2017-18

Permanent URI for this collection

http://172.16.1.132:4000/handle/123456789/9793

Browse

Now showing 1 - 20 of 21

Identification of Landing Sign of UAV Using Faster R-CNN Algorithm
(2018-09-01T10:36:04Z) Aman Kumar, Dubey; Anjali, Muralidharan; Dhanalakshmi, B
In order to realize autonomous landing of the unmanned aerial vehicle (UAV) in power patrolling, a visual method vision based on Faster Regions with Convolutional Neural Network (Faster R-CNN) for UAVs is studied. In this project, we design the landing sign of the Unmanned Aerial Vehicle(UAV’s) and propose the Faster R-CNN recognition algorithm which can be used to identify the target sign. Faster R-CNN successfully identifying the landing mark is the most important step for the UAV autonomous landing. The experimental results show that it can achieve recognition speed of nearly 81 millisecond each frame and 97.8% accuracy by using Faster R-CNN for detection and identification. The proposed method has better identification accuracy compared with three target identification methods, the Support Vector Machine (SVM) classification, the Back Propagation (BP) neural network and You Only Look Once (YOLO) based on deep learning.
3 Strike Dashboard
(2018-09-01T10:40:46Z) Rosy, Guha
Healthcare professionals must use a computerized automated tool to keep a record of the number of SRs which have not received any updates within 7 days. Currently Cerner follows a manual process to achieve 3 Strike Rule which becomes cumbersome when the queue size crosses 25+. Associates might have to keep an eye on every SR to follow 3 strike rule which would then qualify for SR Closure on 8th day. The development of dashboard will help associate to keep automatic count on the number of SRs which has not received any response in 7 working days. Also, associates need not have to manually go and check every SR for next follow-up as dashboard itself will keep count and display a grid(notification) for every SR. Even if the associate doesn’t open the tool, this dashboard will provide an extra leverage by sending the same notification to his/her mail.This will always keep the associate aware on what would be the action item that is required from his/her end without any manual calculation and ensure that the risk of inconsistencies and inaccuracy is minimized. The aim of this project is to create a dashboard to notify associates of Cerner Healthcare India Pvt. Ltd. about the SR (Service Requests) which has not received any response from client for continuous 7 working days. Moreover, it will also display an immediate report with respect to the pending updates for each SR. The achievement of this goal will let the engineers save a huge amount of time instead of doing it manually.
Implementation of Portal Data Into Representable Graphs To Track Associate Performance
(2018-09-01T10:46:35Z) Ashika, A
At Cerner, we receive Service Request (SRs) from the Clients through the Remedy tool where our associates work together and solve issues client face. We not only work on SRs that the client sends but also, we maintain a proactive check on the client domain. This is done by the Portal data. Here the upcoming issues can be observed and proactively cleaned. We receive them in the form of mails from various clients in a particular field. The tool provides analytical information as well as let’s associate be mapped to the client or the reason according the team lead’s requirements. Here we also let the associate understand how much clean up must be done on a that day based on the assignment done. For the managers, to understand all the work done on the portal data, this tool has been developed to understand the trends in the portal data, if any new issue has aroused and the amount of clean up done on the Portal data. It can be very useful to predict the proactive work that an associate has performed. The manager can check how much of Portal data has been cleaned and managed by each team under him by applying filters.
Automation Script Development For RLN
(2018-09-01T10:53:47Z) Yesha, N B
Healthcare domain is one of the important areas that needs integration of Information technology. One of the key problems faced by the patients is that the tests to be carried out as a part of treatment will not be readily available due to lack of information about the laboratories that carry out the tests. As a result, it is overhead on patients to get the required tests done from the reference laboratories to proceed with further treatment. There is also a high barrier for smaller hospitals and laboratories for entry. Due to lack of knowledge about the laboratories that provide the tests and absence of communication between the hospitals and the laboratories there is high risk involved if the tests are not done at the right time. Reference Lab Network (RLN) creates a single point of connectivity for the reference labs. RLN supports reference labs by managing their entire network. The main goal of this application is to ensure that “Right tests go to the right lab using fewer connections”. RLN eliminates the intervention