The purpose of this email client is to manage a user’s email and offer additional personalized functionality. The user will be able to upload his/her contacts, create different groups of contacts, and send personalized mails . It deals with adding of user’s email in the groups using excel files and forms. To produce a hybrid platform which can provide multiple features like Contacts, Groups, Form . It is pretty much useful for beginners because of its simple and easy design as well as for experienced employees in the industry. It is specifically about email design and delivery. Key Words: Email Design
A Deep Learning approach for hate speech and offensive language detection
As the online presence of an average user continues to grow, so does the spread of hate speech. We identify and examine challenges faced by social media giants and govt authorities in detection and removal of foul language and hateful posts online. Some of the challenges of hateful content detection are subtleties in language, no standard definition of hate speech .Among these difficulties are subtleties in language, differing definitions on what constitutes hate speech, and limitations of data availability for training and testing of these systems. Furthermore, many recent approaches suffer from an interpretability problem—that is, it can be difficult to understand why the systems make the decisions that they do. We propose a multi-view SVM approach that achieves near state-of-the-art performance, while being simpler and producing more easily interpretable decisions than neural methods. We also discuss both technical and practical challenges that remain for this task.
Yush Mittal, Ishika Jain
Mr. Pranshu Saxena
An Artificial Intelligence capable of corona virus detectction usingLungs X-Rays
According to the World Health Organization, the coronavirus outbreak poses a daily threat to the global health system. Almost all countries' health resources are insufficient or unequally distributed. There are several issues, such as a lack of health care workers, beds, and intensive care units, to name a few. The key to the country's health systems overcoming this epidemic is to use limited resources at optimal levels. Disease detection is critical to averting an epidemic. The greater the success, the more tightly the covid viral spread may be managed. PCR (Polymerase chain reaction) testing is commonly used to determine whether or not a person has a virus. Deep learning approaches can be used to classify chest X-RAY images in addition to the PCR method. By analyzing multi-layered pictures in one go and establishing manually entered parameters in machine learn-ing, deep learning approaches have become prominent in academic research. This popularity has a favorable impact on the available health datasets. The goal of this study was to detect disease in persons who had x-rays done for suspected COVID-19 (Coronavirus Disease-2019). A bi-nary categorization has been used in most COVID-19 investigations. Chest x-rays of COVID-19 patients, viral pneumonia patients, and healthy patients were obtained from IEEE [ (Institute of Electrical and Electronics Engineers) and Kaggle . Before the classification procedure, the data set was subjected to a data augmentation approach. These three groups have been classified through multiclass classification deep learning models. We are also debating a taxonomy of recent contributions on the eXplainability of Artificial Intelligence (XAI).
Ashish Joshi, Nitish Sharma, Mehtab Jafar, Mohammad Anas
Ms. Preeti Gupta
SSH Based Honeypot using IDS and DB
The considerable increase in usage of the internet has led to an increase in the number of cyber crimes (e.g., identity theft, data theft). There is a need for something that monitors network traffic for suspicious activities. For this purpose various Intrusion Detection Systems (IDS) were designed. We present an architecture that uses basic principles of computer networking, virtualization, and Cyber Security and integrates these with intrusion detection systems, to protect networks characterized by a constantly changing underlying infrastructure and physical topology. The work done in the paper defines a process and a novel architecture that minimizes the security risk in networks. The proposed architecture involves analyzing the behavioral signature of the user logging into the system and capturing new attacks.
Secure and Transparent Mid Day Meal Scheme based on Block Chain and IoT
In the Internet of Things (IoT) vision, conventional devices have become smart and autonomous. This vision is turning into a reality thanks to advances in technology, but there are still challenges to address, particularly in the security domain e.g., data reliability. Taking into account the predicted evolution of the IoT in the coming years, it is necessary to provide confidence in this huge incoming information source. Blockchain has emerged as a key technology that will transform the way in which we share information. Building trust in distributed environments without the need for authorities is a technological advantage that has the potential to change many industries, the IoT is among them. Disruptive technologies such as big data and cloud computing have been leveraged by IoT to overcome its limitations since its conception, and we think blockchain will be the next one to do so. Main aim of this project is to demonstrate how disruptive technologies such as the Internet of Things (IoT) and Blockchain can help address significant challenges faced by mid-day meal production and delivery. The initiative is not only enhanced efficiency while giving a boost to the current operations,but also it ensures that quality standards are met and transparency is maintained while increasing the number of meals served.
Aman Sagar, Aman Gupta,Deepak Yadav, Manvendra Singh
Ms. Shelly Gupta
Sales Forecasting using Walmart Data Set
To analyse the data and based on the trends in the sales predict the output(sales) based on some parameters. Learning various techniques for time series forecasting like Linear Regression, Logistic Regression, ARIMA etc. one should speak what he has made in abstract sense (even a person not aware of technology must be able to understand and appreciate work), whatever the system user have built is going to provide as a service, application, tool etc.
Sound is an inevitable signifier that provides us relevant information about the environment around us. It helps us perceive the environment and our ears can distinguish real world audio signals that are overlapped with each other easily. However, it is difficult to replicate human ear like functionality to recognise each sound signal distinctly with the standard detection methods. With the help of a state-of-art CNN architecture we will classify the sound using spectrogram-based inputs such as mel spectrogram.
