2019-20

Permanent URI for this collection

http://172.16.1.132:4000/handle/123456789/12601

Browse

Now showing 1 - 20 of 86

Face Emotions Recognition Using Image Processing and Machine Learning
(2020-08-31T11:20:54Z) POOJA, NIMJE; ROSHLIN, ACHARYA; PRACHI, SINGH
Recognizing emotion from images has become one of the active research themes in image processing and in applications based on human-computer interaction. Extensive and proper research has being carried out to explore the relationships between these channels and emotions. Thus a neural network based solution combined with image processing is used in classifying the universal emotions. Coloured frontal face images are given as input to the prototype system. After the face is detected, image processing based feature point extraction method is used to extract a set of selected feature points. Our system focuses on live images taken from the webcam. Finally, a set of values obtained after processing those extracted feature points are given as input to the neural network to recognize the emotion contained.
Image Similarity using Deep CNN
(2020-08-31T11:30:32Z) Shreya, Pandey; Nikita, Upadhyay; Yashwardhan, Singh Shekhawat
This project entitled “Image Similarity Using Deep CNN” comes with the chief objective to find the similarity index of two images using deep learning and CNN, even if the images are similar and taken at different time interval (let’s say a time interval of mili-second) it will still have a different similarity index. The same project can be used to delete multiple images with similarity index falling in the same range from the mobile gallery which will not only save time but will also be efficient throughout. An application will be designed to differentiate or in other words find the similarity index between two images, by various algorithms, deep learning and neural networks will be used for application designing. In the process of neural network image recognition, the vector or raster encoding of the image is turned into constructs that depict physical objects and features. Systems can logically analyse these constructs, first by simplifying images and extracting the most important information, then by organizing data through feature extraction and classification. Neural network image recognition algorithms rely on the quality of the dataset the images used to train and test the model. Here are a few important parameters and considerations for image data preparation.
Narcotics Elimination Through Intelligent Ultrasound
(2020-08-31T11:34:59Z) MADHUMITHA, R; ANKIT, TYAGI; AYUSH, BHARDWAJ
Surgery is not easy for everyone. It brings discomfort and often involves significant post-surgical pain. Currently, patient pain is frequently managed through the use of narcotics that bring a bevy of unwanted side effects. Our solution aims to improve pain management through the use of indwelling catheters that block or mitigate pain at the source before the surgery begins. Pain management catheters reduce dependence on narcotics and speed up patient recovery. In our solution, we identify and segment a collection of nerves called the Brachial Plexus (BP) in ultrasound images. Brachial plexus block remains the only practical alternative to general anaesthesia for significant surgery on the upper limb. It provides a superior quality of analgesia and avoids the common side-effects associated with general anaesthesia such as postoperative nausea and vomiting. Our model can identify nerve structures in a dataset of ultrasound images of the neck. Doing so would improve catheter placement and contribute to a painless surgery. The dataset is a large training set of images where the nerve has been manually attributed by doctors. These doctors were trained by experts and instructed to attribute images where they felt confident about the existence of the BP nerve.
Phishing Website using Extreme Learning Machine
(2020-08-31T11:45:33Z) G., JAGANATHAN; W., VISHNU SARAN; S., MANIVASAN
Internet is an essential part of our life. Internet users can be affected from different types of cyber threats. Thus, cyber threats may attack financial data, private information, online banking and e-commerce. Phishing is a type of cyber threats that is targeting to get a private information such as credit cards information and social security numbers. There is no specific solution that could detect whole phishing attacks. Phishing website has become a crucial concern in cyber security applications. It is performed by fraudulently deceiving users with the aim of obtaining their sensitive information such as bank account information, credit card, username and password. In this study, an intelligent model for detecting phishing web pages based on Extreme Learning Machine. Types of web pages are different in terms of the features. Hence, we must use a specific web page features set to prevent phishing attacks. A model based on machine learning techniques to detect phishing web pages has been proposed.
Interactive AI Chat Bot
(2020-08-31T11:50:20Z) SHASHANK, P
