MIT-World Peace University (MIT-WPU) is a top educational institute located in Pune, India. It is recognized within&nbsp;UGC under Government of Maharashtra Act No. XXXV 2017.&nbsp;

It is part of the MIT group of institutions (established in 1983) which covers 10+ campuses across India and has more than 50,000+ students enrolled across various disciplines at a given point of time. The institute houses 15 schools ranging from Engineering &amp; Technology, Design, Pharmacy, Commerce, Governance among others. It offers undergraduate, postgraduate and PhD programs among different disciplines.&nbsp;

&nbsp;

MIT World Peace University

The automobile and transport industries have always tried to reduce downtime by the way of preventive maintenance. In the past decade, electromechanical sensors have become more accurate along with novel innovations in IoT and machine learning, and automobiles have leveraged this. In this paper, an end-to-end open-source predictive maintenance solution is presented to predict the severity of faults in a car using onboard historical and real-time sensor data using IoT and machine learning. Sensor data is collected from a Suzuki Swift VXi Model, and classifiers like logistic regression, random forest, and gradient boosting trees are used to train the data with imputed faults. F1 score and AUC are used as evaluation metrics. An end-to-end onboard diagnostics (OBD) data to the user dashboard pipeline is proposed with final predicted faults visible on a real-time dashboard. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021.

World University root node

School of Electronics & Communication Engineering

Engineering and Technology

Lecture Notes on Data Engineering and Communications Technologies

Maintenance of Automobiles by Predicting System Fault Severity Using Machine Learning

The automobile and transport industries have always tried to reduce down-time by the way of preventive maintenance. In the past decade, electromechanical sensors have become more accurate along with novel innovations in IoT and Machine Learning, and automobiles have leveraged this. In this paper, an end to end open source predictive maintenance solution is presented to predict the severity of faults in a car using onboard historical and real-time sensor data using IoT and Machine Learning. Sensor data is collected from a Suzuki Swift VXi Model, and classifiers like Logistic Regression, Random Forest , and Gradient Boosting trees are used to train the data with imputed faults. F1 score and AUC are used as evaluation metrics. An end-to-end On-Board Diagnostics (OBD) data to the user dashboard pipeline is proposed with final predicted faults visible on a real-time dashboard.

Maintenance of Automobiles by predicting System Fault Severity using Machine Learning

The purpose of this paper is toreview Motion Amplification Techniques in video sequences. Various computational techniques are used to visualize, analyze and efficiently represent imperceptible variation in image and video sequences. These computational methods which are used reveal temporal changes in videos which are not visible to the naked eyes. In these techniques standard video sequence is used as an input on which spatial decomposition is applied then temporal filtering is applied. The filtered output is then amplified to expose the invisible content in the input video. This paper contains four techniques which are used for enhancement of temporal variations in the video: 1. Linear Approximation Method,Eulerian Video Magnification (EVM), 2. Phase Based Video Processing 3. Fast Phase Based Video Processing (Riesz Pyramid) and 4. Enhanced Eulerian Video Magnification (EEVM). Using these techniques, it is possible to magnify and see tiny motions and subtle color variations.After review it is observed that, the noise level in EVM technique is more which reduced in EEVM technique using motion analysis and image warping. The other two phase based techniques supports larger amplification factor and better noise performance.

International Journal of Future Generation Communication and Networking

Comparative Study of Motion Amplification Techniques for Video Sequences

The purpose of this paper is to review Motion Amplification Techniques in video sequences. Various computational techniques are used to visualize, analyze and efficiently represent imperceptible variation in image and video sequences. These computational methods which are used reveal temporal changes in videos which are not visible to the naked eyes. In these techniques standard video sequence is used as an input on which spatial decomposition is applied then temporal filtering is applied. The filtered output is then amplified to expose the invisible content in the input video. This paper contains four techniques which are used for enhancement of temporal variations in the video: 1. Linear Approximation Method, Eulerian Video Magnification (EVM), 2. Phase Based Video Processing 3. Fast Phase Based Video Processing (Riesz Pyramid) and 4. Enhanced Eulerian Video Magnification (EEVM). Using these techniques, it is possible to magnify and see tiny motions and subtle color variations. After review it is observed that, the noise level in EVM technique is more which reduced in EEVM technique using motion analysis and image warping. The other two phase based techniques supports larger amplification factor and better noise performance.

