International Journal on Future Revolution in Computer Science & Communication Engineering
Not a member yet
1384 research outputs found
Sort by
Shadow Detection and Removal using Artificial Neural Network
Get PDFShadow detection and removal is an important task when dealing with colour images. Shadows are generated by a local and relative absence of light or a shadow appears on an area when the light from a source cannot reach the area due to obstruction by an object. Shadows are, first of all, a local decrease in the amount of light that reaches a surface. Secondly, they are a local change in the amount of light rejected by a surface toward the observer. However, they cause problems in computer vision applications, such as segmentation, object detection and object counting. Thus shadow detection and removal is a preprocessing task in computer vision. This thesis work proposes a simple method to detect and remove shadow from a single RGB image using artificial neural network. A shadow detection method is selected based on the phenomena of back propagation algorithm. Back propagation artificial neural network classifier has been used to train and test the neural network based on the extracted feature. The shadow removal is done by multiplying the shadow region by a constant
A Novel Fuzzy Clustering Algorithm for Radial Basis Function Neural Network
Get PDFA Fuzzy Radial basis function neural network (FRBFNN) classifier is proposed in the framework of Radial basis function neural network (RBFNN). This classifier is constructed using class-specific fuzzy clustering to form the clusters which represent the neurons i.e. fuzzy set hyperspheres (FSHs) in the hidden layer of FRBFNN. The creation of these FSHs is based on the maximum spread from inter-class information and intra-class fuzzy membership mechanism. The proposed approach is fast, independent of parameters, and shows good data visualization. The Least mean square training between the hidden layer to output layer in RBFNN is avoided, thus reduces the time complexity. The FRBFNN is trained quickly due to the fast converge of input data to form the FHSs in the hidden layer. The output is determined by the union operation of the FHSs outputs which are connected to the class nodes in the output layer. The performance of the proposed FRBFNN is compared with the other RBFNNs using ten benchmark datasets. The empirical findings demonstrate that the proposed FRBFNN is highly efficient classifier for pattern recognition
IOT Based Smart Agriculture And Soil Nutrient Detection System
Get PDFDevelopment of agriculture using technology will be very much useful in cultivation. For a new agricultural area, without knowing or monitoring the important parameters of the soil, cultivation will be difficult and so the farmers suffer financial losses. This project provides a brief overview of the soil monitoring system using sensors. Various soil sensors are used to measure temperature, moisture and light, humidity and ph value. The information from the sensors in the soil is sent to the MCP3204 A/D converter then from A/D converter it send to the cloud through Raspberry pi. Finally we can see the information saved to cloud on mobile phone as well as laptop. On the basis of information we know which crop is suitable with given soil parameter. Thus this advanced technology helps the farmers to know the accurate parameters of the soil thus making the soil testing procedure easier
Smart Three Phase Crawler for Mining Deep Web Interfaces
Get PDFAs deep web develops at a quick pace, there has been expanded enthusiasm for strategies that assistance effectively find deep-web interfaces. Nonetheless, because of the extensive volume of web assets and the dynamic idea of deep web, accomplishing wide scope and high effectiveness is a testing issue. In this task propose a three-stage framework, for proficient reaping deep web interfaces. In the principal stage, web crawler performs website based scanning for focus pages with the assistance of web search tools, abstaining from going by a substantial number of pages. In this paper we have made an overview on how web crawler functions and what are the approaches accessible in existing framework from various scientists
Location Management in Wireless Sensor Networks with Mobility
Get PDFWireless sensor networks comprise motes which are nothing but small sensor devices. The challenging problems for motes are battery power, storage capacity, and less calculation power of the mote. In this paper developed structure for Real-Time Tracking, Sensing and Management System using IITH motes is proposed. Also the algorithm developed for location management of wireless sensor networks with the aspect of mobility is proposed. This developed framework and algorithm can be utilized in emergency events and safety threats and provides warning signals to handle the emergency
Monitoring File System for Windows Information Security
Get PDFEvery organization�s asset is its data and data are stored in files which are maintained by file systems. Therefore, it is an important role of an organization to keep its File System secure. There is huge amount of changes that are made on daily basis in these files by different users. Hidden among these changes can be the few that are illegitimate and can cause harm to organization. So, File System Monitoring becomes necessary. While many such monitoring tools are available for UNIX/Linux systems [1], very little is done for Windows system. We have developed a File System Monitoring application for Windows operating system which monitors auditing of file systems � specifically, you want to know who read, modified, deleted or created files in a shared area. While there were many options available for implementing such an application, the most appropriate way of doing so is by exploiting native compatibility of C
Dashboard for Learning Outcome based Attainment Measurement
Get PDFAICTE (All India Council of Technical Education) specifically requires colleges under it to adhere to certain guidelines regarding the courses taught. Each course has certain Course Outcomes (CO) which tell what the students are supposed to learn and also indicate to what degree they have learnt the subject. Similarly, Program Outcomes (PO) indicate the same about the programs taught in the college. Professors currently maintain all this data manually which means a lot work is put in Excel documents and huge files are maintained. This system helps in achieving the error free attainment calculations as it eliminates the manual calculation using spreadsheets. It helps to increase the productivity of the faculties as the manual work is eliminated and data is stored and retrieved in a safe and secure manner
Application of the Variance Function of the Difference Between two estimated responses in regulating Blood Sugar Level in a Diabetic patient using Herbal Formula
Get PDFIn this paper response surface methodology has been employed in investigating effectiveness of herbal formula (a mix of six herbs) in regulating the blood sugar level of a diabetic to within acceptable levels. In the experimentation phase, observations are made to investigate the effectiveness for particular level of concentration at regulating the blood sugar level with time. The most feasible of all the identified points of equal yield has been identified as one in which the variance function is minimal. In this investigative research we use the variance function of the difference between two points to provide reliable advice on the range around which the dosage is desirable and time required to effectively regulate the blood sugar level to within acceptable range. In the set up the herbal formula extract has been shown to have successfully regulated the blood sugar level in a diabetic to 11.3898 mMol/L. This is possible by effecting a treatment of herbal formula at a concentration of 66.1125 mg/dl and the effectiveness is within 175.4580 minutes upon treatment
Secure Authentication Model using Grid based Graphical Images with Three Way Validation
Get PDFThe most common computer authentication method is to use text usernames and passwords which have various drawbacks. For example users tend to pick passwords that can be easily guessed. On the other hand, if a password is hard to guess, then it is often hard to remember. This paper provides additional layer of security to normal textual password by using graphical password for authenticating the user. As graphical passwords are vulnerable to shoulder surfing attack so we will send one-time generated password to users and even send credentials to users authorized email-id. Using the instant messaging service available in internet, user will obtain the One Time Password (OTP)
Performance analysis of Handwritten Devnagari Character Recognition using Feed Forward , Radial Basis , Elman Back Propagation, and Pattern Recognition Neural Network Model Using Different Feature Extraction Methods
Get PDFThis paper describes the performance analysis for the four types of neural network with different feature extraction methods for character recognition of hand written devnagari alphabets. We have implemented four types of networks i.e. Feed forward , Radial basis, Elman back propagation and Pattern recognition neural network using three different types of feature extraction methods i.e. pixel value, histogram and blocks mean for each network. These algorithms have been performed better than the conventional approaches of neural network for pattern recognition. It has been analyzed that the Radial Basis neural network performs better compared to other types of networks