Classifying a bird requires lots of knowledge, as Birds exhibit different features like color, beaks, wings, eye size, tail, sound, and live-in different inhabitants. Remembering species-specific information is not easy as there are around 9000 to 10,000 species. In this paper, we proposed a system backed by the CNN model which can identify a bird from an image and can provide information about that species. Our model is trained, validated, and tested on more than 28000 images of 200 species of birds. It model provided 97.44% accuracy on the training dataset and 98.40% on the testing dataset.
The visually impaired people in the world face a lot of problems in day-to-day life. They need either a human or a stick to guide them through their different daily life tasks. They often get hit by objects because they're not able to see them coming towards them. They're not able to recognize people and objects without touching or hearing them. Entertainment is also a luxury for them since they can hardly operate any device. This research paper depicts a project that can solve all the above problems and perhaps more in long run.
Niharika Arora, Nishant Singh
Ms. Neha Verma
Evading Web Applications Vulnerability Scanners and Brute Force Directory Structure Disclosure for Web Servers
The project aims to propose a model that protects the Web servers from the Web Application Vulnerability Scanners (like Nessus or Qualys) and from the tools that are designed to brute force directories and files names on web or application servers. (tools include dirbuster, gobuster, and nikto).It prevents the general first step of an attack on a server, i.e., reconnaissance or Information Gathering, which greatly reduces the risk of compromising the server. This is achieved by understanding the current working of scanners and tools, understanding their dependency on HTTP response codes, and manipulating the HTTP response sent to delay if not stop the scanning attempts.
Live Tracking And Fault Tolerance (LT & FT) For Public Transport
The route framework is fundamental for each individual, let it be making a trip starting with one spot then onto the next, or to discover the separation between different areas or to get data about a spot. In the present life, Time and Money are significant, nobody needs to burn through their time and cash, be it going on a long course or short one. Other than this, strange and sudden conditions on the streets influence the smooth activity of the transport framework and the development of vehicles. Likewise, ordinary issues, for example, traffic blockage, sudden postponements, arbitrariness in traveler request, sporadic vehicle dispatching times occur, and because of which the timetable of the travelers is influenced and they definitely need to hang tight for the appearance of their separate transport.
This traveler burden can be kept away from by presenting a framework that gives ongoing data about the area and the assessed time of appearance of the transports and educates the police headquarters about the obstacles
â€œThe Silent Voice
(Communication Interface for Deaf & Dumb)â€
Hand gesture is a channel of communication between dumb and deaf people. Hand gestures are physical movement by using eyes and hands and non-physical movement is facial appearance, head movement, body position etc. Contemporarily, there are innumerous techniques for capturing and recognizing the hand gestures. Substantial literature exists and extensive analysis, groundwork, exploration and inquisition has been done till now regarding gesture identification systems to resolve communication problem with the disabled (visual and hearing impaired) people and how it can be used with respect to real time. This project presents the method of developing such a system using latest learning techniques such as CNN and its scope of implementing it for because not every single typical person can comprehend the gesture based communication. It also discusses about certain areas and the future research which can be done in regard to CNN for efficient application of the technique in studying the complexities prevailing in the usage of the system.
Dr Neeta Verma
Comparative study for credit card defaulter detection system
A credit card is a payment card issued to users (cardholders) to enable the cardholder to pay a merchant for goods and services based on the cardholderâ€™s promise to the user to pay them for the amounts plus the other agreed charges. Implementation of a credit card default detection system is proposed in this project and a comparative study is proposed to find a suitable solution. The following research reveals the significance of modified classification in estimating new trends. Rigorous evaluation of different classification algorithms viz. Logistic Regression, Decision Tree, K-Nearest Neighbor and Naive Bayesian is explored in this project. These findings forecast the finest techniques for discovery of potential defaulters which can be adapted by banking institutions. Our main motive is to compare the performance measures between original dataset and original dataset on which principal component is applied. Different algorithms can be compared on the basis of various criterions such as Accuracy, Precision, F1-Score, Recall, ROC. Successful contrast between these attributes would yields a efficient model for the given dataset.
Mr Shailendra Singh
Project Image and Video Outpainting using GANs studies the main fundamental issue of extrapolation of images or visual context like videos using deep generative models such as GANs (Generative Adversarial Networks), i.e., extending image and video borders with plausible structure and details. In addition, this seemingly simple job faces several critical technical challenges and has its unique properties. The challenging task of image and video outpainting (extrapolation) in comparison to itâ€™s relative, inpainting (completion),received relatively little attention. So, we followed a deep learning adversarially approach which is based on training a network. The two main problems are the extension of scale and one side constraints.Extensive studies are carried out on various possible alternatives and methods connected with them.We are also exploring our methodâ€™s potential for various interesting applications that may support work in a variety of fields.
