Welcome to International Journal of Emerging Technologies in Engineering Research (IJETER)

Volume 6, Issue 10, October (2018)

S.No	Title & Authors	Full Text
1	An Experimental Investigation on Flexible Concrete M. Sasi Rekha, T. Akshaya Abstract - Flexible concrete is composed of all the ingredients of a conventional concrete but the coarse aggregate is replaced and reinforced with polymer fibers. The flexibility of concrete is determined by ductile behavior of concrete. The scope of this study is to improve the ductility of conventional concrete by addition of polypropylene fiber (PP). Properties of polypropylene fiber reinforced concrete (PFRC) and flexible concrete (FC) were studied by casting samples using this fiber. Totally 178 specimens were cast. Preliminary test were carried out on 30 cubes of size 150x150x150 mm and beams of size 500x100x100 mm with two different mix proportions, replacing cement with fly ash and silica fume and fiber contents of 0,0.06,0.12,0.18,0.5,1.0% using Naphthalene based SP, which was found to be ineffective in achieving the workable concrete. From the preliminary test , 1:1.75:2.18:0.40 mix proportions with fiber contents of 0,0.05,0.10,0.15,0.20% and polymer based SP were chosen for casting 30 cubes , 30 disc, 30 beams. The properties of concrete vise ductility, flexural strength, load – deflection behavior, impact strength, stress – strain behavior and compressive strength were studied for all the five series. Also experimental work on FC was carried out on 16 slabs of size 500x110x15mm with fiber contents of 0, 0.5, 1.0, and 1.5% using CM 1:2 to study the flexural strength and load - deflection behavior. Addition of 0.1% PP fiber in PFRC increases the ductility index by 280%, ultimate mid span deflection by 341%, flexural strength by 16%, Modulus of Elasticity by 42%, impact strength by 163%, but decrease the compressive strength by 38% at 0.2%.Addition of 1.0% PP fiber in FC increases the flexural strength by 7.2%. Post crack deflection is 14 times at 1.5% but the workability decreases beyond 1.0%. Therefore addition of 1% PP fiber in CM 1:2 produced the flexible concrete.	Download
2	Hybrid Encipher to Protect Data on Shared Preferences Anukul Dixit, Krishanu Das, Ronak Harish Patil, Brinda.S Abstract - In this present world different Android applications develop data on internal memory with Shared Preference interface. Due to this many loop holes can access the important information like passwords. Therefore, we propose an encryption system that would secure the confidential data through source code. We create an application and supply some data using Shared Preference. Then we apply the Encryption approach with Android Keystore system and an encryption method to hide important data.	Download
3	Smart Mirror Using Raspberry Pi Model 3 B Anukul Dixit, Krishanu Das, Ronak Harish Patil, J. Sivapriya Abstract - With the never-ending invention of the new technologies there is no limit to making the devices smarter than they exist in today’s world. The reflecting surfaces like mirror provides the user an interactive environment and more personalized feeling along with the display of the information. This paper shows us how to design and develop an interactive smart mirror using Raspberry Pi 3 Model B+. Features like facial recognition helps in the smart unlocking process and features like voice recognition and touch sensitivity makes the user experience easy, better and more interactive. The mirror provides various basic amenities like weather, news, maps, daily schedule and much more.	Download
4	Efficient Distribution of Water using Machine Learning Techniques Aruna.A, Mayukh Dasgupta, Ankur Ankit, Eenesh Chavan Abstract - Every country despite of having proper water distribution database is not utilizing it for the common good. Various water distribution corporations possess valuable data that if used in the proper methodologies can not only reduce human labor and increase efficiency but also be cost effective. Artificial Neural Networks (ANNs) have been brought into the scenario to accomplish the above maintained task. It has been found that there's plenty of scope of applying Machine Learning techniques in efficient water distribution. Delay in implementation of such methodologies are due to lack of in-depth knowledge about the techniques and lack of visualization. Continuous research has been going on in this field where different kinds of ANNs has been used for prediction of water flow in a WDS (Water Distribution System). Such readings can be further used to build an efficient WDS that will efficiently distribute water in every household. Time stamp prediction models are generally preferred for such approach as comparison of previous and current scenario is necessary for better future prediction.	Download
