2022-23

Permanent URI for this collection

http://192.168.75.5:4000/handle/123456789/15946

Browse

Now showing 1 - 20 of 76

A Deep learning model for collective disorder using Visual Geometry Group 16
(NHCE, 2022) V RAVI RAJ : UJWAL V : SACHETH N K
In the current scenario, people have become vulnerable to various diseases due to their lifestyle and their environment. Many of the analyses that have already been done looked at specific diseases. A user needs to use one analysis when they want to analyse diabetes, and another analysis when they want to analyse heart disease. This process takes a while. Moreover, if any user having multiple diseases, but the current method can only anticipate one disease, there is a potential that the death rate may rise as a result of the inability to foresee the other diseases. It is feasible to predict multiple diseases simultaneously using a multi disease model. Hence Users do not need to navigate numerous models in order to predict diseases. Time will be cut short, and there is a likelihood that fatality rates will go down because it can predict several diseases at once.
A Novel Approach for Monitoring Agricultural Production Process using Wireless Sensor Networks and Machine Learning
(NHCE, 2022) ADITYA ARABALE ; CHARANRAJ K R ; NISHANTH S B
The majority of nations rely heavily on agriculture. In India, agriculture directly supports more than half of the country's population. The yield of a given crop is influenced by a number of variables, including the climate, wind speed, soil quality, humidity, etc. The growth of a crop is impacted by these components' ongoing variability. The agricultural industry has benefited from technological advancement. In the agriculture industry, wireless sensor networks and crop yield prediction have had a significant impact. This study introduces a revolutionary Precision Farming method and demonstrates how sensor data can be effectively utilized. The Crop Yield Prediction model receives real-time input from the sensor-generated data. This strategy aids us in getting more accurate results. The project suggests a web application that, every split second, sends data from wireless sensors used in precision farming as an input to a crop production forecast model. Additionally, these sensed parameters from different users can be used as a training dataset. This method not only makes the most of the sensor data but also predicts crop yield with accuracy and promptness.
Accident Evasion and Warning System
(NHCE, 2022)
Worldwide, a sizable fraction of traffic deaths take place every day. Developing automatic methods to recognise traffic accidents and reducing the amount of time it takes for first responders to get on the site after an accident are two efficient strategies to reduce the number of traffic deaths. The automated accident detection and alerting system that is fitted into modern vehicles is used. Although these techniques work, they are expensive, difficult to maintain, and not available in many vehicles. On the other hand, it has only recently been practical to use a smartphone to identify traffic accidents thanks to advancements in the processing speed and sensors used in cell phones. The bulk of smartphone-based accident detection systems use the vehicle's high speed (as determined by the smartphone's GPS receiver) and the G-Force number as their primary data points (extract from the smartphone accelerometer sensor). 90% of on-road collisions occur when the speed is low, according to multiple sources. Therefore, in addition to high-speed accident detection, low-speed accident detection was the main emphasis of our effort. Determining whether the user is inside the car or outside, strolling or gently running, presents the largest obstacle in averting a low-speed accident. In this work, the impact of this obstruction is lessened using a recommended approach that distinguishes between the speed variation of a low-speed vehicle and a walking or slowly moving human. The proposed system comprises two stages: prediction and detection. In the detection phase, both low- and high-speed vehicle accidents are found. Specific information, including images, videos, the location of the accident, and other details, are transmitted to the emergency responder during the notification phase and shortly after an accident is reported to ensure a speedy recovery. The technology performed incredibly well throughout testing in a simulation of real-world conditions. Keywords— Vehicle Tracking, Accident, detection, SMS notification, GPS, GSM
Agranomy Apparatus Using Convolution Neural Network
(NHCE, 2022) C .MOHITH REDDY: RANGANI ROSHINI: L.N.SAI NEHA
Agrobot:- Agricultural Robot Using IOT and Machine Learning
(NHCE, 2022) MOHAMMAD HUZAIFA: RAJSHEKHAR REDDY: PAVAN G
In today’s world agriculture plays a very important role in the manufacturing of textiles, clothes, production of surplus amount of crops that is food which is essential for the everyday livelihood of mankind. Unfortunately, there is a lot of hassles which are taking place in the agricultural industries such as unpredictable natural disasters such as droughts, famines, floods etc which can incur huge loss for the agriculture industries as well as the countries which have an agrarian society not only will it be affected by natural disasters but can also be effected by the diseases which have the potential to destroy the crops. The aim of our project is that it helps the farmer and the agriculture industries to thrive so that there can be less occurrence of food shortages by efficiently increasing the production of crops tenfold, analyzes the soil fertility for better plant growth, helps prevent plant epidemic by analyzing which fertilizer is better suitable for the protection of the plant, analyzes the weather and the plants that is suitable to grow in the particular weather condition and predicts the occurrence of natural disasters such as droughts and floods for early prevention from the crops beings effected by using the ATMEGA controller.