Aditya Priyadarshi Sudhansu
Ms. Preeti Gupta
Virtual Intrusion detection system
Nowadays, the majorly signature based Intrusion Detection System(IDS) are used to detect intrusions in their intranet. We have used the concept of virtualization to develop an Intrusion Detection System which will use Honeypots/Honey Clients to detect and trap the attackers. With the increasing popularity of cloud networks, serious issues around security for ubiquitous hosts will need to be addressed. In this project we focus on a virtual intrusion detection system (V-IDS). We present an architecture that uses basic principles of computer networking, virtualization, and Cyber security and apply them to intrusion detection systems, in order to protect networks characterized by a constantly changing underlying infrastructure and physical topology. Our goal is to define a process and a novel architecture that minimizes the security risk in networks, implementing the principle that network security and reliability is not a â€œproduct,â€ but a well-defined â€œprocess.â€ On the basis of the defined architecture we have implemented a prototype Virtual IDS.
The Internet has proved to be a life-changing invention of our time. Not only it is itself an invention, but it has been the key player to major things that have been invented in the twentieth century. The internet has proven itself in re-inventing almost everything, with education being no exception to it. CodeBox is also an effort, via the internet, to educate the young minds on the how-tos of coding and developing the spirit of competitive coding. It takes a lot of time and practice to get good at coding and CodeBox aims to assist the user in this journey. CodeBox will focus on giving users the required data to learn the basic skills of programming, and then eventually on enhancing the skills required for competitive programming and give the users a sense of the real world competition. It will enable the users to learn, practice and compete with others on the platform. It provides different parameters for self-analysis and evaluation like, time taken to complete the challenge (in competitive programming it is very important to know how fast your code executes so that you can get it to run under the time limit) & complexity of algorithm used.
Dr Rekha kashyap
Secure and Transparent Mid day Meals Using Block Chain
The Mid-Day meal program is an initiative by the government and NGOs to tackle malnutrition among younger generations, enrolled in educational institutions by providing them with nutritious meal. It is a large scale food assistance program with an overwhelming budget. Despite government authorities trying their best to make it a success, still, many problems ranging from poor food quality, meals not being served as per guidelines and food not being delivered are encountered. The present implementation can be improved if there is transparency in the system with the proper distribution of powers from a handful to all stakeholders involved in the process. Blockchain being an immutable, shared and distributed ledger without the control of any single centralized authority can address these shortcomings. The proposed work NutriChain has implemented the idea on a Blockchain and IoT based distribution network by storing meal preparation and distribution data on Ethereum based Permissioned Blockchain. Each batch of meal will be identified by an ID and every successful and unsuccessful transfer of a batch of meal from one stakeholder to another, will be recorded as a transaction on Blockchain. The nutrition value along with the temperature, photographs and weight of the meal are recorded by the IoT sensors as IoT metrics score. All mealDetails are to be stored on the database and its hash along with aggregated IoT metrics score will be stored as append-only blocks of transaction on Blockchain. The initial research and study shows that the Blockchain and IoT based approach in solving the chosen use case is both feasible and promising.
Ms Kumud Alok
Hand sign Recognition using Machine learning
Sign Language is a type of communication that relies on hand signs, gestures and expressions, commonly used in deaf community. But people outside of the deaf community find it hard or almost impossible to communicate with deaf people. They have to rely on an Interpreter which can both be expensive and hurt the privacy of the people trying to communicate. In this paper we are trying to propose a method which uses the power of Convolutional Neural Net to identify and recognize hand signs which are captured in real time through a laptopâ€™s webcam. Since there is no universal Sign Language, we chose our model to specifically be trained on ASL (American Sign Language) alphabets. But this method could also work for other sign languages.
The purpose of Disease Prediction System is to automate the existing manual practice of visiting doctors for initial diagnosis by the help offull-fledged website. Online symptom examination and assessmentservices are used by patients seeking guidance about their health problem.These services generally provide people with a number of possiblediagnoses with their respective accuracies based on their reportedsymptoms and suggest a course of action. According to the survey,people often feel reluctant to go to hospital or physician on minor symptoms, because sometimes even for the small problems doctorsuggest numerous tests and are asked to receive unnecessary antibiotics. Reducing the number of visits saves patientsâ€™ time, energy and money.Furthermore, it makes it easy for the patients to get services and also provides new patients to the doctors. Due to wide usage of mobile phones, users can easily access this online disease prediction system. So, it can be a great head start for users.
Ms Shelly Gupta
voice based email system for the visually impaired
E-mail stands for Electronic mail, is the fastest source of communication. E-mail is considered to be the most reliable way of communication over the internet for sending or receiving important information. Through this project we aim to create an Email system that operates on voice-based commands only. This system will not require any typing work at the part of the user; it works only on mouse operation and speech conversion to text and text conversion to speech. The system is completely based on interactive voice response which will make it user friendly and efficient to use. We also aim to study the various methods and algorithms used for STT and TTS and figure out the most efficient technique that can be deployed for both the conversion processes.