5	Secured E-Banking System using Artificial Intelligence R. Augustian Isaac, Prakhar Chaturvedi, Pradip Gareja, Rohan Grover Abstract - The issue of design and security is very predominant in any financial and business organization, especially such organization as a bank. Therefore, we intend to aid in security of the bank by bringing in an Artificial intelligence system that involves an individual to get an access to some items using face and voice recognition security system. This AI system is not just a normal password lock system that require a user to insert password and gain access to some items, it is a system that has an administrative authentication. In addition, with this kind of security authentication system we intend to implement, a highly secured AI feature, which enables the user with assured and highly secured transactions using their personal frame. Here an individual have to provide the face and voice authentication, which uses different algorithms, and is read by the AI server for clarification and verification. From this project, we hope to build an alternative and highly verified security for banks.	Download
6	Digital Health System in vehicles with Artificial Intelligence Configuration J. Shiva Nandhini, Ranjan Kumar Jaiswal, Prakhar Chaturvedi, Rohan Grover Abstract - After long advancement in human health detection, the need of being alerted about health while seating in a vehicle in real time is paramount to everyone. Cardiac diagnosis in two wheelers and cars for riders and drivers using carotid nerve with heart rate sensing device is aimed at acquiring and analysing the data. The advent of Internet of Things have transcend the communication in real time. In this paper, we conduct feasibility of communication of heart diagnosis in real time with Internet of Things paradigm and detection scenario of the real time images using tenser flow of machine learning and data privacy aspects.	Download
7	Biometric Security System Using Arduino for Vehicles Rohan Chandra Pradhan, Saumya Rai, Smit M Shah, Brinda Abstract - Objectives: This paper advance a noble system to prevent vehicle from thefts. This system provides a fingerprint security to unlock the vehicle. If an unauthorized person tries to steal the vehicle, user will be notified in its registered mobile number. Along with that if the vehicle moves from its position, an alert message will be sends to the authorized user. Methods: Fingerprints of the authorized user is first stored in the memory of the system. If the fingerprint matches with the stored ones then the controller triggers the powering circuit and the engine starts else if the burglar tries to start the car it will sends the alert message to the owner and if the vehicle moves from its location it will send an alert message along with the GPS location of that vehicle. Findings: The system is implemented using Arduino Uno. The GPS, GSM, Wi-fi module and a SD card is additionally associated with the board. Along with that the system incorporates certain other sensors to stop the car burglary. Application: The system works well with low-price range vehicle employed with keyless entry and is unique because it uses IOT to protect the vehicles.	Download
8	Gas Leakage Detection Using IOT Tools T.H. Feiroz Khan, Disha Dikshita Behera, Riya Sidha, Anisah Manouwar Abstract - The security of every single family unit is especially required these days, therefore numerous controls are utilized to stop gas blast dangers. The framework proposed utilizes IOT gadgets to recognize spillage of gas in and aides in the location of human nearness close to the gas spillage source. On the off chance that the distinguishing proof comes up short, then the gadget puts a self-clock of 10 minutes and communicates something specific alarm making the alert ring. Following 10 minutes if there is no human recognition then the primary gas pipeline will itself kill. The essential rule behind the caution discovery is the adjustment in the centralization of air.	Download
9	Flight Delay Prediction using Binary Classification Roshni Musaddi, Anny Jaiswal, Pooja J, Mansvi Girdonia, Minu M.S Abstract - Nowadays, aircrafts have become a necessity because they easy life. They are efficient in carrying goods and passengers around the world. It also supplies emergencies in warfare and takes a vital role in carrying medical necessities. Hence, advent of airplanes is considered important. Delays in aircrafts can affect thousands of people across the globe either directly or indirectly. There are a lot of reasons of delays in aircrafts such as critical weather, security issues, traffic and many more. There are several methods proposed to predict the flight delays but due to various complexities of the ATFM and the huge datasets involved, it has become very difficult to find an accurate solution for this complication. Many algorithms have been implemented to forecast flight delays. We are using Python in Visual Studio Code. We implement Binary Classification to prepare a model that can predict the delays.	Download