A computer program designed to simulate conversation with human users, especially over the internet is called a CHATBOT. The way people communicate with businesses has been evolving at a rapid pace. For years, in-person meetings and phone calls were the dominant means of communication. Then, with the rise of the internet, a multitude of new options became available, from email, to social media, to mobile apps, to filling out a form on a website and waiting for a follow-up. More recently, the rise of real-time messaging has led to a fundamental shift in how people prefer to connect with businesses What all of these communication channels have had in common, including online chat, is that they’ve traditionally relied on humans to do the communicating. But now that chatbots have arrived on the scene, this dynamic is starting to change.
Overcoming Disability: Image Processing Using CNN
(2020-08-31T12:01:32Z) ALANKRITA, SRIVASTAVA; NAGASUHAS, SHASTRY M; ASIF, ALAM
Overcoming Disability: Image Processing using CNN is a program built to enable humans to interact and communicate that ensure to overcome individual disabilities. This project is built using Convolution Neural Network which uses deep machine learning algorithm to compute and manipulate data which are entered by human via webcam and present a favourable output. Behind the neural network a certain amount of data is stored as the final answer which are then match with the predicted output and is presented as a result of what the user wants to convey. The computation can be as simple as “a cat” or can dive deep into complex answers such as “collateral damage”. The CNN in this project acts as the brain of the entire program, and perform all the computation from fetching data from the user, compute with the possible result and finally come up with the most efficient output. CNN is used as it is the backbone of machine learning for image processing, here in this project the input is not entered by any external device, instead it is depicted by the user by simple gesture with their hands, which make it user friendly for those who find difficult to see but can convey their message as they feel the world with their hands. Since, we use deep learning mechanism to train our AI, with each iteration of processing it develop itself and evolve into a better version of itself every time it executes a correct prediction or can correct itself if it encounters with any faults in the processing.
Data-Oriented Computer Simulation
(2020-08-31T12:04:33Z) AKASH, S
The project aims to find problems in developing a computer simulation, by programming a 3D traffic simulation system from scratch. Here, problems refer to both development and performance problems. The idea here is to agree to the fact that most of the modern commercial hardware is pretty generalized, in that, instead of software defining the hardware, the reverse is what is generally practised. So, if our target is to build a high performance, easily scalable software simulation using a new and well suited programming language, our first and foremost objective has to be to analyze how simulations is/were being developed and to put them to criticism.
Emotion Recognition
(2020-09-01T05:54:55Z) BELLAM, BHARATH; BHAVANA, N RAJ; CHAMALA, GEETHIKA
The interest on computing by emotional aspects is now a days increasing as many applications are in demand. The project Emotion recognition basically refers to computer systems that attempt automatically to recognise the facial expression from visual features. For human-computer interaction, facial expressions form a platform for non- verbal communication. As we all know, humans are very capable of detecting emotions based on facial expressions, but the same cannot be said for machines. A significant amount of processing is required in order to account for various features, such as colour, orientation, posture and so on. Our aim is to make a system which is capable of accounting for the mentioned variability, and accurately detects facial expressions by using concepts of machine learning. We’ve implemented our idea on a laptop by using a Haar Cascade Classifier, Tensorflow and OpenCv. The idea behind this implementation was to make the software more portable as possible.
Home Automation Using IOT
(2020-09-01T06:00:08Z) SAINATH, G V; SUHAIL, A; BHARATH, RAJ S
This project presents the general structure of Home Automation System (HAS) with minimal effort and remote framework. This framework is intended to help and give support so as to satisfy the necessities of old and impaired in home. Additionally, the brilliant home idea in the framework improves the standard living at home. The switch mode and voice mode are utilized to control the home machines. The video input is gotten in the android application which streams the video of IP Camera. The fundamental control framework actualizes remote innovation to give remote access from PDA. The structure remains the current electrical switches and gives more security control on the switches with low voltage actuating technique. The switches status is synchronized in all the control framework whereby each UI shows the constant existing switches status. The framework expected to control electrical apparatuses and gadgets in house with moderately minimal effort structure, easy to understand interface and simplicity of establishment.