Opening the car lock with the help of mobile based application is one of the great and important features that are missing from most of the cars sold globally. Although some cars support the remote starting functionality, dealers usually limit such feature to cars sold in certain regions such as the INDIA, US, etc. or to high-end models only. Additionally, manufacturers use short-range remote controllers, which limit the benefits of using the system. This idea will help bring this important feature to almost all cars, either old or new, and will add many important features such as GPS tracking in the system, all without compromising the safety and security aspects. The information is passed onto the central processing system which is in the form of the digital signals, the microcontroller unit reads the signal and sends it to the Global Positioning System (GPS) module and using the triangulation method, GPS module feeds the exact location in the form of latitude and longitude to the userâ€™s mobile. By reading the signals received by the mobile, one can trace the exact location of car. This project involves hardware and software parts construction and the integration of both parts to create the system. We succeed in achieving the objective and in fact, add another feature to the system which will initiates a call to the owner after sending the SMS. In the end of this project, we will document all the hardware and software development and provide a simulation model of the system. An interfacing mobile app is also connected to the microcontroller, which is in turn, connected to the system. Once, the vehicle is being stolen, the information is being used by the vehicle owner for further processing.
Dr. Anjali Singhal
Automated Registration System
The Automated Registration System is developed to provide a web portal for students, faculties, registrar, system administrators, accounts section of a college. The purpose of this system is to save paper and encouraged to access data through the web portal. Through this,students will be able to complete their registration process of each semester through this portal rather than filling the registration forms. Accounts section will manage the online generation of the fee receipt on the bas payment details received from each and every student of the college.
Ms. Kirti Jain
World Tours and Travel and sentiment analysis
Tours and Travel Sentiment Analysis System is developed to provide accurate and reliable information on country, travel destination and accommodation. The purpose of developing this travel information system is to search travel destination based on traveler preference and view travel destination, location information. Another purpose for developing this web portal is to engage users by using opinion poll and review. System will also apply sentiment analysis on the feedback given to the hotel and give overall rating of the hotel.
Ms. Alka Singhal
Nowadays with growing amount of event taking place, users find it difficult to decide which is most suitable event for them. In this regard, a recommender system can help them to get an overview about a particular event in advance. In this paper we explore the user of collaborative filtering to recommend event to the user on the basis of personalised past user data and current event parameters such as location of event, total number of registrations and likes. We have tried to combine results of collaborative filtering with
content filtering to generate the final result. The two main attributes we have used for event recommendation are registration and likes. The weightage given to the registration is 70% and to the likes is 30% as we have considered users who have registered for the events are more likely to visit the event.
Dr. Anchal Jain
Neural Network based Autonomous Car
The project aims to build a neural network based monocular vision autonomous car prototype
using Raspberry Pi, ultrasonic sensor and camera. Raspberry Pi as a processing chip. An HD Pi camera along with an ultrasonic sensor is used to provide necessary data from the real world to the car. The car handle three tasks: self-driving on the track, stop sign detection, and front collision avoidance and therefore, capable of reaching the given destination safely and intelligently. Using the concept of machine learning, a car can be automated (self-driving). Neural network is used to train the car with specific images and whenever it detects the trained images,it operates according to the trained instruction. The microcontroller used in the car is raspberry pi which is used to control the ultrasonic sensor and the raspberry pi camera. We use different components like pi camera which is used to train and detect the objects, Arduino which control the wheels of the car and the ultrasonic sensor to calculate the distance. This autonomous car is the prototype to the self-driving cars which is the present growing advanced technology in the present scenario.
Akshay SachanAman Verma
Ms. Chahat Sharma
Automated essay evaluation and feedback system
Automated Essay Scoring (AES) is a great research area to analyse the text andmatching human expertise. AES is one of the most challenging activities in Natural Language Processing (NLP). It makes use of NLP and Machine Learning (ML) techniques to predict the score and match with the human like grading system. It has been around since the 1960s, with the first automated grading system â€“ Project Essay Grade (PEG), proposed by Ellis Page. Since then, there have been multiple systems that look at providing either a holistic score to the essay or to score individual attributes of the essay. Examples of a few online systems include Grammarly, Paper-Rater, ETS e-rater, IntelliMetric by Vantage.
Monika YadavNeeraj Negi
Dr. Neeta Verma
Intelligent traffic lights control by fuzzy logic
Traffic congestion is a major concern for many cities throughout the world. Developing a sophisticated traffic monitoring and control system would result in an effective solution to this problem. In a conventional traffic light controller, the traffic lights change at constant cycle time. Hence it does not provide an optimal solution. Many traffic light controllers implemented in
current practice, are based on the â€˜time-of-the-dayâ€™ scheme, which use a limited number of predetermined traffic light patterns and implement these patterns depending upon the time of the day.
A traffic light controller based on fuzzy logic can be used for optimum control of fluctuating traffic volumes such as oversaturated or unusual load conditions. The objective is to improve the vehicular throughput and minimize delays. The rules of fuzzy logic controller are formulated by following the same protocols that a human operator would use to control the time intervals of the traffic light.
Sumit SharmaShobhit Singh
Library management system using Alexa
This is a speech processing application which helps in operating a website, library management system in this case. It will control a website using speech through the alexa interface in Amazon Echo.The website can be easily controlled through voice after alexa has been trained and the data is saved across the AWS account.
This project hence is an alexa skill which helps in automating the library management system which makes issuing and returning of books easier removing the need of the librarian and hence saving time and energy.