An Experimental Analysis on Mitigating the Effects of Malicious Nodes in a Federated Learning System
(NHCE, 2022) SATHISH KOTTURI : SHIBI STEPHEN K : SHUSHANK BALAJI REDDY
This paper describes how deep learning can be used to provide security for IoT devices by analyzing the data packets that arrive at an IoT device and classifying them as packets part of the normal operation of the device or packets sent with a malicious intent. An experimental analysis is performed to check the effectiveness of such an approach with the help of the data present in the MQTT dataset. Federated learning approach is suitable for the IoT platform as IoT devices tend to contain less computing power. But a consequence of this is that the networks can contain malicious nodes which send wrong updates to the model decreasing its accuracy. We propose the introduction of verifier nodes into the system which verify the given updates sent by a node and check if it actually increases the accuracy of the model before appending it to the global model. The extent to which the malicious nodes impact the accuracy of the model and the remedy provided by the introduction of verifier nodes is also studied in this paper.
Analysis of Trending NFTs using Time-Series Data and Machine Learning
(NHCE, 2022) Kasish S V; Jashwanth M S; Umashankar Reddy M
Blockchain technology has reshaped the financial ecosphere. The first recorded use of blockchain may be found in a whitepaper from 2008 that was produced by a person going by the pseudonym Satoshi Nakamoto. NFTs, or Non-Fungible Tokens, are a blockchain product that has sparked a lot of interest from the general public. A NFT is a digital asset that utilizes blockchain technology. Because to its impossibility to be copied, replaced, or divided, it is used to demonstrate ownership and authenticity. The ownership of an NFT is recorded in the blockchain and transferable by the owner, making it possible to buy, sell, and trade NFTs. In this paper, we make an effort to establish a relationship between NFT value and several factors, such as social media, OpenSea data, and others. By using algorithms like Random Forest classifier and Support vector machine we have developed an algorithm which can accurately classify the price range of the NFT taking into consideration the current market trends. Through these adopted algorithms, we have obtained has better accuracy and ability to classify the NFT market. The model has been trained using a large dataset comprising around 70000 records and 130 factors. Thus, the model has shown an accuracy of over 95%, improving the accuracy of previous model already in place. Keywords: Blockchain, NFT, prediction, PCA, SVM, Random Forest classifier, Regression
Analysis of Women Safety In India Using Machine Learning on Different Social Media Platform
(NHCE, 2022) Konapalli Sai Chaitanya Reddy; Guggulla Geetha Priyanka
In the current scenario, women community facing issues like gender discrimination, schooling, child marriage, sexual assault and harassment, and much more, not just from society but also from social media. Women are protected by organizations like the She Team, Disha Act, and many others in society, but these organizations are much less in social media. Social networking sites like Twitter, Instagram, Facebook, etc. cause problems for women. This paper focuses on the safety analysis and monitoring of women using various social media platforms in Indian cities. The posts on Facebook and Instagram, as well as tweets on Twitter, that abuse women are considered and show the percentage of threats that women face from social media, which aids in understanding by the youth of India who misuse the women's safety and harass them in social medias via tweets, posts, and text should face strict action. People may grasp the threats to women with the help of this, and it demonstrates that women face challenges not only from society but also from social media platforms. The outcome is easily comprehended in the form of a graph and a pie chart. Algorithms such as Nave Bayes (NB) and XGBoost are used in the analysis of women's safety on various social media sites. The goal is to use classification techniques to categorize or forecast the Type based on dataset properties. Using categorization algorithms, we can determine whether social media content is positive, negative, or neutral. It has been substantiated that Naive Bayes algorithm has proved better accuracy compared to random forest and decision tree algorithms.
AR: A Visual Aid for Classrooms
(NHCE, 2022) MEGHNA ASUTI : SALONI BELLIAPPA B : ATHARVA VISHAL KAPADNIS