Social Media Sentimental Analysis Using Machine Learning
(2020-09-01T06:06:22Z) DEEKSHITH, D V; M KALYAN, NAIDU; BRIJESH, DOBARIYA
Sentiment analysis or opinion mining is the computational study of people’s opinions, sentiments, attitudes, and emotions expressed in written language. It is one of the most active research areas in natural language processing and text mining in recent years. Its popularity is mainly due to two reasons. First, it has a wide range of applications because opinions are central to almost all human activities and are key influencers of our behaviours. Whenever we need to make a decision, we want to hear others’ opinions. Second, it presents many challenging research problems, which had never been attempted before the year 2000. Part of the reason for the lack of study before was that there was little opinionated text in digital forms. It is thus no surprise that the inception and the rapid growth of the field coincide with those of the social media on the Web. In fact, the research has also spread outside of computer science to management sciences and social sciences due to its importance to business and society as a whole. In this project, we will start with the discussion of the mainstream sentiment analysis research and then move on to describe some recent work on modelling comments, discussions, and debates, which represents another kind of analysis of sentiments and opinions. Sentiment classification is a way to analyse the subjective information in the text and Then mine the option. Sentiment analysis is the procedure by which information is extracted from the opinions, appraisals and emotions of people in regards to entities, events and their attributes. In decision making, the opinions of others have a significant effect on customers ease, making choice with regards to online shopping, choosing, events, products, entities. The approach is of text sentiment analysis typically work at a particular level like phrase, sentence or document level. This project aims at analysing a solution for the sentiment classification at a fine-grained level, namely the sentence level in which polarity of the sentence can be given by three categories as positive, negative and neutral.
Classifications on Online Shoppers Purchasing Intention
(2020-09-01T06:12:32Z) ELDHO, M M; CHETAN; KAVYA, A S
Due to today’s transition from visiting physical stores to online shopping, predicting customer behaviour in the context of e-commerce is gaining importance. It can in-crease customer satisfaction and sales, resulting in higher conversion rates and a competitive advantage, by facilitating a more personalized shopping process. By utilizing clickstream and supplementary customer data, models for predicting customer behaviour can be built. This study analyses machine learning models to predict a purchase, which is a relevant use case as applied by a large German clothing retailer. Next, to comparing models this study further gives insight into the performance differences of the models on sequential clickstream and the static customer data, by conducting a descriptive data analysis and separately training the models on the different datasets. The results indicate that a Random Forest algorithm is best suited for the prediction task, showing the best performance results, reasonable latency, offering comprehensibility and a high robustness. Regarding the different data types, models trained on sequential session data outperformed models trained on the static customer data by far. The best results were obtained when combining both datasets.
Convenience Store Assisting System for Visually Adaptive Using OCR
(2020-09-01T06:17:54Z) HARSHINI, K; SANDHYA, MN; SOWMYA, PB
Assisting a person with a visual impairment means, being able to handle cases that people without visual impairment accomplish without even realizing that they do so. Making a really helpful system with a range of interfaces exhibits a progression of difficulties, not which are all at first self-evident. These difficulties incorporate managing the individual through store, which incorporates a test of limitations, obstacles,and avoidance of individual person, and shopping for food. As there is no shop or framework where daze individuals or blind people can do shopping without the assistance of any third individual, and the visually impaired individuals can't satisfy their desires, so we require a framework where daze individuals can do shopping by their own particular without the assistance of others. Our proposed system is to design and develop a visual assistance system which will recognize the label on the product by Optical Character Recognition [OCR] in grocery store and then translate that text into speech. Through the speech, he/she will get the details about that product, which will help them to decide whether he/she wants to take that product or not. The principle goal of Ambient Assisted Living is to give blind people individuals the power to but themselves without the assistance of anybody or identify family unit stuffs independent from anyone else and to keep up their protection.