IJFGCN

<a href="http://sersc.org/journals/index.php/IJFGCN/article/view/29174">Comparative Study of Motion Amplification Techniques for Video Sequences</a>

Remote sensing image scene classification has a wide scope of applications and hence has been receiving remarkable attention. Nowadays, Deep learning (DL) algorithms have seen a massive rise in popularity for remote-sensing image analysis over the past few years. This paper first provides a comprehensive review of the recent progress in remote sensing applications. The existing remote-sensing classiﬁcation methods are categorized into four main categories according to the features they use: manually feature-based methods, unsupervised feature learning methods, supervised feature learning methods, and object based methods[15]. This article then focuses on evaluating the available and public remote-sensing datasets. Finally, a conclusion regarding the current state of art methods and directions for future research are presented.

Review on Deep Learning in Remote Sensing Image Classification

Remote sensing image scene classification has a wide scope of applications and hence has been receiving remarkable attention. Nowadays, Deep learning (DL) algorithms have seen a massive rise in popularity for remote-sensing image analysis over the past few years. This paper first provides a comprehensive review of the recent progress in remote sensing applications. The existing remote-sensing classification methods are categorized into four main categories according to the features they use: manually feature based methods, unsupervised feature learning methods, supervised feature learning methods, and object based methods[15]. This article then focuses on evaluating the available and public remote-sensing datasets. Finally, a conclusion regarding the current state of art methods and directions for future research are presented.

Review on Deep Learning in Remote Sensing Image Classification

This project includes the designing of 16-Bit RISC processor and modeling of its components using Verilog HDL. The processor is based on Harvard architecture. The instruction set adopted here is extremely simple that gives an insight into the kind of hardware which should be able to execute the set of instructions properly. Along with sequential and combinational building blocks of a processor such as adders and registers more complex blocks such as ALU and memories have been designed and simulated. The modeling of ALU which has been done in this project is fully structural starting from half adders. At the end the semi-custom layout has been developed for ALU. Complex blocks such as memories have been modeled using behavioral approach, whereas simple blocks such as adders have been done through structural approach. The tools which have been used throughout the project work are ModelSim and Intel Quartus prime 18.1. For synthesis purpose the targeted FPGA device technology is ALTERA.

International Journal of Creative Research Thoughts

Design and Implementation of 16 bit RISC Processor on FPGA

Automobiles are perhaps the domain that represents the 21st century. The vehicles play a major role in our day to day life for transportation. The term Cruise Control refers to the concept of assisting drivers in the task of longitudinal vehicle control to avoid any accident or collision. With the increase in the world population the need for automobiles as well as its usage daily has increased drastically. This leads to heavy traffic, rush, collisions, and accidents. The automobile black box is used to analyze the cause of vehicular accidents and prevent a loss of life and property from vehicle accidents. This Paper is based on the Project topic of Adaptive Cruise Control. In our proposed system, we tried to implement ACC using the less expensive devices, also increasing the ease to use it and adding a few other features in it. The project aims to build a Semi-Autonomous system; that assists the driver for the longitudinal control on their vehicle while driving by using the ultrasonic sensor (HCSR_04). An alert (led or buzzer) indication is given when the inter-vehicle gap approaches the safety limit. A switch is provided to ensure the semi-autonomous use of the vehicle. Also, it aims to use IoT to keep track of the distance readings, speed readings, and stability of the vehicle using the NodeMCU and ThnigSpeak platform. This system also uses the GPS (Neo-6m) and GSM (SIM 800A) to send a text message to a pre-defined number in case of detection of an accident.

Adaptive Cruise Control System

The increase in population leads to increment in pollution and accident as well. Technological development makes great efforts to detect the accident and monitor the increasing pollution. This paper is based on enhancement in the smart vehicle system. Different units implemented in this paper which enhance the vehicular system. The main objective is to detect accidents in real time and minimize the response time of medical help. For accident avoidance, tire pressure is measured whereas in accident detection is implemented with the help of node MCU. MQ7 is used in order to monitor the pollution. The proposed system is useful in reducing the vehicular accidents and pollution monitoring will help to know the environmental status. © 2018 IEEE.

