Browse Articles
Discover research articles across all indexed journals
IMPLEMENTATION AND DESIGN OF EYE SPY ATTENDANCE TRACKER
Artificial intelligence enabled computer vision allows for the automation of various processes, computer vision for face recognition and time tracking can be quintessential in sectors where attendance and presence of various individuals needs to be tracked efficiently with as much handsfree experience as possible by employing data training sets and deep learning algorithms. The integration of this system with a robust and secure database to register information creates an extremely useful system for automated attendance tracking with minimalist human intervention.
AGRICULTURE ANALYSIS USING ENSEMBLE LEARNING TECHNIQUES
Agriculture is an important occupation in India. Agriculture contributes to approximately 17% of India’s Gross Domestic Product and it’s still the most popular occupation amongst 70% of India’s population. Due to the various climatic factors the agriculture productivities in India are continuously decreasing over a decade. Farmers are still struggling when it comes to picking the right decision for better crop production. Thus, to accelerate the crop production, different technologies are being proposed worldwide. Data mining and machine learning in agriculture is a new approach for predicting crop yield production. The reasons for continuously decrease in production over a decade was studied mostly using regression analysis. The emerging technologies can change the situation of farmers and help in decision making. For this we are using Machine learning algorithms called k-means algorithm, AdaBoost and Random-forest algorithm. So, these algorithms take in different parameters to give us best recommendations which not only leads to better yields but also minimum use of resources and capital. In this model, the algorithm classifies the dataset into 4 classes namely, 0-poor yield, 1- average yield,2-good yield, 3-very good yield. This model predicts the expected accurate yield by considering the parameters like, state name, crop, production of crop, and season.
GLOBAL IMAGE IDENTIFIER
Many of the things, signs, and symbols we encounter when exploring the world might not be familiar to us. A global image identifier must be created to minimize confusion and misunderstanding. We shall use the less-than-universal traffic signs as an example. Road signs are strategically positioned to safeguard drivers’ and tourists' safety. Additionally, they offer instructions on when and where cars should turn or not turn. The traffic signs on the road express several cautions. In India, there are 400 traffic accidents per day, according to official statistics. Road signs ensure the safety of both automobiles and pedestrians by preventing accidents from occurring. Additionally, traffic signals reduce the incidence of traffic offences by ensuring that drivers follow certain laws. All users of the road, including pedestrians and automobiles, should give priority to traffic signals. For a multitude of reasons, including difficulty focusing, tiredness, and lack of sleep, we fail to see traffic signs. Other reasons for ignoring the indicators include impaired vision, the outside world's influence, and environmental factors. There is a critical need for a system that can recognize traffic lights automatically.
DIABETES PREDICTION USING MACHINE LEARNING
Diabetes is a fatal chronic illness that adversely affects every system in the body. This disease affects many people, and its adverse effects cause a sizable number of sufferers to pass away each year. High blood glucose levels are a concern for diabetic patients. It is quite challenging to make a robust and accurate prediction of diabetes due to the tiny amount of labeled data and outliers (or incomplete data) in the datasets for diabetes. We are putting up a reliable framework for diabetes prediction in this literature. Untreated diabetes may result in hearing loss, renal damage, heart and blood vessel disease, poor wound healing, and a few skin problems. For the classification, early diagnosis, and prediction of diabetes, a machine learning (ML)-based strategy has been put forth.
BLOCK CHAIN-BASED BIOMETRIC AUTHENTICATION SCHEME
Data security is one aspect of the digital era, and it is recommended that biometrics and block chain be used together as an original and reliable method of protecting sensitive data. As high-level authentication methods, biometric technologies are often utilized in the modern world in place of passwords because they are more difficult for hackers to crack. At this point, the block chain can actually serve as a rapid access method while providing high degrees of security. Any authentication process generates digital signatures using a common hashing method. This paper describes the hashing method's use in the block chain system to process identical digital signatures.
