizer consumption. An automatic agricultural system with an automated irrigation system having a universal nozzle for spraying water, fertilizer, pesticides based on the need is implemented. The field is monitored by having a soil moisture sensor, humidity sensor, and temperature sensor. The sensing units are placed in various locations for observation. The growth of the plant is monitored using drone NDVI and NIR sensors. This module consists of a Programmable Logic Controller (DRONE) for its overall automation. NDVI Sensors are connected to the IoT controller and the output is given to the solenoid valve. A pumping motor is implemented for irrigation depending upon the requirement the values are opened by using an electrical valve named Solenoid valve (a logic function of ON and OFF as output). As soon as the required level of water is irrigated, the sensing element senses and stops the pump preventing excess irrigation. This DRONE automation is more efficient in automatic water drip Irrigation system, pesticide and fertilizer spraying with float level switch. IoT networking connected to the DRONE controller using the IoT multispectral camera of Drone Controller for damage plant detection, Sprayer controlling and saving the daily database.
S. Meivel, K. Dinakaran, N. Gandhiraj, and M. Srinivasan
IEEE
New technologies are launching on every day in Indian agriculture. But it is expensive and high power consumption. We proposed a reliable system with low power usage and introduced an agricultural engineering project that is known as UREA spraying system using Unmanned Aerial Vehicle. That is rapidly upcoming method for remote-sensing data acquisition, mostly aerial green field images and derived products. By now, the systems are less weight and cost-effective, the improvement of the sensors and their good enable a relative safe operation with good satisfaction of customer. This Project introduced a high-performance quad-rotor model that is built around a high implementation, which is all direction (0°-360°) supports with high payloads through less weight materials and advanced brush-less BLDC motors. The UAV is 1.8m in diameter, weighs 3litre in total, and generates a maximum lift of 8 kg. Without payload the maximum flying time is around 60min. The UAV programmed using Arduino through network. The UAV solve the problem using RTOS for any critical situation. The paper also focuses to land edge detection, atmospheric climate detection through camera required to obtain accurate remote sensing products that are useful for plants monitoring and Urea Spraying equally to each ground in sector-wise. UAV speed is 100m/minute. UAV spraying time is 1 hour/acre with 4 rounds. UAV carrying weight is 5 liter of urea content. It has Manual mode for manual operation and auto mode for repeated operations. Arduino Controller is used for UAV lifting control and BLDC motor controller. We have designed four BLDC motor and designed a micro-strip patched antenna of RF transmitter with 2.919GHz for controlling the payload. This paper mainly proposed to Improvement in weight lifting capacity may leads to adding a more function in account of UAVs. The payload of our Quad-rotor is around from 1 litre to 5 litre water content which adds function of weight lifting in Agricultural fields for Urea spraying operations.