Traffic Sign Detection Using Deep Learning
(2020-09-01T06:25:24Z) MADHURA, RAMOJWAR; MEGHANA, MOHANDAS; HASIKA, CHELLADURAI
Automatic detection and recognition of traffic signs play a crucial role in management of the traffic-sign inventory. The study is about trying to determine and measure the advantages of training a detector and more generally full autonomous driving systems inside a virtual environment. Proper management of traffic-sign inventory is an important task in ensuring safety and efficiency of the traffic flow. Most often this task is performed manually. Traffic signs are captured using a vehicle-mounted camera and manual localization and recognition is performed off-line by a human operator to check for consistency with the existing database.
Cancer Prediction and Prognosis Using Deep Learning
(2020-09-01T06:34:23Z) INDUCURI, SWETHA; ISHITA, NILESH JOSHI
Early diagnosis of any disease can be curable with a little amount of human effort. Most of the people fail to detect their disease before it becomes chronic. It leads to increase in death rate around the world. Breast cancer is one of the diseases that could be cured when the disease identified at earlier stages before it is spreading across all the parts of the body. The medical practitioner may diagnose the diseases mistakenly due to misinterpretation. The computer-aided diagnosis (CAD) is an automated assistance for practitioners that will produce accurate results to analyze the criticality of the diseases. This method employed deep neural network (DNN) as classifier model and recursive feature elimination (RFE) for feature selection. DNN with multiple layers of processing attained higher classification rate than SVM. So, the researchers used deep learning method for hyper-spectral data classification The DNN with multiple layers of processing is applied to classify the breast cancer data. The system was experimented on Wisconsin Breast Cancer Dataset (WBCD) from UCI repository. The dataset partitioned into different sets of train-test split. The performance of the system is measured based on accuracy, sensitivity, specificity, precision, and recall.
Secured MapReduce Layer for Big Data Privacy and Security
(2020-09-01T07:10:52Z) J C, MONISHA; MEGHA, P
Hadoop has become a promising platform to reliably process and store big data. It provides flexible and low cost services to huge data through Hadoop Distributed File System (HDFS) storage. Unfortunately, absence of any inherent security mechanism in Hadoop increases the possibility of malicious attacks on the data processed or stored through Hadoop. In this scenario, securing the data stored in HDFS becomes a challenging task. Hence, researchers and practitioners have intensified their efforts in working on mechanisms that would protect user's information collated in HDFS. This has led to the development of numerous encryption-decryption algorithms but their performance decreases as the file size increases. In the present study, we have enlisted a methodology to solve the issue of data security in Hadoop storage. We have integrated Attribute Based Encryption with the honey encryption on Hadoop, i.e., Attribute Based Honey Encryption (ABHE). This approach works on files that are encoded inside the HDFS and decoded inside the Mapper. The ABHE algorithm shows considerable improvement in performance during the encryption-decryption of files.
Heart Disease Prediction Using Machine Learning
(2020-09-01T07:20:03Z) KARAN, D K; MAQDOOM, BAIG; NINGARAJ, SINDOGI
Heart disease is one of the most critical human diseases in the world and affects human life very badly. In heart disease, the heart is unable to push the required amount of blood to other parts of the body. Accurate and on time diagnosis of heart disease is important for heart failure prevention and treatment. The diagnosis of heart disease through traditional medical history has been considered as not reliable in many aspects. To classify the healthy people and people with heart disease, noninvasive-based methods such as machine learning are reliable and efficient. In the proposed study, we developed a machine-learning-based diagnosis system for heart disease prediction by using heart disease dataset. We used seven popular machine learning algorithms, three feature selection algorithms, the cross-validation method, and seven classifiers performance evaluation metrics such as classification accuracy, specificity, sensitivity, Matthews’ correlation coefficient, and execution time. The proposed system can easily identify and classify people with heart disease from healthy people. Additionally, receiver optimistic curves and area under the curves for each classifier was computed. We have discussed all of the classifiers, feature selection algorithms, preprocessing methods, validation method, and classifiers performance evaluation metrics used in this paper. The performance of the proposed system has been validated on full features and on a reduced set of features. The features reduction has an impact on classifiers performance in terms of accuracy and execution time of classifiers. The proposed machine-learning-based decision support system will assist the doctors to diagnosis heart patients efficient.