DETECTION OF DEPRESSION USING ACOUSTIC AND TREATMENT RESPONSE PREDICTION USING MACHINE LEARNING
The easiest way to beat the depression blues is to recognise the symptoms and seek therapy right away. Depression diagnosis is based on the patient's openness and level of assistance. There are many diagnostic and assessment tools for depression, including as interview-style testing, objective screening methods, and automatic speech, video, and text identification. The entire procedure takes time, and remote diagnosis is also not an option. The automatic diagnosis from speech, video, and text requires much less time and no highly trained medical professional. If depression is found, the patient can be directed to a skilled and knowledgeable physician. In our study, the website established aids in rapid self-diagnosis of depression with the use of machine learning algorithms. Here we take the audio of the patient to detect whether the person is depressed or not and then we implement in naïve bayes and get the decision as whether that person is depressed or in normal state.
AUTOMATIC FACE MASK DETECTION AND TEMPERATURE SCANNER
Face masks detection and Temperature scanner is an IoT device, that prevents the spread of Corona virus to people. This is done by ensuring that each and everyone are wearing mask and body temperature is less than the ambient temperature. Mask monitoring often requires additional staff resources. Some of the symptoms of covid-19 include fever, tiredness, sore throat, nasal congestion, loss of taste and smell. In most cases, it's transmitted directly (person to person) through respiratory droplets, but also indirectly via surfaces. Therefore, by using face mask and sanitizer, we can prevent the spread of corona virus. However, the crucial problem is that they are lack of approved vaccine and drugs to fight against corona virus. Histogram of oriented gradients (HOG) is a feature descriptor used to detect objects in computer vision and image processing. The HOG descriptor technique counts occurrences of gradient orientation in localized portions of an image - detection window, or region of interest (ROI). This helps in accurate detection of mask. MLX90614 is a temperature sensor, that efficiently detects the body temperature of a person.
AI BASED FARMER’S ASSISTANT
The foundation of the Nation's development is agriculture. The rapid advancement of technology has left our farmers behind. This project gives awareness regarding latest technologies that is necessary for them. The farmer can get access to the agricultural data like best crop for plantation and suitable fertilizers and also get to know about the plant leaf disease. This project will help farmers by providing accurate information about farming. This will also help common users to access the information based on their requirements regarding crops and fertilizers prediction and crop diseases. A website powered by machine learning that, in response to user input like Nitrogen, Phosphorous, Humidity, Rainfall, suggests the best crops to produce and fertilizers also detects the disease caught by their crops. This project uses Random Forest Algorithm for crop suggestion and ResNet for Plant leaf disease detection.
INFORMATION MANAGEMENT SYSTEM FOR MONITORING THE EMOTIONAL STATE OF STUDENTS USING AI
In this current changing era, everything has become digitalised, education is conducted online. Teachers use online platforms to teach their coursework, but it has recently been realised that the classes were not conducted efficiently. The challenge here was to know if the students were understanding the concepts or not. In an offline class, the teacher could get the information just by looking at the students. However, in an online class, it is not possible to look into each student and get collective data. So, in order to cope with these changing demands of the era, we have attempted to build a system that can be helpful in analysing the emotions of students. A framework is built with the convolutional neural network (CNN). The network is trained on the FER2013 dataset and the emotions are analysed and live feedback is given to the teachers by showing the statistics. That represents the proportion of feelings that are angry, disgusted, afraid, joyful, neutral, sad, and surprised. In addition to that, the recording of the complete online class with the label of individual emotions is displayed in the recorded video, and the videos are compressed, so whenever the teachers wish to see the recorded emotions of the class. They need not view the 1-hour video for a 1-hour class, the video is compressed to a few minutes. This saves both time and space.
NETWORK AUTOMATION
In the traditional ways of networking, we have a lot of manual operations to complete any task in networking. This leads to a heavier cost in network management than the cost of the network. To overcome cost and efficiency in network configuration and management, we are automating the tasks using software, where they can control, configure and manage networks. In this project, we are upgrading the Arista device operating system by taking pre and post conformations using python. Networking programming is based on software networks, where scripting is used for network elements control.