10	Route Planning for Ships during Emergency Situations using Genetic Algorithm MS.Minu, Ram Abhilash V, Bhaskar Dey, S.Venkatesh, Vignesh Kannan Abstract - During sea navigation, plotting an efficient route is highly essential. Simple target specific algorithms like Dijkstra Algorithm are generally used for this purpose. But these algorithm are not helpful during emergency situations when solutions are not optimal enough due to technical limitations.Thus, the authors propose the use of Genetic Algorithm(GA) in critical conditions to reach destination optimally or as closely as possible. GAs were proposed many times for sea navigation but are replaced with target specific algorithms due to delayed time in path-finding. This paper discusses the need of GA in emergency situations such as low fuel or ship damage (time constraint) where it can be used to find all possible paths by modifying the fitness function. The obstacles are plotted using ARPA system installed in the ship. The collective data is used to run simulations to find the optimal route. Once the optimal route is found, the ship can follow the course and will not exceed the constraints.	Download
11	Digital Health System with Artificial Intelligence Configuration R. Augustian Issac, Ranjan Kumar Jaiswal Abstract - After long evolution in human health detection, the need of being alerted about health in real time is paramount to everyone. Cardiac diagnosis in two ways can be done using carotid nerve and wrist nerve with heart rate sensing device is aimed at acquiring and analysing the data. The development of Internet of Things have overstep the way of communication in real time. In this paper, we conduct feasibility of communication of heart diagnosis in real time with Internet of Things paradigm and detection scenario of the real time images using tenser flow of machine learning and data privacy aspects.	Download
12	Image Compression over MNIST Dataset using Autoencoder R.J.Poovaraghan, Ankush Rai, Ronit Dharmik, Siddharth Bhattacharjee Abstract - Image compression has been the area of investigation for many years. A huge amount of images are being stored by Multinational companies and different organisations. For the last few year, deep learning has been very successful in computer vision tasks, and now its being used in image compression. Deep Autoencoder neural network trains on a large set of images to find similarities between the images in the set. Then those similarities are used to compress and present them using simpler and less codes then current image compression techniques. In this paper, we demonstrate how Deep Autoencoder neural network is used to train data and large sets of images are compressed. MNIST dataset is used for training and testing purpose. A 28 x 28 image is first converted to a vector of 784 x 1, which is then given to the network. Then we take this vector and compress it in each iterations of training phase of the network. After the vector is compressed, a compressed feature vector is hold by the middle hidden layer of the network to be stored or transmitted, Later, that vector is decoded back to obtain the original image.	Download
13	Weather Forecasting System using Machine Learning R.B.Saroo Raj, Ankush Rai, Ronit Dharmik, Siddharth Bhattacharjee Abstract - Weather prediction is an ongoing testing issue seen by the world in the most recent decade. The prediction is hard because of the adjustment in weather. Different weather prediction models have been outlined utilizing machine learning and Artificial Intelligence over the recent years. The genuine issue of precisely anticipating or forecasting the weather still remains. Numerical weather prediction is taking the current numerical information on weather conditions and applying machine learning algorithms on it to forecast the weather. This paper is the use of machine learning algorithms, direct relapse demonstrate from insights, and two enhancement techniques, Normal condition strategy, and Gradient plunge technique to anticipate the weather based on a couple of parameters. Two enhancement techniques have been utilized to look at the execution of the algorithms. The got results exhibited that the typical condition technique forecasts the weather with high exactness, though the inclination plummets strategy forecasts the weather with next to no accuracy.	Download