High End Security Systems for Banking
(2020-09-01T07:26:40Z) KARTHIGA, R; MONI KRITHIKA, S; POOMARI
Hacking a bank’s server is only one of the many known ways that criminals can get their hands on account holders’ card details and their hard-earned cash. Some methods are clever and tactical. Some can be crude. And others are more destructive and dangerous. Whatever the cost, we can bet that criminals will do whatever it takes to pull off a successful heist. This has led to major structural changes in the banking system. With the ever increasing number of ATM systems and self-service stations to carry out mobile banking operations the need for higher, better and advanced security systems for the banking systems has become mandate. Due to the advent of online services and ATM facilities provided by these financial institutions providing the users more conveniences, the security in the banking sector can no longer be limited and restricted to physical security systems such as video surveillance and access control devices but it must be widened and broadened to include logical access controls as well. This gives rise to another challenge in the banking sector as to secure and protect the customer information on the Internet and fortify the online transactions.
Secure File Sharing on HDFS Using Encryption
(2020-09-01T07:34:16Z) KARTHIK, MK; MANTHAN, KAMILA
This project intends to solve the problem of file sharing by providing security. The objective of this project is to develop an application which provides secure sharing of files in a Hadoop environment. File sharing is the process of sending and receiving of different types of file within the same network or different network. File sharing is done using techniques like file storage, distribution and transmission. It might be actualized through an assortment of ways. Hadoop is an open-source programming structure utilized for circulated capacity and handling of dataset of enormous information. Managing big data is being a big problem to many organizations. Due to this reason, file sharing faces many problems. The project is built to tackle the shortcomings of putting into use the outmoded methods of data record storage and management and also support the automation of tasks carried out in the process of data management. Cloud computing is currently getting considerable attention in several communities, which provides the user’s software resources, storage, and massive computing on demand. Focused on the operation of secure file sharing we use encryption algorithms as well as one time password algorithm to provide security to the file. This provides a centralized system that presents a platform to perform basic operation of handling a wide range of distinct information in an efficient way according to the convenience of the user with the help of a simple, user friendly interface provided as application software.
Road Accident Analysis Using Machine Learning
(2020-09-01T07:40:05Z) KIRTI R, NAMBIAR; SHANMUGAPRIYA., J; SUSHMITA, PATIL
Today, traffic safety is one of the main priorities of governments. Considering the importance of topic, identifying the factors of road accidents has become the main aim to reduce the damage caused by traffic accidents. In this paper, we have applied the concepts of machine learning and data mining to identify the various factors that affect road accidents and its severity. The application takes various inputs such as weather conditions, road conditions, time of day etc. and uses machine learning algorithm (Decision tree algorithm) to calculate the severity of a possible accident on a scale of 1 to 4 (1 being the least and 4 being the most severe). This data can be used for analysis of future inputs and improves the accuracy of the system output. This model can further be improved to send the report of the accident to the concerned authorities, such as hospitals, ambulance and insurance agencies and can therefore prove to be very helpful in reducing accident fatality rates in the country.
Detecting Phishing Website Using Machine Learning
(2020-09-01T07:46:16Z) M JAYA, BHARATHI; TEJA PRAVEEN, KUMAR; B PREETHI, REDDY
The criminals, who want to obtain sensitive data, first create unauthorized replicas of a real website and e-mail. The e-mail will be created using logos and slogans of a legitimate company. The nature of website creation is one of the reasons that the Internet has grown so rapidly as a communication medium. Phisher then send the "spoofed" e-mails to as many people as possible in an attempt to lure them into the scheme. When these e-mails are opened or when a link in the mail is clicked, the consumers are redirected to a spoofed website, appearing to be from the legitimate entity. We discuss the methods used for detection of phishing Web sites based on url importance properties.

Browse

Browsing 2019-20 by Issue Date

Results Per Page

Sort Options