The conventional methods of education as we previously knew them are becoming obsolete. They are getting more and more digital and are impacted by technological improvements. There are numerous applications for augmented reality in the classroom. It facilitates the students' learning, comprehension, and retention of information. AR also increases the enjoyment and engagement of learning itself. Among the benefits include a learning process that is quicker and more efficient, enhanced teamwork skills, and practical learning that is relevant worldwide. When and where they are needed, accessible learning resources are available and do not require any special tools. The goal is to create a computer vision application that uses the idea of augmented reality to show a 3D representation of the image that the user of the software has scanned. One of the impressive implicit properties of Augmented reality is its ability to blend reality with virtual data. In most fields, augmented reality is employed as a system to help people execute activities. Particularly in the fields of surgery and aero plane manufacturing, AR has shown to be helpful in improving task efficiency and accuracy. In the event of surgery, it can be used as a tool to create 3D models of the patient's operated organ or body part, which can assist doctors in performing procedures with the least amount of risk and difficulty. AR also increases the enjoyment and engagement of learning itself. always Having access to educational resources and locations without the need for specialized equipment is only one benefit. Any level of education or training can benefit from practical learning.
Artsy: An Ml Routine Detecting Desktop Assistant
(NHCE, 2022) NISHITHA TANUKUNURI : VIJAY KUMAR REDDY P :KARTHIK SURYA J
The voice assistants that we have today such as Apple Siri, Amazon Alexa, and the Google Assistant are a complex Artificial Intelligence technology. People now connect with computers in novel ways thanks to personal assistants, conversational interfaces, and chat bots. A personal virtual assistant may even perform certain basic duties like launching apps, reading out notifications and messages, taking personalized notes for you, etc. with just a voice command. Users can ask inquiries to them in the same way they would to a real person. This paper targets on describing the importance of digital assistants and how they can be further enhanced compared to the versions we have in the market today. These Voice assistants can be used to do basic personal task management duties as well as more advanced capabilities and connected-device integrations. However, the functional and topical use of a digital voice assistant varies from person to person. As a result, the respondents' contextual knowledge is reflected in this study. Only a little amount of empirical evidence of client satisfaction with digital voice assistants exists now. To investigate this research gap, PLS-SEM was largely employed to analyze 245 survey responses. While Siri users made up much of the sample (72 percent), additional digital voice assistants (28 percent) were also identified and included. The findings also revealed that customer expectations and confirmation of those expectations have a very favorable and considerable impact on the customer satisfaction when it comes to digital assistants. As many of the multinational corporations continue to integrate digital voice assistants into many or all their operations, they must ensure that consumers are effectively assisted and that customers understand what to expect from the firm's interactive experience. Regardless of the circumstances, consumers have always prioritized the validation of these expectations. Keywords: Voice Assistants, voice command, Artificial intelligence, conversational interfaces, chat bots, voice command.
Bitcoin Price Prediction
(NHCE, 2022) TANUJ SINGH RATHORE: KRISHNA AGRAHARI: SAUMIK SHUBHAM
The new methods for tackling prediction issues in time series include machine and deep learning-based algorithms. It has been demonstrated that using these strategies yields more accurate findings than using traditional regression-based modelling. Long Short-Term Memory (LSTM), a type of artificial Recurrent Neural Network (RNN) with memory, has reportedly been shown to outperform Auto-regressive Integrated Moving Average (ARIMA) by a significant margin. Additional "gates" are incorporated into the LSTM-based models in order to memories longer input data sequences. The main query is whether the LSTM architecture's built-in gates already provide a strong prediction and whether further data training is required to further enhance the forecast. By traversing the input data twice (from left to right and from right to left), bidirectional LSTMs (Bi-LSTMs) offer extra training. If Bi-LSTM outperforms standard unidirectional LSTM after adding training capabilities, this raises the research topic of interest. This study compares and analyses the behaviour of the Bi-LSTM and LSTM models. The goal is to determine the extent to which additional layers of training data may be used to fine-tune the relevant parameters.
BMyVision: A Virtual Eye For the Visually Impaired
(NHCE, 2022) ADITI VYAS : ARPITA GADO : CHIDANANDITA MOHANTY :