of the patient while carrying out the tests. RLN improves the overall efficiency and reduces the number of interfaces. Developing efficient application ensures leading edge of care by connecting the ordering labs (hospitals) and the reference labs (performing labs) and help facilitate sharing of patients’ tests and returning timely results. The aim of this project is to automate the validation process of web application called RLN (Reference Lab Network) that is used for maintaining the reference lab orders of clients at Cerner Healthcare India Pvt Ltd. The project focuses on writing automation scripts for testing the application’s UI to ensure that it meets its required business specification document. Automation of plans in RLN takes place using various current technologies along with few frameworks. The development of automation scripts for the application ensures that the workflow is secure and accurate.
Software Integration to Achieve Poka-Yoke in Prepacking Tightness Automation
(2018-09-01T11:14:49Z) Monish, H S; Nayanashree, S; Poornima, V; Sahana, C
PF pump is a diesel fuel injection pump. This pumps are assembled by operators; after the assembly it is verified in the pre-packing tightness automation station. In this pre-packing tightness automation, the defects in the pf pumps are detected. This defects have to be eliminated otherwise it causes the customer complains. The common defects that occur in pf pumps are:  Air leakage  Pressure check  Leakage in pump joints If the pf pump has any one of these defects, it may lead to the serious damage in the engine management system. This defects are detected with the help of pre-packing tightness application. This application is developed by using the concept of interlocking and traceability which is used to test the pump defects and to achieve the mechanism of poka-yoke i.e. zero defect pumps. The aim of our project is to achieve zero defect pumps, reducing customer complains, enhancing productivity by minimizing time and cost. Keywords: poka-yoke, pf pumps, safety, quality, tightness test.
The Detail Study of Blackhole Mechanism on Manets
(2018-09-01T11:32:52Z) Pranish, Pokharel; Pratik, Sanjel; Kshetiz, Pokhrel; Kumar Sonali, Anand
A mobile ad hoc network (MANET) is a gathering of self-governing hubs that speak with each other by shaping a multi-jump radio system and keeping up associations in a decentralized way. Ensuring the system layer of a MANET from malignant assaults is a vital and testing security issue, since the greater part of the steering conventions for MANETs are helpless against different sorts of assaults. Ad hoc on-demand distance vector routing (AODV) is an extremely prominent steering calculation. Be that as it may, it is helpless against the outstanding dark gap assault, where a noxious hub dishonestly promotes great ways to a goal hub amid the course disclosure process yet drops all parcels in the information sending stage. This assault turns out to be more serious when a gathering of malevolent hubs participate each other. The proposed instrument does not have any significant bearing any cryptographic natives on the steering messages. Rather, it secures the system by recognizing and responding to malevolent exercises of the hubs. Recreation comes about demonstrate that the plan has an altogether high discovery rate with direct system movement overhead and calculation overhead in the hubs.
Efficient Data Deduplication for Cloud Storage
(2018-09-01T11:42:33Z) Anusha, A; M N V Kashyap, Sai; Khushdeep, Kaur; Mustafa, Sanwari
The explosive growth of digital contents continues to raise the demand for new storage and network capacities, along with an increasing need for more cost effective use of storage and network band-width for data transfer.With this increase of data production, shared network multi-user cloud storage systems are becoming very popular.However Security and privacy are among top concerns for the public cloud envi-ronments.Researchers have proposed many dynamic PoS schemes in single-user environments.In this project we propose an improved data deduplication scheme by improving its efficiency by identifying and removing the duplicate copies at block level from the cloud storages in multi-user environments. For implementation we are using message digest hash algorithm (MD5) to generate hashcode and Logical block addressing (LBA) to improve the performance of the proposed deduplication scheme .The security of the system is maintained by DES(Data Encryption Standard) which is a symmetric-key algorithm.
A RFID-Based Intelligent Child Security System
(2018-09-01T11:53:16Z) Sai Dhinesh Kumar, K; Karthigeyan, S
Children’s safety is one of the important aspects of the life. A system to increase the safety of the children is mandatory. The focus is on the daily route from home to school and vice versa, assuming the use of school buses. The bus safety system uses different localization techniques i.e. RFID (Radio Frequency Identification) and GPS (Global Positioning System), in order to design a solution for parents who can track the movements of the child i.e. taking the school bus and entering school or leaving school and entering the school bus. The applicability of RFID technology efficient tracking capabilities is tested in children’s tracking and monitoring during their trip to and from school by school buses.