Akansha SinghHimanshu Kapoor
Mr. Gaurav Srivastava
Home security system: A face recognition system
A real-time face recognition system for home and office security using Local Binary Pattern Histogram (LBPH) algorithm. Facial Recognition has been effectively used to develop systems for the purpose of security in offices and homes as well as keeping track of criminal activities through identification by these surveillance systems. Variations in pose and angles of human faces as well as illumination has been a major limiting factor for most face recognition algorithms but LBPH algorithm takes care of both these variations. This paper highlights the advantage of this algorithm over the others and also shares the reasons for preferring its OpenCV implementation over MATLAB.
Internet of Things (IoT) conceptualizes the idea of remotely connecting and monitoring real world objects (things) through the Internet. When it comes to our house, this concept can be aptly incorporated to make it smarter, safer and automated. This IoT project focuses on building a smart wireless home security system which sends alerts to the owner by using Internet in case of any trespass and raises an alarm optionally. Besides, the same can also be utilized for home automation by making use of the same set of sensors. The leverage obtained by preferring this system over the similar kinds of existing systems is that the alerts and the status sent by the wifi connected microcontroller managed system can be received by the user on his phone from any distance irrespective of whether his mobile phone is connected to the internet. The microcontroller used in the current prototype is the TI-CC3200 Launchpad board which comes with an embedded microcontroller and an onboard Wi-Fi shield making use of which all the electrical appliances inside the home can be controlled and managed
Stock Prediction using Yahoo Finance & Google Trends
Stock price forecasting is a popular and important topic in financial and academic studies. Time series analysis is the most common and fundamental method used to perform this task. This project aims to combine the conventional time series analysis technique with information from the Google trend website and the Yahoo finance website to predict weekly changes in stock price. Important news/events related to a selected stock over an eighteen-year span are recorded and the weekly Google trend index values on this stock are used to provide a measure of the magnitude of these events. The result of this experiment shows a significant correlation between the changes in weekly stock prices and the values of important news/events computed from the Google trend website. The algorithm proposed in this project can potentially outperform the conventional time series analysis in stock price forecasting.
Driving Bot Using Machine Learning
This project attempts to solve real-world problems via machine learning techniques. The idea is to build a system that can learn to play a game using the training dataset. The input is given to the system frame-per-frame via Image Grabber. The machine then processes the frame using OpenCV (Open Computer Vision Library) and analyzes the changes in the environment. The machine will generate inputs for the game environment using old dataset and learns at the same time. The inputs are keystrokes use to control the game environment. To keep the project simple we have focused on only one task in the game i.e to drive a car. We would make a bot that will make decthe ision to steer the car on the road. At first we used edge detection to detect the boundaries of the road and tried to keep the car on the road, at this stage we, are not concerned about the collisions. Later we collected training dataset and trained our neural network to avoid cola lision and drive more efficiently.
Avnish Kr. YadavPranay RajputUtkarsha Kappar
Ms. Nidhi Agarwal
Person of Interest â€œThe Machineâ€
The aim of the project is to develop a system that is capable of collating all sources of information to predict and identify, in advance, people planning terrorist acts. As an artificial super intelligence, its objective is to predict and prevent imminent terrorist attacks and it does so by analyzing immense amounts of surveillance data.The machine sorts through all available information and categorizes persons of interest into relevant (national security) and irrelevant (ordinary risk) cases. The project uses machine learning to determine the identity, location and intentions of monitored persons by infiltrating domestic organizations to analyze their databases from various sources such as video footage, phone calls, GPs, electronic transactions, e-mail, social media.
Lalit Kr. SharmaKunwarPrashant
Ms. Shelly Gupta
Behaviural Analyzer of Twitter Data using HIVE
In todayâ€˜s highly developed world, every minute, people around the globe express themselves via various platforms on the Web. And in each minute, a huge amount of unstructured data is generated. This data is in the form of text which is gathered from forums and social media websites. Such data is termed as big data. These opinions can be mined using various technologies and are of utmost importance to make predictions or for one-to-one consumer marketing since they directly convey the viewpoint of the masses. Here we propose to analyze the Behaviours of Twitter users through their tweets in order to extract what they think. Hence we are using Hadoop & Hive for Behaviour analysis which will process the huge amount of data on a Hadoop cluster faster.
Abhishek MishraPratik Singh
Rehan Ali Mehdi
Mr. Amit Kaushik
Optimised and Energy Efficient Routing algorithm in Heterogeneous Wireless Sensor Network
The Wireless sensor network has been highly focused research area in recent times due to its wide applications and adaptability to different environments. In Wireless sensor networks, several routing protocols have been developed to improve the various parameters such as stability period, throughput etc . After a homogenous protocol leach, many heterogeneous protocol were developed which successfully improved the routing in WSNs. The energy-constrained sensor nodes are always under consideration to increase their lifetime. In this project simulation work we have used the advantages of two approaches i.e. fuzzy c-means clustering and neural network to make an energy efficient network by prolonging the lifetime of the network. The cluster formation is done using FCM to form equally sized clusters in the network and the decision of choosing cluster head is done using neural network having input factors as distance from base station, heterogeneity and energy of the node etc. Also, we have incorporated a reactive wireless senor network protocol i.e. TEEN in our project to make our network communication more efficient and optimised one .