BMyVision is an advanced assistive technology designed to assist people who are visually impaired. It is a wearable device that functions as a virtual eye and provides users with an enhanced visual experience through the use of haptic feedback. This innovative technology works by capturing real-time images of the user's surroundings using a camera, which then processes the images to provide the user with a tactile representation of their environment. The device consists of a pair of glasses that houses a miniature camera and a microcontroller. The camera captures the images, and the microcontroller processes the information and converts it into tactile feedback, which is then transmitted to the user through vibration or pressure. The user can then use this feedback to navigate their environment, avoiding obstacles, and identifying objects in their surroundings. BMyVision has several advantages over traditional assistive technologies for the visually impaired. The most significant advantage is that it provides users with real-time information about their surroundings, which allows them to move around independently and confidently. Moreover, unlike traditional assistive technologies, BMyVision is not limited by the user's ability to process or interpret visual information. Instead, it provides users with an intuitive and easy-to-use system that can help them navigate their surroundings more effectively. The device is still in the early stages of development, and there are ongoing efforts to improve its functionality and usability. However, the potential benefits of this technology are enormous. It has the potential to significantly improve the quality of life for visually impaired individuals by providing them with greater independence, access to the world around them, and a more fulfilling and active lifestyle.
Breast Cancer Detection Using Deep Learning Methodology
(NHCE, 2022) MILINA M: MOHAMMAD AINAIN: NEEL GANAPATHI SABHAHIT
Breast cancer is a significant health issue, with an estimated 2.3 million new cases diagnosed each year worldwide. Histopathology images are commonly used in the diagnosis of breast cancer, as they can provide detailed information about the tissue structure and characteristics. However, accurately classifying these images can be a challenging task, requiring specialized expertise and experience. Many Deep Learning and Machine Learning models have been applied to solve this problem. Convolutional neural network (CNN) is a model applied for image recognition and processing pixel data. The Support Vector Machine (SVM) algorithm is based on Supervised Learning. It is often used for Regression problems and Classification problems. Random Forest, a collaborative algorithm for learning, is also largely used for both Regression and Classification problems. This project is a comparative analysis of the CNN, SVM and Random Forest Algorithms based on various evaluation metrics like accuracy, precision, specificity, recall and F1- score. The dataset used in this project is the Kaggle Breast Histopathology Image Dataset. We also create an easy-to-use web application to aid in the selection of the best model for classification.
Copy-Move Forgery Detection
(NHCE, 2022) BASIREDDY MOUNISH KUMAR REDDY: BHARATH B S: ANKIREDDY RAMA LINGESWARA REDDY
Images have been used as evidence for a long time. With the introduction of digital images, there was a rise in manipulation of these images. There were several manipulations that could be done to an image to hide important details. Some of the most popular types of forgery include image splicing where parts of other images are superimposed on another image and Copy-Move forgery where parts of the same image are copied and moved over different regions of the same image. These types of forgeries posed a serious issue as morphed images cause misinterpretation of information available in the image. To counter these forgeries, many methodologies were developed. While some of them were computationally intensive, others suffered with less accuracy. In this project we have focused on Copy-Move forgery detection and have implemented a novel approach to localize the regions where forgeries were detected. We have modified the traditional DCT based approach and have added a filter namely LoG (Laplacian of Gaussian) to smoothen the image and generated edges. In the project, the user needs to give an image and the software developed will use DCT and LoG to detect the regions that are similar which are the regions where forgery could have taken place. The software produces an image which will mark those regions as forged. Using this software, the detection of forged images can become simpler and faster.
Crowd Counting for Risk Management using Deep Learning
(NHCE, 2022) SANJAY RAGHAVENDR: SRI TANMAYI CH: RISHTI HIREMATH
CYTOCOL – Online Law System
(NHCE, 2022) GAGAN P: MAITRI SIDDHARTH MEHTA : NIHARIKA MAHESH
The Indian Penal Code (IPC) is a comprehensive legal document that encompasses a wide range of criminal laws in India. However, navigating the IPC can be a daunting task due to its vast size and complexity. To make legal research more accessible and efficient, the Online Law System has been developed. This web-based platform offers a comprehensive collection of laws organized by keywords and a powerful search function. The system provides legal professionals, researchers, and the public with an invaluable tool for accessing relevant legal information related to criminal activity in India. In this paper, we examine the features and benefits of the Online Law System, highlighting its user-friendly design and extensive collection of legal resources. The Online Law System not only makes legal research more accessible, but also improves the accuracy and efficiency of legal proceedings by providing relevant legal information in a timely manner. This study emphasizes the importance of the Online Law System in facilitating access to legal information and its crucial role in promoting transparency and accountability within the Indian legal system. By providing an extensive collection of laws categorized by keywords and a powerful search function, the Online Law System is a valuable resource for anyone interested in criminal law in India. Overall, the Online Law System is a critical tool for navigating the complexities of the Indian legal system, and its user-friendly design makes it an essential resource for legal professionals, researchers, and the public alike.