IOT Based HealthCare Monitoring System
(2018-09-01T11:59:18Z) N Phani, Bhaskar; D.T., Rajesh; Kishan, Kunal; Vishal Anand, Jha
Advances in the information and communication technology leads to the development of IOT.IOT is the combination of the sensors, physical devices and computer that communicates with each other. Health is an important thing in our life due to the lack of doctors and on time health care services at the critical time many people are dying. So we proposed a system that uses the sensors attached to the patient’s body to collect the patient’s health data and send to the system and system will notify the doctors and the patient’s family members in critical situations through message service. These collected data is stored in database for further use. Security plays an important role in any system to achieve this a password protected Wi-Fi module is used.
Credit Card Fraud Detection Using Big Data
(2018-09-01T12:16:10Z) Rashmi, R; Sampurna, J; Shilpa Shree, N; Swati, Maurya
Credit card fraud detection is a wide-ranging term for theft and fraud committed using or involving a payment card, such as credit card or debit card, as a fraudulent source of funds in a transaction. Due to the fast growth of E-commerce, use of credit card for online purchases has dramatically increased and it caused and increase in the credit card fraud. It is relevant problem that draws the attention of Machine Learning. In the fraud detection task, there are some peculiarities present, such as the unavoidable condition of a strong class imbalance, the existence of unlabelled transactions, and the large number of records that must be processed. The present paper aims to propose a methodology for automatic detection of fraudulent transactions that tackle all these problems. The methodology is based on a Balanced Random Forest that can be used in supervised and semi-supervised scenarios through a co-training approach, which allows to compensate the class imbalance problem. In FDS, it has described the alert-feedback interaction, which is mechanism of providing recent supervised samples to train/update the classifiers. The main objective of FDS is to identify the fraud as soon as possible in order to take the necessary actions to revert it. Feedbacks play a central role in the proposed learning strategy.
Monitoring the Soil Moisture, Water Level Using Sensors and Early Crop Disease Detection Using Image Processing
(2018-09-01T12:25:34Z) Nishant, Kumar; Payal Jain, C; Rupendra Pratap, Singh
The prime need of this world is best agriculture which decides the development of each country as the survival of human being is completely dependent on farming and its best production. The main problem that is observed in most of the regions of farming is the early diseases in crops, no proper monitoring of soil moisture, water level and pH because of which the production results in low level identification of diseases on plant and It is also important to take the primitive measures in monitoring of soil moisture, salinity and PH of the growing crops. In this work, we propose a solution for early disease detection of crops using image processing and it also illustrates, how basic requirements of growing plants such as soil moisture, salinity and pH are controlled and monitored using IOT enabled Arduino sensors. The developed processing scheme for detecting the disease in plants consists of four main steps, first a color transformation structure for the input RGB image is created, this RGB is converted to HSI because RGB is for color generation and is for color descriptors. Then green pixels are masked and removed using specific threshold value, then the image is segmented and the useful segments are extracted, finally the texture statistics is computed. Finally the presence of diseases on the plant leaf is evaluated and if the result is disease in plant then an email is sent to the farmer else not. In the other part of the processing scheme soil moisture sensor is used for monitoring the moisture in the soil and the value of moisture is sent to farmer via
An Automated Approach to Headlamp Positioning and Accident Recognition
(2018-09-05T10:49:22Z) Akshay, V; D.K Ramachandra, Chowdary; Deepak, Bhat; Naonidh Signh, Ahuja