Medical Image Indexing Using Edge Pattern and Histogram Oriented Gradient
Increasing use of computer imaging in the medical field leads to the large number of digital images generated at hospitals every day such as Computer Assisted Tomography (CAT), Magnetic Resonance Imaging (MRI), X-ray and ultrasound, Mammography etc. In this scenario, it is necessary to develop appropriate information systems to efficiently manage these collections. Another problem is the complexity of image data, and these data can be interpreted in various ways, thus raising the question of how to work in order to manipulate these data and represent or establish policies to its content. Content-Based Image Retrieval is an approach that retrieves images similar to a user-defined specification or pattern i.e image retrieval based on content properties (e.g., shape, color, texture), usually encoded into feature vectors. This project implements a CBIR technique that uses Peak valley edge patterns(PVEP) and a histogram of oriented gradients (HOG) as a feature vector for image retrieval. The technique is tested on the images from ABIDE(Autism Brain Imaging Data Exchange)database which includes Autism Spectrum Disorder(ASD) magnetic resonance imaging (MRI) images. Results have shown that PVEP and HOG together give better results than using PVEP alone as HOG considers Gradient intensity and orientations, along with directional edge recognition of PVEP.
Ms. Alka Singhal
RAIS â€“ Remotely accessible irrigation system
Animesh BasakAbhay Jain
Ms. Neeta Verma
Simple Term for Artificially Intelligent Reminiscent (S.T.A.R.)
Neural machine translation or NMT is a proposed defined approach to the machine translation. Unlike the traditional SMT i.e. statistical machine translation, NMT focuses on constructing a single neural network that can be jointly aligned to maximize the performance, translation and efficiency. The models that are proposed for NMT belongs to a group of encoders and decoders.It e ncodes a source text or sentence into a vector of fixed length from which a decoder generates an appropriate translation.Neur al Machine Translation (NMT) Model on Python with Tensthe or flow is used and valuthe e of different parameters are analyzed during the analysis of the model.Para meters used for analysis are Perplexity, Bleu rate, WPS(words per second) on different models were examined.
Abhishek DubeyAyush Gupta
Mr. Pranshu Saxena
Improved Recommender System using sentiment analysis
Traditional Collaborative filtering algorithm works by using only the past experience of a user. To overcome the limitations of the traditional collaborative algorithm, an item based collaborative filtering system was introduced. In this algorithm, items are recommended by finding similar items to the ones rated by the user and the close ones are recommended. However, this approach does not consider the overall opinion of users about an item before recommending it and so does not always makes the best recommendations. In this paper, an improved recommender system is proposed. A dictionary of sentiment scores is created. These sentiment scores are calculated by finding the probability of the reviews to be positive. This sentiment score is used by an item based collaborative filtering system to improve the recommendations and filter out items with overall negative user opinion. The performance of the proposed system is compared with previous work done in this field.
Mayank KumarAbhishek kr Rai
Dr. Tripti Sharma
Air Monitoring and Process Management
The level of pollution has increased with times by lot an of factors like the increase in population, increased vehicle use, industrialization and urbanization which results in harmful effects on human wellbeing by directly affecting health the of the population exposed to it. In order to monitor the quality of air, a Sensor network (SN) based new framework is proposed which is based on data acquisition and transmission. The parameters of the environment to be monitored are chosen as temperature, humidity, volume me of CO, volute me of CO2, detection of leakage of any gas â€“ smoke, alcohol. The values of these parameters are transmitted by using Arduino Uno R3 microcontroller to a base station where they are being monitored. The value of temperature and humidity are also transmitted over Application , so that every person in the range of the system can check it over their smart phones and laptops as these parameters hold importance to everyone. CO, a dangerous parameter is monitored with an extra precaution.
Combining Raster and Vector Images for Secure Steganography
Securing information today is increasingly obnoxious as hackers are at ease of internet to detect vulnerabilities for exploitation and communicate same between malicious peers through various means including â€œdark websâ€. Cryptography alone cannot withstand attacks by such adversaries. Thus another information security domain steganography must also be employed. Steganography is hiding of confidential information in covert channel called cover media. There are several covert channel based steganographyâ€™s, where covert channels can be text, network protocol s, images, audios, and videos. The most common is digital image steganography with digital image as cover media. Most digital image steganography algorithms are weak. A new algorithm is proposed in this project that works by combining raster and vector graphic based digital image steganography to counterfeit the loop holes in standardized steganography. Here raster and vector graphics are two different forms of representing digital image in computer graphics.
Healthcare Analytics is a term used to describe the healthcare analysis activities that can be undertaken as a result of data collected from areas within healthcare that includes claims and cost data, pharmaceutical data, research and development data, clinical data (collected from electronic medical records), patient behavior and sentiment data (patient behaviors and preferences) and other types of health related data. Big Data in healthcare is being used to predict epidemics, cure diseases and to avoid preventable deaths. The main aim of research is to describe the nascent field of big data analytics in healthcare and to perform real-time analytics on healthcare using Big Data, and help executives and research professionals from hospitals, health systems and other healthcare provider organizations to identify and understand big data analytics for innovative uses of data assets that can enable them to reduce costs and provide more accessible care.
