Dr.B.Hemalatha
@drngpit.ac.in
Associate Professor and Department of Information Technology
Dr.N.G.P.Institute of Technology
RESEARCH, TEACHING, or OTHER INTERESTS
Multidisciplinary, Computer Engineering, Dental Hygiene, Computer Vision and Pattern Recognition
Scopus Publications
Scopus Publications
B Hemalatha, B Karthik, and C V Krishna Reddy
European Alliance for Innovation n.o.
INTRODUCTION: Image Processing (IP) methods play a vital role in medical images for diagnosing and predicting illness, as well as monitoring the patient's progress. The IP methods are utilized in many applications for example in the field of medicine.
 OBJECTIVES: The images that are obtained by the MRI magnetic Resonance imaging and x rays are analyzed with the help of image processing.
 METHODS: This application is very costly to the patient. Because of the several non-idealities in the image process, medical images are frequently tainted by impulsive, multiplicative, and addictive noise.
 RESULTS: By replacing some of the original image's pixels with new ones that have luminance values which are less than the allowed dynamic luminance range, noise frequently affects medical images.
 CONCLUSION: In this research work, the Speckle type noises are eliminated with the help of Mean Filter (MF) and classify the images using Multi-SVM classifier. The entire system developed using python programming.
B. Hemalatha, B. Karthik, C.V. Krishna Reddy, and D. Gokulakrishnan
IEEE
Machine Learning(ML) algorithms are used in cervical cancer categorization to determine whether a person has cervical cancer based on pertinent information from medical records. This procedure is critical in healthcare for prior treatment. For this kind of categorization problem, many ML approaches such as RF(Random Forest), SVM(Support Vector Machine), and LR(Logistic Regression) can be used. We employed ML techniques to increase the accuracy, precision, and recall of cervical cancer diagnosis in this study. SMOTE preprocessing was used to resolve data imbalance by producing synthetic samples for the minority class. Furthermore, t-SNE feature extraction was used to capture complex structures in the data. SVM consistently outperforms RF and Logistic Regression when combined with SMOTE preprocessing and t-SNE feature extraction, demonstrating improved accuracy, precision, and recall rates. This research highlights the efficacy of these strategies in enhancing cervical cancer categorization, indicating the possibilities for precise and reliable forecasting in medical investigations. The proposed SVM gave accuracy of 92.60%, precision of 0.91 and recall of 0.90 respectively. The tool used for execution is python.
Irene Michelle, Deepa Jose, Jhonatan Meza, B. Hemalatha, Nithya T, and Saptarishi Sengupta
IEEE
The prediction of the COVID-19 patients at an early stage is an open challenge in the research community. CT is the tool used to diagnose lung diseases, such as pneumonia. However, the CT scan procedure has the fastest turnaround time when compared with the molecular diagnostic test achieved in the ordinary laboratory that offers comprehensive information relevant to pathology such that it offers quantitative measurement in the dimension of lung lesion. The faster-increasing number of death and confirmed cases causes overburden in various medical systems worldwide. Recently, ReverseTranscription Polymerase Chain Reaction (RT-PCR) test is mainly used for detecting COVID-19 disease. However, the RT-PCR test consumes more time usually days for exploring the result of patients. Screen large number of cases for treatment and successful isolation is a high priority for controlling the COVID-19 spread in the entire world. Due to the inconvenience caused by CT scans, Chest X-rays are now emerging for automotive detection. Conventional neural network (CNN) is favored for classifying COVID-19 patients using chest CT images]. The three-dimensional (3D) CNN is used for classifying pulmonary blood vessel veins using CT images. Deep CNN is used to categorize interstitial lung disease with CT images. However, it achieves better accuracy in the classification of a lung nodule. For real-time detection, a web application using python is developed in order to reduce the burden on health care practitioners. In addition to accurate and effective predictions, it also prevents radiologists from getting infected.
B Hemalatha, B Karthik, C.V Krishna Reddy, and A Latha
Elsevier BV
B. Hemalatha, B. Karthik, S. Balaji, G. Vijayalakshmi, and Rabindra Nath Shaw
Springer Singapore
B. Hemalatha, B. Karthik, S. Balaji, K. K. Senthilkumar, and Ankush Ghosh
Springer Singapore
The industrial communication systems of present time are to a large extent based on commercial operating systems, open protocol implementation & communication application which are not secure. By connecting to the internet or other public networks these risks are exposed to potential attacker and cause damage to the industries. The data send by the network can be tampered causing network failure. This results in an increase in the downtime of the industry causing major loss. The industrial communication channel can be accessed by unauthorized person and uncorrect information can be transmitted, i.e. the originality of the information is violated. Cryptography is the technique, which is used to improve SCADA security. Cryptography is hiding of actual information. But SCADA protocols do not support any type of cryptography due to limited computational capabilities of SCADA device and low rate data transmission in SCADA network. To tackle this situation we are introducing an encrypting/decrypting module (ENIGMA) which can be connected at one end of the plant and one end of the control room. The encrypting module encrypts the data at the transmitting end and transmits it through the communication channel. The decrypting module decrypts the data received via the communication channel.