SMART DUST BIN
In our project “SMART DUSTBIN” the thought comes for up Smart cities. The main aim of our project Smart Dustbin is to keep clean buildings, Colleges, Healing centers and Transport and society. The Smart Dustbin is a way of an improvement of ordinary dustbin by elevating it by utilizing sensors. Smart Dustbin is a thought of implementation which makes a typical dustbin smart and modern way of utilizing ultrasonic sensors for hand detection and to indicate message to the control room over the status for utilizing GSM modem. The LED light is utilized to show the level of garbage collected. The LED light appears when the bin is empty, the bin is partially filled and the bin is completely filled. Hence this “SMART DUSTBIN” makes a big difference to keep the city clean by indicating the level and sending an alert to the garbage management system.
IOT-BASED SECURITY SYSTEM USING RASPBERRY PI
By providing immediate monitoring of commodities, people, the environment, and property, video surveillance has established its significance and advantages as a crucial part of many organizations' security and safety concerns. The goal of this project is to build an Embedded Real-Time Surveillance System (ERTSS) based on the Raspberry Pi SBC for intruder detection, which improves surveillance technology and provides essential security for our daily life. The suggested security system is built on our ground-breaking integration of cameras and motion detectors into a web application. The Raspberry Pi runs and manages motion sensors and cameras for remote sensing and monitoring, sends live footage, and records it for later replay. The goal of this study is to develop a surveillance system that can recognize strangers and act fast by taking pictures and sending them to a wireless owner-based module. Additionally, this project will show how to detect and track motion using image processing. These kinds of tools are essential for security and surveillance. Live video broadcasts will therefore be used to show how objects can be identified and followed. The detection and tracking process will employ pixel threshold.
PHISHING URL DETECTION WEBSITE
Phishing websites are one of the types of internet security issues that target human vulnerabilities rather than software flaws. It's the technique of luring online users in order to get sensitive information like usernames and passwords. We present an intelligent technique for detecting phishing websites in this research. The technology works as an add on to an internet browser, notifying the user when it finds a phishing website. The system is based on supervised learning, which is a type of machine learning. We chose the Random Forest approach because of its high categorization performance. Our goal is to develop a higher performing classifier by examining the characteristics of phishing websites and combining them in the best way possible.
AI BASED LAKE CLEANING ROBOT
Recently, water pollution has become a global problem. Water resources such as lakes, rivers and lakes are polluted due to the accumulation of domestic, industrial, medical and sewage. The AI Based Lake Cleaning Robot project addresses this issue by providing an efficient and environmentally friendly remote pool cleaning machine. This project has two operating modes which are Manual or Android operating modes and default mode. In manual mode the device is used by the user, the android app is used to control the device to collect garbage. In automatic mode, AI technology that assists the wireless fidelity camera and ultrasonic sensors are used, the camera captures a real-time image of a liquid body and sends it to the Convolutional Neural Network (CNN) for image analysis and CNN will control the boat's performance. The device is equipped with a waste bin. The device is designed to detect any obstacles through the use of ultrasonic sensors to avoid damage. A pool cleaning robot reduces a man's effort by performing an automatic cleaning process. The use of this function can be done in rivers, lakes, lakes and other areas of water to clean up surface water waste.
VOICE ASSISTANT DRUG RECOMMENDATION
Since the coronavirus emerged, the shortage of appropriate clinical resources—such as specialists, care personnel, the right equipment and medications, etc. has reached an all-time high. The medical profession as a whole is in difficulty, which leads to many people passing away. Due to inconvenience, people began taking medications frequently without getting proper consultation, which made their health situation worse than usual. Recently, machine learning has proven useful in many areas, and creative work for automation is on the rise. It aims to provide a drug recommendation system that will significantly reduce the workload of doctors. We are developing a method for prescribing medications that analyzes patient feedback to forecast the emotional response to various vectorization techniques, including Bow, TF-IDF, Word2Vec, and Manual Feature Analysis, which may facilitate suggest the highest drug for a given sickness by totally different classification algorithms. The anticipated sentiments were evaluated by preciseness, recall, flscore, accuracy, and AUC score. The results show that classifier LinearSVC victimization TF-IDE vectorization outperforms all different models with ninety-three accuracies.