Mukul Sharma Nupur Pandey
Dr. Anchal Jain
Congent- Sentiment Analysis of Google Search Result for Automatic Summarized Review Generation
Sentiment classification system is designed in mobile environment, that works as a mobile application. System extracts reviews and information present about a product or service over the internet and uses sentiment analysis to categorize this data as positive or negative based on the calculated polarity. Finally, the collected data is presented in the form of charts showcasing the percentage of positive and negative mentioning about the product or service over the internet. Machine learning approach, NaÃ¯ve Bayes algorithm is used. The classifier takes unstructured product reviews as input, performs pre-processing, calculates polarity of reviews, outputs the best suited category. In propose system sentiment accuracy and response time is focused.
Domotics system are basically the intelligent system for houses and apartments to control several issues as security and light or climate devices. Modern systems generally consist of switches and sensors connected to a central hub sometimes called a â€œgatewayâ€ from which the system is controlled with a user interface that is interacted either with a wall-mounted terminal, mobile phone software, tablet computer or a web interface.
Real Time Analysis of Data in Credit Card Transaction System
Today is the time of Plastic money. Cards are accepted worldwide for every transaction. This system provides offers at the instant you process one transaction and promotes you or motivates you to attempt a consequent second transaction. We aim to sell this software to a network provider like Visa, MasterCard or American Express who, with the latest technology would be able to conquer the market with the promotional offers. The Algorithms for offer generation, uses location of the person which is the key element and the it is based on the Spark framework which is latest and trending in Real Time Analysis of Data.
Name of the student
1303010187 1303010210 1303010195 1303010133
Shubham Kumar Singh
T M Mithun
Ms. Shweta Chaku
2d Platform Multiplayer Gaming
The game is a hybrid of endless runner and fighting game genres with multiple types of enemies to be fought using unique moves thus making every move a challenging and exciting one.
MS. PRACHI VATS
CROSS PLATFORM REALTIME MULTIPLAYER GAME
As the title suggests, the project aims at developing a game that features Inderprastha Engineering College and allows multiple players to interact with each other in real time without being bounded by the platform they are using.
1303010115 1303010062 1303010124 1303013075
MS. ANJALI SINGHAL
THE-VOTE REPLACEMENT OF EVM
This project was related with online voting. In this a Android App was designed to give vote. In this Thumb impression was used for identification.
The project is made with an objective to develop or improve existing algorithms to make the iris recognition accurately in possible on noisy iris images. Further, Performance evaluation of proposed frame work has been done by comparing the performance of existing iris recognition system.
1303010036 1303010041 1303010048
DR. TRIPTI SHARMA
PARKING MANAGEMENT SYSTEM
Parking is an ever-growing challenge in cities and towns across the country. This project allows only authorized user to access various function and processed available in the system. It provides parking facility and parking information to user.
Linear Feature Extraction of Indian Ocean Sea Surface Temperature
The Reynoldâ€™s Reconstructed SST data (ERSST) has been used for determining the first Principal Component of the Indian Ocean SST index. It is observed that the leading PC explains a better portion of the data. This can be used for the prediction of the Indian Summer Monsoon Rainfall forecasting using a reduced feature space.
Dr. K. C. Tripathi
Expert System for Sea Surface Temperature Characterization
Artificial Neural Network model has been used for the time series prediction of the Sea Surface Temperature (SST) anomalies over the Indian Ocean region. The correlation analysis has been used for the determination of the predictors. The prediction accuracy and other statistical verification scores are determined and explained.
BTech Cafe is a portal which provides a platform for B.Tech. students where they can get all the relevant resources e.g. old question papers, sessional papers, list of practicalâ€™s, abstracts of projects etc. at one single abstract joint.
Mr. Gaurav Agarwal
Multi-Purpose Card System
At present, smart cards are not implemented in a unified way. Deploying multiple applications on a single card is the need of the hour as it saves time and money. The objective of the present project is to create a multi-purpose card system in which multiple functionalities are utilized through a single card.
This project describes the advantages and possible applications of network traffic and profiling. The techniques presented here in allow extraction of exceptionally useful an important information about the particular network under consideration.
Aman Deep Singh
Dr. Tripti Sharma
It is an Internet based application to solve the problem of availability of specialist doctors in the areas where they are not available. This software provides a platform for local doctors to communicate with specialist doctors at remote locations.
Ms. Kumud Kundu
Scale Invariant Feature Transformation
This project deals with robust feature selection process. It selects features which are scale invariant. This application may be used in Face Recognition.
The project provides a platform for party goers to interact with the DJ via android application. A person in the party sends songs of his liking to the DJ who adds the same to the playlist. The project is made on Android platform.
Mr. Alok Katiyar
The project aims to create an educational application which can represent the solar system. The software shall be useful for geography students. This project is based on image/video simulation through mobile in 3D scenario.
Mr. Gaurav Agarwal
Image Steganography Using DWT Technique
An algorithm is proposed for embedding and extracting the secret image embedded behind the cover gray scale image. Signal to noise ration and mean square criteria are used to evaluate the performance.
Dynamical Aspects of Prediction by Artificial Neural Network When Predicting in Chaotic Regime
The project aims to analyse the aspects of prediction by statistical algorithms such as the neural network in the areas of dynamical error growth and decay. The observation is that the prediction capability by statistical models is independent of dynamical error growth and decay.
