DIVYA K S
@kristujayanti.edu.in
A.P ,CSE,
Kristujayanti college
EDUCATION
M.Tech
RESEARCH INTERESTS
Cryptography and Network Security
Scopus Publications
Scopus Publications
V. Pavani, K. Divya, V. Venkata Likhitha, G. Sai Mounika, and K. Sri Harshitha
IEEE
The vehicle's license plate number can be read by an image recognition system, called number plate recognition. The goal is to implement a license plate-based modified authorized vehicle recognized verification system that works smoothly. The system can be implemented at the entry point for controlling access to a highly restricted area, such as a military base or the area around the most important government buildings (such as the Parliament or the Supreme Court). The built system first identifies the car, and then takes a picture of it. The license plate area of the vehicle is then converted to grayscale. At that time, the license plate is taken off the car. After that, a KNN (K- Nearest Neighbors) algorithm is used to identify the numerical and alphabetic sequences. With this information, the vehicle's owner, preferred location, current residence, etc. are tracked. Python is used for the implementation, and the system is tested with real images. The test results show that the proposed framework successfully differentiates between fake and real images of license plates. The purpose of this extension is to develop an illustration capable of correctly selecting a license plate based on a photograph of the plate.
Mudit, K Divya, Sanjeev Kumar Joshi, and Sahil Verma
IEEE
A face expression is a clear representation of an individual's affective state, cognitive activity, intention, personality, psychopathology, and it serves as a means of communication in interpersonal relationships. Automatic facial expression recognition is a critical segment of usual interface between human and a machine, and it must be used for behavioristic psychology therapeutic practice as well. Face identification and placement in a chaotic scene, facial feature extraction, and facial feature categorization are all tasks that an autonomous face features Recognition system must complete. Convolution Neural Networks are used to create a facial expression detection system. LeNet Architecture is the foundation of the CNN model. During this experiment, we have used the Kaggle facial features dataset in which we acquire seven types of face expressions which indicates our emotions swiftly. Kaggle facial feature data set with 7 face expressions labeled as (happy, sad, surprise, fear, rage, disgust, neutral) was taken for the research. Through this paper we are able to obtain 56.7% accuracy and 0.576 precision while testing the dataset.
Pawanjot Kaur, Divya K, and Lakshmipriya Vinjamuri
IEEE
Google form is a free online tool that lets users construct surveys and quizzes. It's part of Google's web-based apps package, which also include Google Docs, Google sheets and Google Slides. It's a flexible tool that can be used for a variety of tasks, including gathering RSVPs for an event, producing a pop quiz, and designing stunning forms. Google forms makes it easy to create a form that looks professional. These are the most widely used form filling tool used by people these days. It has now become a daily thing to fill google forms every other day. Many people find annoying filling the same information repetitively. In this paper, an application to automate the google form submission has been proposed. In this application a user would have to enter all the information like name, age, gender, qualifications and other personal information only once and that information would be saved in a database and can be further automatically filled in any of the google forms which the user has to fill. This application will reduce the tedious task of the user consumed in filling the forms with the same information every time.
Shubham Bhardwaj, Divya K, and Abhishek Kumar Pathak
IEEE
Hand gesture is the initial method for communication that was used by cave men. The verbal communication was developed by the more developed civilization. Hand gesture has been the oldest form of non-verbal communication. In current time, hand gesture is not only a way of communication between people who may or may not be physically challenged but also between humans and machines in diversified fields such as aviation, surveys, military simulation, medical training etc. It is the best way of interaction between man and machine as no peripheral devices are required and a wide range of gesture can be used for communication with high accuracy. For such human and machine interaction a real-time gesture recognition system is proposed. Detection of motion and skin is used for capturing region-of-interest. The region of interest is further processed by convolutional neural network in order to get a proper result.
Divya K. S, Roopashree H. R, and Yogeesh A C
The Science and Information Organization
—Security has always been a prominent concern over the network, and various essential requirements are required to cater to an efficient security system. Non-repudiation is a requirement about the non-deniability of services acting as a bridge between seamless relaying of service/data and efficient security implementation. There have been various studies carried out towards strengthening the non-repudiation system. There are certain pitfalls that render inapplicability on dynamic cases of vulnerability. The conventional two-party non-repudiation schemes have been widely explored in the existing literature. But this paper also advocates the adoption of multi-party computation, which has better feasibility toward strengthening a distributed security system. The current work presents a survey on the existing approaches of non-repudiation to investigate its effectiveness in the multi-party system. The prime aim of the proposed work is to analyze the current research progress and draw a research gap as the prominent contribution of the proposed study. The manuscript begins by highlighting the issues concerning multi-party strategies and cryptographic approaches, and the security requirements and standardization are briefly discussed. It then describes the essentials of non-repudiation and examines state-of-the-art mechanisms. Finally, the study summarizes and discusses research gaps identified through the review analysis.
B. Manimekala, K.S Divya, and Sreeparna Chakrabarti
IEEE
Vehicular Adhoc Networks (VANET) is an emerging revolution which ensures to improve the safety measures of the vehicles. To ensure safety and to keep track of accident prone zones, there is a need for communication between vehicles. The positive side of this networks that helps to improve the traffic safety and vehicle security while protecting the drivers from assaults executed by enemies. Security is a the most basic issues that identified with VANETs since the data transmitted is spread in an open get to environs. VANETs confront many difficulties. This paper shows a study of the security issues and the difficulties they create. The different classes of uses in VANETs are presented, and additionally some security prerequisites, dangers and certain structures are proposed to take care of the security issue. Here, various authentication and secure key management mechanisms in VANET are reviewed and their advantages are listed. The performance metrics used in different technique for analysis are also studied.