The main objective of this project is to significantly increase the safety of motorcyclists by using easily available technologies to automate the control of headlamps and detection of an accident thereby making the Helmet Intelligent. The proposed idea is achieved by using an Accelerometer, Piezoelectric Sensor, RF Transmitter and Receiver, Servo Motor attached to the Headlamp and GSM Module. All these components are connected to an Arduino Board which is the brain of the Helmet. Headlamps are automatically positioned as required which is calculated by the amount of Rider's Facial movement. The values from the Accelerometer and the Sensor are used to detect an accident, and if yes, alerts are sent to prioritized contacts through the GSM Module. To achieve the objective we make use of a Tri-axis Accelerometer used to measure acceleration forces in three perpendicular axes. The values obtained by the device are sent to Arduino board. This Arduino board is used for controlling the switches that are sent by the accelerometer. The RF Module is used for transmitting and receiving the signals that is connected to servo motors which is attached to headlamps. The Helmet is also equipped with a piezoelectric sensor to detect an accident. The pressure values obtained from the Accelerometer are sent to Arduino board. If both values get sudden changes, result in accident event being triggered. The GSM module is used for mobile communication. It provides a connection within the Helmet and sends an accident alert to phone. Thus, the development of this system marks a new step in the importance of road safety for motorcyclists thereby encouraging them to use the helmet at all times.
A Novel Approach to Detect Acute Lymphoblastic LEUKEMIA
(2018-09-05T10:55:21Z) Blessy, Thomas; Harshitha, R
This project focuses on the early detection of Leukemia can be made possible using recent emerged technologies in image processing wherein different algorithms and techniques can be used for processing the medical images digitally. The advancement in the devices such as microscopes and digital cameras has led to a high-quality input dataset. Firstly, the patients’ blood smear image is subjected to pre-processing, which includes image sharpening, contrast adjustment, and noise removal. Secondly, extract the features; reduce its size to minimize the computation time and storage requirement, selection of the best classifier. The aim here is to develop a foundation for a system which detects the harmful cancer cells, allowing a person to diagnose such hematic conditions at an early phase. The proposed system can be extended to a prototype of a handy machine in the near future. Ultimately, the motto of this proposal is to promote the principle - “Prevention is better than Cure”.
Sentiment Analysis of a Topic on TWITTER
(2018-09-05T11:16:24Z) Dhanush, M; Ijaz Nizami, S; Abhijit, Patra
The wide spread of World Wide Web has brought a new way of expressing the sentiments of individuals. It is also a medium with a huge amount of information where users can view the opinion of other users that are classified into different sentiment classes and are increasingly growing as a key factor in decision making. Social networking sites like Twitter, Facebook, Google+ are rapidly gaining popularity as they allow people to share and express their views about topics, have discussion with different communities, or post messages across the world. There has been lot of work in the field of sentiment analysis of twitter data. This survey focuses mainly on sentiment analysis of twitter data which is helpful to analyze the information in the tweets where opinions are highly unstructured, heterogeneous and are either positive or negative, or neutral in some cases. Twitter is one of the most popular social media sites and often becomes the primary source of information. Twitter provides both researchers and practitioners a free Application Programming Interface (API) which allows them to gather and analyse large data sets of tweets. Twitter data are not only tweet texts, as Twitter’s API provides more information to perform interesting research studies. The purpose of this work is to evaluate a topic and also calculate how many people have a positive or a negative view, based on the text reviews using sentiment analysis taking reviews from Twitter. Keywords: Popularity prediction, Sentiment Analysis, Natural Language Processing, Lexical Analysis I
Disease prediction using SVM and Machine Learning
(2018-09-05T11:26:53Z) Divya, Balaji; Deekshita, S; S, Harshita
Disease prediction is of critical importance with respect to advancements in health-car e as well as technology. With the advent of big data and analytics, it has become feasible to develop methods to diagnose and predict diseases in cost effective ways. However, the accuracy of analysis is decreased when the nature of medical data is incomplete. Furthermore, different regions exhibit distinctive characteristics of certain regional diseases, which have the probability to weaken the prediction of disease outbreaks. Thus, there lies a need to establish effective methods by which the predicted values can be separated for successful prediction. The predicted values are such that 0 denotes absence and 1 denotes presence, which is again used to check model performance. An algorithm in python is used to conduct this classification. 1
Home Automation and Security Using Speech Recognition
(2018-09-06T12:21:44Z) Diya, Jolly; Jennifer, Joseph; Maseera, Shajid