Jasmin. M, , Ramya. S, Hemalatha. B, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
In any type of communication in robots one too many communication is only possible. I.e. One master node can communicate with one slave node only. But here we are going to build up to two slave robots that communicate with the master node. This is possible only by using signee as it is alone having a separate unique id for it. Here we have used only two slave robots for one master node but, we can add a number of robots for the master node having of different id’s using of ZigBee transfer protocol.
Mohanraj, , Hema. R, Hemalatha. B, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Although more than 10 years have elapsed from the proposal of Spatial Radio Value system, in these years the analysis was based around evaluation in order to update the convenience of flexible braking for versatile philosophical radios specially built.In versatile specially appointed systems (MANETs), the system topology changes as often as possible and eccentrically because of the subjective portability of hubs. This element prompts successive way disappointments and course reproductions, Which creates an increase in control friction influence course disclosure can't be dismissed. Consequently, it is basic to lessen the overhead of course revelation in the outline of steering conventions of MANETs taking into account .Here we concentrates on a novel steering convention in light of probabilistic rebroadcast and neighbor scope data got utilizing Cognitive radio worldview toward decrease and directing above .Cognitive Radio is a late worldview that goes for added adaptable besides proficient utilization of radio range. It permits remote radios to deftly get to partitions of whole radio range with no unsafe impedance to authorized clients. This joined methodology empowers to give better execution of steering in portable specially appointed Networks and effective use of Radio Spectrum.
Balaji. S, , John paul Praveen A, B. Hemalatha, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Given a lot of pictures that contain objects from a typical classification, object co-division goes for naturally finding and sectioning such regular articles from each picture. In the proposed structure, we initially present the idea of association foundation and use it to improve the power for smothering the picture foundations contained by the given picture gatherings. At that point, we likewise debilitate the necessity for the solid earlier learning by utilizing the foundation earlier. For the feeble foundation earlier, the model which is called the MR-SGS model is utilized. This is characterized as complex positioning with a self-learned diagram structure.it can derive the reasonable chart structures as opposed to fixing diagram structures in a given plan.
Hemalatha, , Sowmiya Manoj, Ramya, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
In latest days, the amount of cars has risen dramatically. As a result, free parking space has become a issue, particularly in metros. This raises the demand for the development of a semi-automated car parking system, which makes effective use of room and prevents traffic congestion. But the operation and maintenance of such a car parking system is very difficult. Earlier, the parking spaces available for parking were tracked by human staff, which was very time consuming. Here, we have created a semi-automatic parking scheme that provides a distinct parking slot for each vehicle. Therefore, the driver does not need to search for a vacant slot. This scheme also requires less manpower. Any number of cars entering or leaving the zone at the same time may also be accommodated. This stops the driver from stopping at the door. The system is extremely energy efficient by turning on the lamps in the parking area only when the vehicle is on the move, which makes it distinct from most parking schemes. A research on current systems shows that there are automated systems based on sensor networks or video systems that monitor the existence of vehicles and the accessibility of free parking spaces. They lack the benefit of giving each vehicle a distinctive slot, thus generating confusion when many vehicles arrive at the same time.
S. Philomina, B. Hemalatha and R. Hema
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Now days we are sending many satellites into the space which are mainly useful for communication. Till now we have sent many satellites and we have dropped many things in the space like those dead particles revolves around the earth in Low Earth Particles (LEO). This derbis became a problem. It will be a risk to other particles which are revolving in LEO. If we doesn’t take any steps it will lead to collisions and destroy all communication system. The Japan Aerospace Exploration Agency (JAXA) Aerospace Research is studying a micro-satellite space derbi removal scheme. This paper discuss about the Key technologies, strategies and specifications of removing derbis satellite
Karthik. B, , Hemalatha, Jasmin. M, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
In many image processing applications, a wide range of image enhancement techniques are being proposed. Many of these techniques demanda lot of critical and advance steps, but the resultingimage perception is not satisfactory. This paper proposes a novel sharpening method which is being experimented with additional steps. In the first step, the color image is transformed into grayscale image, then edge detection process is applied using Laplacian technique. Then deduct this image from the original image. The resulting image is as expected; After performing the enhancement process,the high quality of the image can be indicated using the Tenengrad criterion. The resulting image manifested the difference in certain areas, the dimension and the depth as well. Histogram equalization technique can also be applied to change the images color.