14	Implementation Procedure on Efficient Distribution of Water using Machine Learning Techniques Feiroz Khan, Mayukh Dasgupta, Ankur Ankit, Alankrit Yadav Abstract - This paper mainly deals with the functionality prospective of implementing machine learning to water distribution in an urban or rural locality. To this date all the work has been done through human labors. This process although seems to be full proof but has quite a number of loopholes. Firstly, the above process is time consuming as humans can work only at a certain rate. Secondly, human error cannot be left out of the equation as well. When taken into considerations, all these problems cater largely to inadequate water distributions and loss of water in huge quantities. All these problem scenarios can be checked with the implementation of a simple Machine Learning algorithm into the system. Although much of the work still has to be taken care by human methods itself. This also renounces the possibility of job opportunities becoming less due to automation in the system.	Download
15	Obstacle Avoiding Robot Using Infrared and Ultrasonic Sensor Ajay Krishna Gamidi, Saran Reddy Ch, Aravind Yalla, R.Augustian Issac Abstract - Now a day’s many industries are using robots due to their high level of performance and reliability and which is a great help for human beings. Proposed project robot will move in a particular direction in which Ultrasonic Sensors and Infrared sensor sense a signal while avoiding the obstacles in its path. Ultrasonic Sensor and Infrared Sensors are controlled by an Arduino Microcontroller. Ultrasonic and Infrared are less in cost and has high ranging capabilities. They can Navigate Robot in Unknown Environment By avoiding collisions and it is also used for blind people. It is used to stop accidents and reach our homes safely. We have programmed the Arduino Microcontroller so that it detects the obstacle and avoid the collision.	Download
16	A Model to Predict Loan Defaulters using Machine Learning R. B. Saroo Raj, Gurpartap Singh, Balaji S, K. H. Ajit Baskar Abstract - With the upgrade in the managing an account part bunches of individuals are applying for bank loans however the bank has its restricted resources which it needs to give to constrained individuals just, so discovering to whom the loan can be allowed which will be a more secure alternative for the bank is a regular procedure. So, in this paper we endeavour to lessen this hazard factor behind choosing the protected individual to spare heaps of bank endeavours and resources. This is finished by mining the Data of the past records of the general population to whom the loan was allowed previously and based on these records/encounters the machine was prepared utilizing the machine learning model which give the most precise outcome. The primary target of this paper is to anticipate in the case of allotting the loan to specific individual will be protected or not. This paper is separated into four segments (i)Data Collection (ii) Comparison of machine learning models on gathered data (iii) Training of framework on most encouraging model (iv) Testing.	Download
17	Distributed Database for Water Supply Data Based on Hbase Adlene Ebenezer, Gurpartap Singh, Adarsh Malav, Kishan Agarawalla Abstract - Managing massive water supply data is a typical big data application because water supply systems generate millions or billions of records every single day. To guarantee the safety and sustainability of water supply systems, massive water supply data need to be processed and analysed quickly to make real-time decisions. Traditional solutions typically use relational databases to manage water supply data. However, relational databases cannot efficiently process and analyse massive water supply data when the data size increases significantly. In this paper, we show how water supply data can be managed by using HBase, a distributed database maintained by Apache. Our system consists of clients, zookeeper, HBase database, Hadoop Distributed File System (HDFS). We show’s how HBase’s parameters can be tuned to improve the efficiency of our system.	Download
18	RFID Authentication System Integrated with the Enhancement of the Security of OTP using Cryptography J. Sivapriya, P. Vamsi Mohan, V. Surya Teja, N.V. Harsha Vardhan Gupta Abstract - The technology of RFID has been evolving in various aspects of implementation in every industry. We believe that RFID microchip is been erupting as a new technology which can be implanted in the body of a human being for entry into more secured locations and to protect the data which is most important. The main approach for the RFID environment is for the safety purpose of the data which can be achieved in a highly secured manner .In this paper, we purposed a system for E-commerce websites to avoid fraud and ensure safe bank transaction to the customer, we provided RFID login authentication which is connected to database and the OTP is encrypted for the safety of the customer’s bank transaction using Cryptography.	Download