Proceedings - 2018 IEEE Global Conference on Wireless Computing and Networking, GCWCN 2018

Vehicle Collision Detection and Avoidance with Pollution Monitoring System Using IoT

This paper shows that E-Toll booking can be completely managed using the 'Internet of Things' concept based on the RFID and QR/Barcode scanning technology using Raspberry Pi. In reference to survey of annual toll collection on toll plazas, conducted by government of Maharashtra in year 2010, we examine and report on the conditions of total time spent, for money transaction on toll plaza, by a particular vehicle and further evaluate total fuel wastage as well as human errors involvement, while doing so. Finally to provide an optimal solution, we consent with the idea of making toll plazas completely automated, with assistance of IoT technology. By using this system no separate lane required at Toll Plaza. Keywords IoT: Internet of Things, RFID: Radio Frequency Identification, Webcam, Stepper Motor, RFID Reader: EM18, Raspberry Pi: 3B Model, WordPress Software OpenCV

International Journal of Research Publications

IOT BASED E-TOLL BOOKING SYSTEM

Water is an important natural resource in our life. Quality of water needs to bemonitor in real time in order to ensure safe and clean supply of drinking water.For that purpose IOT (Internet of Things) based water monitoring systemssuch as tank water level sensing monitoring and water pollution monitoringhas been proposed. This paper describes design of IOT based water qualitymonitoring system, with the objective of notifying the user of the real- timewater quality parameters. The system measures physical and chemicalparameters of water quality such as pH, level, turbidity, temperature andhumidity. This parameter detects water contaminants. The measured valuesfrom the sensors are processed by ESP8266 Wemos d1 mini and theseprocessed values are transmitted remotely to the raspberry pi via Wi-Fi andmake the raspberry Pi display the data on a simple graphical interface overdifferent time periods.

International Journal of Electronics Engineering Research

Water quality monitoring system for water tanks of housing society

Smart helmet safety system is the idea that has been developed for the social responsibil-ity towards the society. Recently the number of bike riders has increased so there is over-all increase in the number of motor bike acci-dents. The system assists in reducing major number of accidents. It consists of microcon-troller along with IR and PIR sensors for hel-met authentication followed by alcohol detec-tion with the help of MQ3 sensor. As soon as the bike rider try to start the bike it shall rst authenticate if the person is wearing the hel-met through IR and PIR sensors. Secondly, MQ3 sensor detects the presence of alcohol, if detected the bike wont start. It also detects fall detection through accelerometer. The project focuses on bike rider safety and reduction of accidents which are fatal.

International Journal of Research in Engineering and Technology (IJRET)

Smart helmet safety system using Atmega 32

Sign language is widely used by individuals with hearing impairment to communicate with each other conveniently using hand gestures. However, non-sign-language speakers find it very difficult to communicate with those with speech or hearing impairment since it interpreters are not readily available at all times. Many countries have their own sign language, such as American Sign Language (ASL) which is mainly used in the United States and the English-speaking part of Canada. The proposed system helps non-sign-language speakers in recognising gestures used in American Sign Language. The system described in this paper is implemented using MATLAB. In this approach, firstly, the signs are captured using a webcam. The images captured are then processed further and the features are extracted from the captured images using PCA. Comparison of the features is done using Euclidean Distance with the training sets. Minimum Euclidean distance helps to recognise the character. This system will enable non-sign-language speakers to better understand and communicate with those with impaired hearing.

Proceedings of the International Conference on Inventive Computation Technologies, ICICT 2016

Online hand gesture recognition &amp; classification for deaf &amp; dumb

Object counting is a popular operation in computer vision. It is also useful for visual inspection. When counting objects for visual inspection count of objects with similar features is calculated. In object counting feature extraction from input image is done and count is calculated by comparing these features values with cutoff values. These cutoff values are statically specified and used for particular application. In this paper an automatic counting method is proposed which requires only tolerance to be specified to calculate count of object. The proposed method calculates cutoff values at runtime and therefore it can be easily applied to different applications. The object count generated by proposed method is useful for visual inspection of input samples. The proposed method has basic four steps as Input image, Preprocessing, Segmentation and Counting. In fourth step 'Counting' cutoff values are calculated to generate final object count. Proposed automatic counting method is tested on two applications and it shows high accuracy. © 2015 IEEE.