Home automation systems have become more efficient and cost effective in recent years due to the availability of cheaper hardware and relentless work of start-ups striving to make a dent in this industry. The advancement of technologies like Bluetooth, Wi-Fi, RF, ZigBee, WiMAX, Zwave, etc. paved way for developers to explore many alternatives to design smart home system. On the other hand, that many options also increases the ways in which security breaches can occur. Hence, maintaining security and integrity of such automation systems to prevent theft has not only become an area of interest but also an area of need. One such method which is hard to circumvent is speech recognition – easily integrated with the home automation system and requiring minimum user interaction. There are advantages and disadvantages to this method like any other and, are addressed in this paper. The sheer range of options available to help secure semi-automated or fully-automated home systems makes it cumbersome to address every point of failure. The following study helps an individual to choose a best fit method that suits his/her requirements based on the home automation system.
Programmed Maximization OF User Interface For Superior Customer Acquaintance
(2018-09-06T12:29:58Z) Rakshith, Sridharan; MD, Naushad; Sashikant S, Singh; Raj, Pratap
A literature survey or a literature review in a project report shows the various analyses and research made in the field of interest and the results already published, taking into account the various parameters of the project and the extent of the project. A literature survey includes the following  Existing theories about the topic which are accepted universally.  Books written on the topic, both generic and specific.  Research done in the field usually in the order of oldest to latest.  Challenges being faced and on-going work, if available. Literature survey describes about the existing work on the given project .It deals with the problem associated with the existing system and also gives user a clear knowledge on how to deal with the existing problems and how to provide solution to the existing problems.
Intelligent Approach for Classification of Grain Crop Seeds Using Machine Learning
(2018-09-06T12:34:04Z) Ranjith, Bose M; Ranjith, K; Subham Kumar, Singh; Suraj, Prakash
Seeds are the foundation of agriculture. Seed quality plays a vital role in the high yield production of crops. Seed analysis and classification helps in maintaining and achieving better quality of seeds. Generally, these activities are performed by specialists, by visually inspecting each sample, which is a very tedious and time-consuming task. Nowadays, the machine vision technology is widely used in this field to reduce time as well as work load on human, considering more samples at a time. The results achieved may not be much accurate, using this technique. So, in this paper, we use convolutional neural network with increased parameters to achieve better results. We use more feature extraction techniques, to improve the accuracy of classification.
Big Data Analytics for Anomaly Detection is Online Reviews
(2018-09-06T12:37:21Z) Rashmi I, Chamakeri; Roshnee Krishna, Murthy; T, Roja; Vaishnavi, M
Big Data is an evolving term that describes any voluminous amount of heterogeneous data with respect to (3v) variety, volume and velocity of the data. In a business model, the feedback of the customers is a primary concern as their review plays a crucial role in the business development. In the existing system, the detection of anomalies in wireless mobile networks is carried out by using spatio-temporal information about a user through big data analytics. This method involves anomaly detection but does not specify the type of anomaly. We propose a system that determines if the review given by a user about a particular event is true by tracking the location of the user in a backend process. We achieve this by using truth discovery algorithm, sentimental analysis and android-web server communication using WAP. This increases the security of the user and also enhances the performance of the system, regardless of the size of the dataset along with providing accurate results.
Performance and Evaluation of IPV4 & IPV6 Implemention for Transition Coexistence in Networks
(2018-09-06T12:40:33Z) Satish, Raj; Saurabh, Sunny; T Sudarshan, Rao; Yashish, Thacker
Due to the increasing demand of the Internet, we are facing a great problem of the depletion of our existing IPv4 (Internet Protocol version 4) network. To solve the situation, we have to use IP version 6 in coming years. But the IPv4 network will not be opt out, but also coexist with IPv6 network. For the transition from IPv4 to IPv6 and vice versa, there are three prominent transition mechanisms are used. They are Dual Stack, Tunneling and Translation. In this paper, the performances of these three mechanisms have been analyzed. IPv6 header format, its security and the routing also have been focused. For the simulation Packet Tracer simulation software has been used. Keywords: IPV4, IPV6, Dual stack Server, Tunneling, Translation, Cisco Packet Tracer.

Browse

Browsing 2017-18 by Issue Date

Results Per Page

Sort Options