Mr. Pranshu Saxena Dr. K. C. Tripathi
Pattern Classifier for the Diagnosis of Dengue
Diagnosis of mosquito borne diseases is a tough task as the symptoms are almost similar. The project aims to differentiate between two prominent mosquito borne diseases i.e. Dengue and Malaria based on the symptoms that may appear as common by nascent practitioners. The sampling for data is done by local hospitals. The Artificial Neural Network model is used for the training purpose. The accuracy of the model outcasts the accuracy claimed by registered practitioners.
Ms. Anchal Jain
Dynamic Encryption Algorithm using DNA Operation
Use of dynamic S box in substitution is explored to increase the security of encryption algorithm. In this regard, here a dynamic S box based encryption algorithm is proposed. The S box is based on DNA computing coupled with the power of chaotic maps. The proposed algorithm has reasonable security features.
The project envisions a fully automatic anti-collision train signaling system. It is expected to be helpful to Railway collision detection. The primary goal of the anti-collision system is to identify such collision points and to report these error cases to main control room.
Biometric Based Authentication Mechanism for Voting Systems
A new voting method based on complete biometric traits of voters which are saved in the government database as Adhar (U-ID) is proposed. It is an effort to prevent fraud and illegal voting. A comparison of conventional methods of voting is done.
Ms. Kumud Kundu
Image Dimension Reduction
The project deals with Data Dimensionality curse problem. An application in the area of classification of images for a large data size is demonstrated.
Dr. K. C. Tripathi
Gesture Based Image Processing
The project is an application of image processing tools. Various image processing techniques have been applied. The input is fed by human gestures given by hand movements. The application area includes signaling through human gestures from long distances.
Comparative Study on Facial Recognition and Detection Techniques
The project aims to explore various face recognition techniques. Dimension reduction techniques such as the PCA and LDA are explored and applied on facial images. The outputs of PCA and LDA have been discussed and explained.
The project focuses on the image security sharing techniques for safe transmission purpose using the wavelet transform. The study uses the images as digital data. The wavelet transform coefficients of a copyright image are used to embed the watermark. The logo as the image copyright is used in the form of a binary image.The technique used for encryption is â€˜Secure Force algorithmâ€™ (SF). SF has a simple architecture consisting of basic mathematical operations and requires less computational resources as compared to other standard encryption algorithms.
Ms. Aruna Bhat Dulloo
GPS identification using radio frequency locator system
The project suggests a new perspective to use technology to protect a woman. This system keeps track of the location of the user and sends the information to authorities and relatives in a risky situation.
The project aims at exploring the various spatial based object tracking approaches. Two popular methods: the correlation method and the template matching method have been implemented and the results discussed. It is observed that the spatial based tracking methods suffer from limited scope of movement detection and hence the tracking in the frequency domain such the Fourier and Wavelet domains is recommended.
Approximate Semantic Matching of Events in Publish/ Subscribe System
This project is used to address semantic coupling within event based systems. A vocabulary free subscription is proposed together with the use of approximate semantic matching of events.
Mr. Aditya Kumar
Improving Peak Signal to Noise Ratio of Color Images Using Moment Preserving Algorithm
Block Truncation Coding (BTC) is a streamlined and competent application for image compression. An improved BTC algorithm, namely Enhanced Block Truncation Coding (EBTC) is presented in this project.
Ms. Anjali Singhal
Delhi Metro Alert
This project was based on Android Application. It has Delhi metro map, alarm. User can set their destination, while boarding. When the metro reaches that place. It gives alarm. This application was GPS connected.
1003010076 1003010018 1003013035
Vyomesh Km Singh
Dr. Anchal Jain
Improved Indexed Based Symmetric Encryption Algorithm Using DNA Approach
A new image encryption method is proposed. The algorithm presented in the report is an integrated approach which couples the power of S-Boxâ€™s and the two-dimensional logistic map. Theoretical analysis and experimental results show that the proposed algorithm is reasonably secure and faster in run time compared to indexed based approach.
Ms. Diksha Dani
It is database based project to run the SQL queries automatically to assist novice users. The project embeds a layer between the database and the user so that a comparatively unskilled user can make queries related to database.
Ms. Aruna B Dullo
Extraction of Information Through Augmented Reality
The project attempts to superimpose graphics, audio and other sensory enhancement over a real world environment, in order to give a better perception of the environment.
The project aims at providing a platform for efficient transfer of visual information from one system to another. The application area includes providing the platform for laboratories. The faculty can carry out the tasks on his system and the same will be visible to all the students on their system.
Mr. Gaurav Agarwal
Chaos Based Watermarking Using DWT
This project provides a novel watermarking technique based on Discrete wavelet transform. Further a chaos- based authentication mechanism is used to improve the security aspect. The algorithm is resistant to many cryptographic attacks and can be used to provide authenticity for the digital document.
Aggrandisement of Qualitative and Quantitative Assessment in Sequence of Frames
Video compression plays an important role in real-time scouting/video conferencing applications. For the entire motion based video compression process, motion estimation is the most computationally expensive and time consuming process. In the present study we have developed a new algorithm to reduce the temporal and spatial redundancy in the video sequence.
Ms. Neeta Verma
This project is used for Human Computer Interaction (HCI) where the control of mouse cursor movement and click events of the mouse are controlled using hand gestures.