FAKE LOGO DETECTION
This project aims to find fake logos by matching and detecting them against the original logo. This is accomplished by dividing the logo picture into rows and columns and assigning an index value to each cell. The values in each cell are referred as pixels which can be compared with logo image to classify whether it’s a fake logo or original logo. To say whether the logo is fake or the original by looking at the relationship between the pixel values of both the original and the logo in training dataset. The original logo is determined by the pixel values of all cells exactly matching the genuine logo otherwise its fake logo. The convolution neural network algorithm is used for detecting and classifying the logo.
3D BRAIN TUMOR DETECTION
The efficacy of self-supervised learning techniques in a variety of application fields has helped them become more popular recently. In this work, we give 3D proxy tasks for five various self-supervised learning approaches using these principles. By enabling neural network feature learning from unlabelled 3D images, we hope to minimise the cost of expert annotation. The techniques include Relative 3D Patch Location, 3D Exemplar Networks, 3D Rotation Prediction, 3D Jigsaw Puzzles, and 3D Contrastive Predictive Coding. Our findings show that pretraining models using our 3D tasks generates more rich semantic representations and enables completing downstream tasks more accurately and quickly compared to training the models from scratch and pretraining them on 2D slices.
WIFI BASED SMART AGRICULTURE
Smart agriculture is an emerging concept, because IOT sensors are capable of providing information about agriculture fields and then act upon based on the user input. The goal of this paper is to create a Smart Agriculture System that takes advantage of cutting-edge technologies like Arduino, IoT, and Wireless Sensor Network. The goal of the article is to use developing technologies, such as the Internet of Things (IoT) and smart agriculture with automation. The ability to monitor environmental variables is a key aspect in increasing the production of efficient crops. The invention of a system that can monitor temperature, humidity, wetness, and even disease detection in tomato leaves is a characteristic of this research. TCP/IP is used in the development of the application for the farmer's Smartphone. The system uses a duplex communication link based on a cellular-Internet interface, allowing data inspection and irrigation scheduling to be configured using an android application. The device has the potential to be beneficial in water-scarce, geographically isolated places due to its energy independence and low cost.
AN ANALYSIS OF WIRELESS NETWORKS
Wireless networks have a unique capability of maintaining same contact even if one moves from one location to another. It provides the flexibility and versatility for wireless communications. The wireless and mobile communications have found usefulness in the area of commerce, education, defense, home-based and industrial applications, and military environment. Wireless networks have had a significant impact on the world through the use of wireless networks; information could be sent overseas or behind the enemy lines easily, efficiently and more reliably. Wireless networking is an essential productivity tool for today's mobile workforce. With wireless networking, we can stay connected to companies and others information resources virtually anytime, anywhere. Wireless networks are becoming more and more ubiquitous in recent years, ranging from mobile analog and digital cellular telephony up to satellite broadcasting. Therefore, supporting a high Quality of Service (QoS) in delivering voice, video and data in this context has emerged as one of the most important challenge of the new century. In this research article we present a survey on introduction to wireless networks, classification of wireless networks, quality of service(QOS) parameters like bandwidth, Delay, Delay Variation (Jitter), Throughput, Energy, and finally to provide the Security to information about different parameters for end-to-end communication over wireless networks.
VEMANA INFORMATION BOT [VIBO]
Generally websites contain large volume of data or information. A person visiting the website for the first few times will find it difficult to search for the required information. To reduce the difficulty and time of a person by providing quick access to the needed information, user friendly chatbots were developed. Similarly, in order to make our college website more interactive and to provide ease of access to the information VIBO is developed which is a text-based chatbot.