19	Survelliance and Controlling of Smart Agriculture System using IOT V. Satya, N. Vamsi Kishan, V. Surya Teja, P. Vamsi Mohan, N.V. Harsha Vardhan Gupta Abstract - Now-a-days agriculture is being automated in many ways as per the requirement. As there is need of automation in farms due to decrease in the strength of the farmers working in day by day life. The main reason for the agriculture farms is automated is because of the unseasonal rains, soil erosion, attack of the animals. The automation of agriculture can be done using IOT (Internet of things) with the help of sensors and RFID (Radio Frequency Identification) tags and readers as IOT is the most evolving technology in recent days. In this proposed system, helps farmers by providing various sensors like motion, image, gas (particularly methane), soil moisture sensors that sense the various parameters of the field. In addition, based on soil moisture values sensed, the motor will be ON/OFF as per the requirement and if any sensors captures any anomaly then an SMS is sent to the farmers mobile and an appropriate action is taken.	Download
20	Multi Pixel Visual Cryptography for Colour Images in Meaningful Shares Baby D Dayana, Jitesh Akaveeti, Kotha Karthik, M Sushmitha Abstract - Visual Cryptography is an encryption method to hide secret information in images so that when decrypted by using a correct key image secret information gets revealed. Visual Cryptography uses two transparent images. One of the images has random pixels while the other has secret information stored in it. Retrieval of the secret information from one specific image is impossible. Both transparent images/layers are required to reveal the information. Using the properties of the human visual system to force the recognition of a secret message from overlapping shares, the secret image is decrypted without additional computations and any knowledge of cryptography. Encryption is done using simple algorithm which generates n copies of shares depending on the type of access structure schemes.	Download
21	Enhanced Smart Parking System using Sensor Technology J Sivapriya, Jitesh Akaveeti, Karthik Kotha, M Sushmitha Abstract - The idea of Smart cities has gained a great popularity around the world nowadays. Parking spaces these days are used by of people at every commercial complex like shopping malls, cinema halls, etc., and finding an empty spot there is a very difficult task, people find it burdensome most of the times. Many face these problems in finding an appropriate slot. So when this comes to large commercial complexes we see people even wasting their valuable time every day. Sometimes it takes hours to find an appropriate parking spot. On removing all these constraints and making everything automated by implementing a computerized system which simplifies the existing problem. Thanks to the advancement of the Internet of Things. To do all this we need to introduce a system which includes proximity sensors, OCR reader, Fare collecting machine which collects the amount to be paid by the user, QR code generator that prints on the ticket that is to a relative vehicle and parking slot allotted to it in the complex and is given to the user. The parking space is allotted depending on the requirements of the mall, concern of the user as well as the availability of the parking space in the complex. As soon as the user goes to the assigned parking space, it is updated in the admin database that it is occupied. Before leaving, QR code is scanned by a QR code reader including the fare collecting machine at the exit gate by the assigned user where he has to pay for the usage of parking lot for that time period and also it makes sure that he exited the slot and now that slot can be assigned to the next user. This whole information is stored and monitored by the database and is displayed to the administrator on a screen.	Download