Divya K and Sukhvir Kaur
IEEE
Analysis of the tree rings by hand is a difficult task and agitated for dendrochronology domain area. Detection of the tree rings are quite popular in numerous fields of science. As, the detected results enables the users to determine the age of tree, tree with good ring and environment changes. The evaluation of the tree rings requires previous detection of the tree ring boundaries that is usually performed physically with devices like stereoscope, moving table, along with data recorder. To ease the manual work of users, this paper presents detection of actual ring as good for denoised images using denoising neural network (dncnn). In existing paper, author worked for 3 images with median filter. In comparison to the existing work, this paper provides 100 images which were denoised first and then detected rings.
K Divya and Sukhvir Kaur
IOP Publishing
Abstract This paper outlines an introduction to dendrochronology. The tree rings are being identified through image processing and statistical simulations. The study of the tree rings is known as dendrochronology but the techniques do need to be modified for better performance. Image analyses convert the tree ring into digital data using imaging tools. This method involves the scaling, piling, width calculation. Technology is required to evaluate the factors of different tree ring pattern. In this review paper, literature survey has been provided which deals with the contribution of different researchers, which methodologies they preferred and what were their limitations. And two attributes are being calculated named MSE and PSNR, as these obtainable values indicates to work further with the enhanced image. Also, researcher can get idea what all the attributes are important to implement this technique.
K. Divya, Akash Sirohi, Sagar Pande, and Rahul Malik
Springer International Publishing
J. Amudha, K.V. Divya, and R. Aarthi
IOS Press
Kota Divya, Prakash M., and P. Pabitha
IEEE
Multiple Online Social Networks and its applications have emerged rapidly at a large scale made dependence of global population on it for various reasons. The individuals data in each social network is just partial. Mapping these individuals across various online social networks is having more importance in forming communities and identifying the most influential node. The difficulties of fragmentary data which is the biggest challenge in the present social network era can be solved by forming groups and comparing their information which can be useful in several applications like influential node detection, spammer identification etc. The occurrence of different characterization of users leads in identifying the influential community by classifying them from other communities. Decision tree technique is used in splitting into different communities from multiple social network. Various number of trees are generated by changing class labels. Greater the performance metrical values (accuracy, precision, recall, f1-score) of the tree is considered for the effective formation of communities. Experimental results of the proposed method achieves 0.54, 0.58, 0.54, 0.54 in accuracy, precision, recall, and f1-measure, respectively.
P. A. Reshma, K. V. Divya, and T. B. Subair
Springer International Publishing
P A Reshma, K V Divya, Geethu J Therattil, and T B Subair
IEEE
Recent biometrie research has examined the possibility of obtaining attributes such as hair color, age, gender, weight, ethnicity, height, etc. from biometric traits, face, hand geometry, fingerprints and iris. This paper examines detailed study about different process, for each step of predicting gender from iris for achieving better accuracy and authentication. Capturing the iris image in high quality specification camera is used to achieving specific features of iris. Different algorithms and software systems are used to locate the boundaries of an iris image. Mutual Information(MI) is better used to compare other feature in geometry, texture etc. UND_V and GFI datasets are used to get more accuracy in SVM classifier for better gender prediction.
P. Srikanth Reddy, P. Satyanarayana, G. Sai Krishna, and K. Divya
Springer Singapore
Anusha Bilakanti, Anjana N.B., Divya A., K. Divya, Nilotpal Chakraborty, and G. K. Patra
IEEE
Computing by a cloud service has not succeeded much due to concerns about confidentiality and privacy assurances by the service providers. This challenge can be addressed if computations can be performed on encrypted data directly. In this paper, we have developed a scheme based on fully homomorphic encryption that can perform operations directly on encrypted images. Our aim is to take the first step towards solving the problem of those which are in dire need of computing services by a third party, namely cloud, without any security breaches. We believe this will be a major breakthrough in which security has been handled while computing.
Sreena Narayanan and K. V. Divya
Springer India
K. Divya and S. Jeyalatha
IEEE
Cloud computing is an emerging technology which is gaining wide interest both in IT industry and academic research. Instead of buying, setting up and managing hardware resources, users begin to rent virtual machines and storage space. This paper focuses on basic concepts like layered structure of cloud models, deployment models and cloud storage. And also describes how cloud computing services are utilized in Academic environment. Various security risks involved in cloud storage is discussed.
R Aarthi, K Divya, N Komala, and S Kavitha
IEEE
Medicine is one of the major fields where the application of artificial intelligence primarily deals with construction of programs that perform diagnosis and make therapy recommendations. In digital mammography, data mining techniques are used to detect and characterize abnormalities in images and clinical reports. In the existing approaches, the mammogram image classification is done in either clinical data or statistical features of an image using neural networks and Support Vector Machine (SVM) classifier. This paper is proposed to evaluate the Application of Feature Extraction by means of combining the clinical and image features for clustering and classification in mammogram images. Initially, mammogram dataset is divided into training and test set. For the training and test sets, preprocessing techniques like noise removal and background removal are done to the images and Region of Interest (ROI) is identified. The statistical features are extracted from the ROI and the clinical data are obtained from the dataset. The feature set is clustered using k-means algorithm followed by SVM classification to classify the image as benign or malignant. The accuracy obtained from the proposed approach of clustering followed by classification is 86.11% which is higher than the direct classification approach where the accuracy is 80.0%. From the above results, the superiority of the proposed approach in terms of accuracy is justified.