Mohanraj, , Hemalatha. B, Sowmiya Manoj, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
In today’s world, with increasing price of fuels, depletion of natural resources and pollution levels mounting up, an eco friendly and cheap alternative fuel in automobiles can be implemented. A hydrogen fuel cell makes it possible by converting the hydrogen to electricity that powers the vehicle. The proposed green fuel is hydrogen which is combined with oxygen inside the fuel cell. The byproduct is water which can be used for purposes like irrigation, domestic and industrial uses after treatment. The fuel cell efficiency is dependent on its moisture content and so a MEMS based humidity sensor is proposed for measuring the humidity
Karthik. B, , Hemalatha. S, Jasmin. M, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
This paper provides a solution for the hand impaired people who cannot use the mouse to operate the computer. This is a device which controls the mouse using voice commands using HM 2007 Voice DSP Processor. A specific function is assigned to specific instructions. According to the instructions given, the mouse is controlled wirelessly using RF.Therefore, to make the assistive device for the disabled people is the main aim of this paper. Here, we design amouth-activated mouse applying face detection technique. By using a webcam, the pictures of mouth position can be captured in a computer. We can then control the cursor and click the mouse by moving mouth. Through this equipment, the living quality of disabled people can be improved.
Hemalatha B, , Subbulakshmi. K, Kanagavalli. G, , and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
There is a clear need to enhance security action to prevent any malicious use or inadvertent abuse of radiation sources. Some of these radioactive sources are consistently transferred beyond the office or laboratory grounds for work and consulting reasons. This paper presents the early development of the radioactive citation tracking, which combined the Arduino microcontroller, Global Positioning System (GPS) and Global Mobile Communication System (GSM) systems. The monitoring scheme will assist the proprietor track the motion of irradiated sources. The system is completely capable of monitoring the motion of the toxic substance through GPS satellite signals. The GPS position could either be transferred to the headquarters at a set intervalShort Messaging Service (SMS) enables real-time surveillance or storage on a flash drive for offline surveillance and information capturing.
B. Hemalatha and S. Srinivasan
Diva Enterprises Private Limited
A wireless sensor network is a correspondence framework which detects and accumulates data from a specific zone and sends it to the base station or sink node. WSN can be very exceptionally valuable for target detection and recognizable proof by military and ecological applications like agrarian farm monitoring and woodland fire supervising. A genetic algorithm is a meta-heuristic approach propelled by the process of natural selection that belongs to the class of evolutionary algorithms. It is utilized to create high-quality solutions to optimization and search problem by relying on bio-inspired operators such as mutation, crossover, and selection. The goal is deploying a minimum number of sensor element nodes without overlapping each other based on antenna height and separation between sensor nodes. The experimental result shows that better coverage in heterogeneous wireless sensor network using NS2 software simulation.
E. S. Sujit, N. Kusuma, and B. Hemalatha
IEEE
Pressure measurement is a key part of many commercial and industrial systems. Piezoresistive pressure sensors are simpler to integrate with electronics, they are inherently shielded from RF noise and their response is more linear while compared to capacitive pressure sensors. And piezoresistive devices have always dominated the pressure sensor market. The analyticals that are typically used to model the diaphragm of the pressure sensor have been analysed by many researchers. To optimize the pressure sensor for parameters like linearity and sensitivity, the Finite Element Method (FEM) is incorporated. The selection of appropriate parameters of piezoresistors such as the shape and the position of the piezoresistor on the pressure sensor diaphragm, thickness of diaphragm are important. This study shows the scope of using analytical solutions and design techniques for a piezoresistive pressure sensor.
Aditya Sripada, Abhishek Warrier, Arpit Kapoor, Harshit Gaur, and B. Hemalatha
IEEE
This work presents a control algorithm to achieve dynamic lateral balance in humanoid robots on unstable and geodynamic surfaces such as a seesaw or a suspension bridge. The proposed method generates a stable pose at any given point using the feedback from an Inertial Measurement Unit (IMU) in realtime which enables the robot to balance on dynamic surfaces. The algorithm provides a robust control of the Center of Gravity (CoG) of the robot and constricts the CoG from moving away from the most stable pose of the robot in the double support mode thus achieving stability at any given point in time. The method implemented is relatively less heavy in the terms of computation and easy to implement. The proposed system works on any humanoid platform irrespective of size and weight, given the robot has the required degrees of freedom needed to achieve balance.
Bakthavatchalam Hemalatha, , Asokan Juliet, and
The Intelligent Networks and Systems Society
The water level in boiler drum is the critical control parameter to be addressed for providing the efficient and safe operation of steam generators in power plants. The regulation is achieved by using the Artificial Intelligence (AI) technique, namely, Adaptive Neuro-Fuzzy Inference System (ANFIS). It has the capability of self-learning as ANN with the linguistic expression function of fuzzy inference, whose membership functions and fuzzy rules are acquired from a large lot of existing data instead of experience. So, it is well renowned for its control performance in the power system area due to the nonlinear nature and robustness to system parameters uncertainties in multidisciplinary domains and power system too. Here, ANFIS controller takes the error and change of error in water level as inputs to control the actuator of inflow water such that the steam flow out of the steam generator meets the requirement of turbine unit set point which indirectly resembles the demanded load by the Generation System Operator (GSO). The performance of the proposed ANFIS controller is analyzed in terms of tracking the water level set point of boiler drum. For this purpose, a boiler model is developed using the MATLAB/SIMULINK working platform. Then, the proposed ANFIS controller is designed and tested on it for water level regulation. The simulation results are examined and compared with existing FIS technique.