22	The Radar System Using Arduino Thanasekaran, Sibi Sankar, Dina Dayanithi, Sellam Abstract - RADAR is an object detection system in which it uses radio waves to determine RADAR is an object detection system which uses radio waves to determine the range, altitude, direction, or speed of objects. RADAR systems come in different variety of sizes and have different performance specifications. Some RADAR systems are used at air-traffic control at airports and others are used for long range surveillance and also early-warning systems. A RADAR system is known as the heart of a missile guidance system. Small portable RADAR systems which can be maintained and operated by one person are available as well as system that occupy several large rooms . The investment required in developing RADAR is for less critical purposes like surveillance in close proximity, auto parking systems in vehicles, and object detection in small ranges it would unreasonable to spend capital in large amounts. So, this is an attempt to use the Ultrasonic Sensor (HC SR04) to implement the similar working concept as used in RADAR to detect any nearby objects. Arduino UNO board is sufficed to control theultrasonic sensor and to interface the sensor and also as a display device.	Download
23	Educational Website Based on Dynamic Recommendation and Filtering Techniques M. S. Bennet Praba, Prangad Khanna, Ashutosh Saxena, Sachit Jain, Raunak Pal Abstract - The main objective behind creating an educational website is to provide accurate notes and study material to students who wish to learn more about the subject in a precise manner. There is a need to efficiently deliver an especial, pertinent data to attenuate the issue of information overloading, which has created barrier for various education-based website users. To solve the growth of this problem our website follows dynamic recommendation system, content based filtering and collaborative filtering techniques which ensures that the best of the content is shown to the user. To make learning even more fun, we are giving an opportunity to our users for earning money by uploading notes. This project will serve its impetus to the ones who are going to download material from our website too. The mechanism follows simple methods and algorithms. A scratch card will be rewarded on each review the downloader gives to the notes uploaded on the website. Our vision is to be one of the leading sites providing education and a simple channel to earn money.	Download
24	Automatic Control on Railway Gates Using IoT Baby Dayana, A.K.R.N.Supreeth, K. Somesh, A.Hruday Kumar, N. Harinath Abstract - Majority of the people like to travel in trains because it is affordable comfortable and fastest mode of transportation. In spite of large usage of railways, It is still not developed in technology in certain areas like level crossings. level crossing gates are still manually operated and many unmanned railway gates are present. Railway gates are mainly used to save the road users to prevent accidents from occurring. In every level crossing there are so many of the railway accidents are occurring due to the carelessness and lack of experienced workers. The main objective of this project is to control and operate the railway gate automatically and prevent accidents to save precious human lives. It helps to prevent accidents from occurring and reduce the manual labour. Here, we use Arduino and IR sensor.IR sensor use the IR transmitter to pass the arrival of train information, IR receiver receives the information and pass it to the micro controller. Arduino use the DC motor to close or open the gate.	Download
25	Wireless Body Sensor Network Rohan Chandra Pradhan, Saumya Rai, Sirigiri Keerthi, Gowthamy. J Abstract - Wellbeing is one of the fundamental parts of our life. In order to make the human life free from all the crucial diseases there are numerous frameworks which help to defeat with the medical issues. This paper presents a kit which consists of a GSM module, GPS module and three different types of sensors i.e. Heartbeat, Pressure and Temperature Sensor and all of this sensor is programmed using an Arduino. This system is implemented using Internet of Things (IOT). The high and low limits can be set by the user for its blood pressure, the pulse rate as well as its body temperature and if the user’s body crosses any of the limits an alert message will be send to the user register device along with a medical health guideline and if the body symptoms cross the critical range then an alert message will be sent to the nearby hospital along with its GPS location.	Download
26	Preliminary Tests and Progress Report of Multiple Myeloma Using Computer Vision and NLP Sellam.V, Nimisha Asthana, Ritik Raj, Yash Singh Abstract - Multiple myeloma refers to bone marrow cancer which is the most common amongst all. For the detection of it various methods are being used which include MRI, biopsy, CT-scan and bone marrow examination.The methods are invasives.This work proposes a system in which the eradication of existing systems will be performed. In, this approach a chip will be designed using trabecular titanium (TT) to identify and report the progress or declination of the cancer.The proposed method will be evaluated on matlab using real clinical images.	Download