Data Leakage Detection Using Cloud Computing
(NHCE, 2022) K. ABHILASH REDDY : K. SAI GIRISH :M. NANDI VARDHAN REDDY
The growth of cloud computing has led to an increase in the amount of data being stored and processed in the cloud. However, data leakage remains a significant concern for cloud users, as data can be accessed by unauthorized parties. To prevent data leakage, various techniques have been proposed, such as access control, encryption, and anomaly detection. In this paper, we conduct a comparative study of these techniques and their effectiveness in preventing data leakage in the cloud. We also discuss the limitations of each technique and propose potential solutions. Our results indicate that while access control and encryption are effective, anomaly detection can provide additional security by detecting unauthorized access attempts in real-time. Furthermore, we discuss the impact of the COVID-19 pandemic on data security in the cloud and provide insights on recent advancements in this field. Cloud computing has become a popular computing model due to its scalability, flexibility, and cost-effectiveness. However, the security and privacy of data stored on the cloud remain a significant concern. Data breaches and unauthorized access can result in financial losses, damage to reputation, and legal liabilities. With the increasing adoption of cloud computing for storing and processing data, the need for data security and privacy has become crucial. One of the widely used encryption algorithms for data security is the Advanced Encryption Standard (AES).
Deep Learning Implemented Communication System for the Auditory and Verbally Challenged
(NHCE, 2022) ADITYA GOSWAMI : ANANT SHAYNAM :BINDYA S :
Sign language is a natural way of communication for challenged people with speaking and hearing disabilities. There aren’t a lot of text to sign language conversion system, due to the lack of abundance of sign language libraries including all collection of written texts and learning sign language is not an easy task to begin with. This shows a need for developing a system that facilitates the communication of people with such disabilities with the rest of the world. Our project would bridge the gap between people untrained in sign language and people who rely only on sign language for their day-to-day activities, without having to assume or rely on someone who already knows sign language. This is a way to not only encourage communication with people with disabilities without worrying about miscommunication, but to even enjoy it.
Deepfake Detection Using Deep Learning
(NHCE, 2022) BHARAT PURI; SOMNATH MUKHERJEE : JAGADEESH KUMAR
In a narrow definition, deepfakes (stemming from "deep learning" and "fake") are created by techniques that can superimpose face images of a target person onto a video of a source person to make a video of the target person doing or saying things the source person does. This constitutes a category of deepfakes, namely face swap. In a broader definition, deepfakes are artificial intelligence-synthesized content that can also fall into two other categories, i.e., lip-sync and puppet-master. Lip-sync deepfakes refer to videos that are modified to make the mouth movements consistent with an audio recording. Puppet master deepfakes include videos of a target person (puppet) who is animated following the facial expressions, eye, and head movements of another person (master) sitting in front of a camera. We have created a project that tries to push the boundaries. and make the life of the user easier by telling the user if the image or the video uploaded is fake or real.
Design and Implementation of a method for diagnosing various stages of Alzheimer’s Diseases using Deep Learning
(NHCE, 2022) KAVYA.S: LIKITHA.R: GUNA KEERTHI.P
Deep learning, a cutting-edge approach to machine learning, has demonstrated superior performance over classical machine learning at recognizing detailed structures in complex, high-dimensional data, particularly in the field of computer vision. Due to the rapid advancement of neuroimaging techniques and the resulting large-scale multimodal neuroimaging data, the use of deep learning to automate the early identification and categorization of Alzheimer's disease (AD) has recently attracted a lot of attention. An organized review of articles using neuroimaging data and deep learning techniques for diagnosing AD was carried out. When fluid biomarkers and multimodal neuroimaging were coupled, the classification performance was at its greatest. Using multimodal neuroimaging data, deep learning algorithms appear to have promise for the diagnostic classification of AD since their performance keeps improving. Deep learning-based research on AD is still in its early stages, and it continues to advance in terms of performance by embracing new hybrid data types, such as omics data, and transparency by using explainable methods that incorporate knowledge of particular disease-related traits and mechanisms. The most frequent factor contributing to a deterioration in cognitive function is Alzheimer disease (AD). Language, memory, understanding, attention, judgement, and reasoning are all affected by this neurological condition, which often affects persons over the age of 65. Keywords—Alzheimer’s Disease, MRI, ADNI, Deep Learning, Diagnosis.

Browse

Browsing 2022-23 by Title

Results Per Page

Sort Options