Information Technology

Document segmentation is a method of rending the document into distinct regions. A document is an assortment of information and a standard mode of conveying information to others. Pursuance of data from documents involves ton of human effort, time intense and might severely prohibit the usage of data systems. So, automatic information pursuance from the document has become a big issue. It is been shown that document segmentation will facilitate to beat such problems. This paper proposes a new approach to segment and classify the document regions as text, image, drawings and table. Document image is divided into blocks using Run length smearing rule and features are extracted from every blocks. Discipulus tool has been used to construct the Genetic programming based classifier model and located 97.5% classification accuracy.

The main aim of this paper is the briefing of secure locker systems having authorized input methods of fingerprints as well as an embedded burglar alarm system. The system can be used for domestic, commercial or industrial purpose. The purpose of this project is to provide a secured and reliable environment to the users for their safety valves by providing a unique identity to every user using the FINGER PRINT identification technology. Finger print authorization is required for activation of the scanner. Scanner is interfaced to the micro controller with the serial interfacing. The micro controller reads the data from the scanner. The micro controller allows those users, who are authorized to operate the account. If any unauthorized user tries to operate the account the micro controller, a warning is made. After a certain trial if finger print is not matched, scanner will block the system. The total information about the account holders is stored in the EEPROM. Keypad is used to enter the password to operate the account or Locker. A special alarm system is also used in this project. If a person is threatened by anyone, user can use any other finger for critical conditions, such as emergency. During such cases, system gets locked and a warning is sent to the emergency services using alarm burglar system.

The liver supports almost every organ in the body and is vital for our survival. Liver disease may not cause any symptoms at earlier stage or the symptoms may be vague, like weakness and loss of energy. Symptoms partly depend on the type and the extent of liver disease. Liver diseases are diagnosed based on the liver functional test. Though this disease cannot be predicted at earlier stage due to lack of symptoms and signs, in this paper we attempt to apply soft computing technique for intelligent diagnosis of liver disease. The classification and its type detection are implemented in three phases. In first phase, ANN classification is applied for classifying the liver disease. In second phase rough set rule induction using LEM algorithm is applied to generate classification rules. This rule induction overcomes the drawback of MLP and hence improves the accuracy. in third phase fuzzy rules are applied to identify the types of the liver disease. Using LEM algorithm 6 rules are generated with accuracy of 96% in correct classification. On applying rules generated by LEM, improves the classification accuracy by 6% compared to MLP. The 4 fuzzy rules are framed to identify the types of liver disease.

Now a days Public and private organizations increase their database system requirement for day-to-day business. Hence database security becomes more crucial as the scale of database is growing. A signified approach for protecting information which enforcing access control policies based on subject and object and their characteristics. There are many security models for database systems. The database security systems have developed a number of different access control policies for assuring data confidentiality, integrity and availability. In this paper we survey the concepts under access control policies for database security. We review the key access control policies such as Mandatory Access Control policy(MAC), Discretionary Access Control Policy(DAC), and Role Based Access Control Policy(RBAC) and propose a concept on RBAC policy that is instead of access control through role assigned to the users, the users are assigned by some level of access control.

The paper proposes Handwritten Character Recognition method using 2D view and Support Vector Machine (SVM). In this all the character images are Pre-processed (includes Normalization and Noise Removal), which are further used for feature extraction using two dimensional (2D) views. From each character four different views (Top, Bottom, Left, and Right) are obtained called as basic views. All basic views are not able to collect the complete information of character image. The hidden information is capture separately called as extra views. From each view 16 features are extracted and combined to obtain 80 features. These features are used to train SVM to separate different classes of characters. Handwritten Character database is used for training and testing of SVM classifier. Support Vector Machine provides a good recognition result for lower case characters and upper case characters are 82%.

Mobile ad hoc networks will appear in environments where the nodes of the networks are absent and have little or no physical protection against tampering. The wireless nodes of MANET are thus susceptible to compromise and are particularly vulnerable to denial of service (DoS) attacks launched by malicious nodes or intruders. Flooding attack is one such type of DoS attack, in which a compromised node floods the entire network by sending a large number of fake RREQs to nonexistent nodes in the network, thus resulting in network congestion. In this paper, the security of MANET AODV routing protocol is investigated by identifying the impact of flooding attack on it. A simulation study of the effects of flooding attack on the performance of the AODV routing protocol is presented using random waypoint mobility model The simulation environment is implemented by using the NS-3 network simulator. It is observed that due to the presence of such malicious nodes, average percentage of packet loss in the network, average routing overhead and average bandwidth requirement? all increases, thus degrading the performance of MANET significantly.

Data Mining is the use of algorithms to extract the information and patterns derived by the knowledge discovery in databases process. Classification maps data into predefined groups or classes. It is often referred to as supervised learning because the classes are determined before examining the data. The Verification of handwritten Signature, which is a behavioral biometric, can be classified into off-line and online signature verification methods. The feasibility and the benefits of the proposed approach are demonstrated by means of data mining problem: online Signature Verification. This paper addresses using ensemble approach of Genetic Algorithm for online Signature Verification. Online signature verification, in general, gives a higher verification rate than off-line verification methods, because of its use of both static and dynamic features of problem space in contrast to off-line which uses only the static features. We show that proposed ensemble of Genetic Algorithm is superior to individual approach for Signature Verification in terms of classification rate.

In this study an extended geophysical survey using gravity and high angle refracted seismic methods with recorded earthquake has been carried out in Kerman province in southeast of Iran. The purpose of the applied geophysical surveys was to provide information on crustal structure and lithospheric thickness. The purpose of this research is focused on determination of Moho depth and for this reason several profiles for both methods were performed and by use of topography and gravity data with a reasonable combination with high angle refracted seismic data and earthquake record, the crustal and lithospheric thickness of Kerman province was calculated. Oldenburg- Parker algorithm was the base of gravity survey of the present work but by use of Matlab programming environment and generated C Sharp computer codes was optimized, improve and then executed. The key factor of this study was generated computer code, which named as “MOHO R.A.T 1.01”. For seismic data, a least-squares analysis of the travel-time data was made and the uncertainties were taking in to account. Finally, depth calculations for the velocity discontinuities and gravity anomaly contour maps were made. The comprehensive comparison between the obtained results and other studies on the selected area showed good agreement, which verified the ability of produced code.

Bloom Filter (BF) is a simple but powerful data structure that can check membership to a static set. The trade-off to use Bloom filter is a certain configurable risk of false positives. The odds of a false positive can be made very low if the hash bitmap is sufficiently large. Spam is an irrelevant or inappropriate message sent on the internet to a large number of newsgroups or users. A spam word is a list of well-known words that often appear in spam mails. The proposed system of Bin Bloom Filter (BBF) groups the words into number of bins with different false positive rates based on the weights of the spam words for spam filtering. Simulated Annealing (SA) is stimulated by an analogy to annealing in solids. It is used to search for feasible solutions and converge to an optimal solution. In this paper SA is applied to minimize the total membership invalidation cost of BBF. The experimental results are analyzed for various sizes of bins. The results show that, the BBF using SA with different false positive rate has lower total membership invalidation cost than the standard BF.

Weather forecasting is a very fundamental application in meteorology and technologically challenging problem. The estimation of temperature values are needed for agricultural, technical and environmental applications. Meteorological dataset has historical data of all weather parameters and the temporal analysis of this dataset can help to mine meaningful knowledge. There are number of techniques available in data mining, but Association Rule Mining is one of the most popular technique to mine large amount of dataset for finding the hidden relationship between various dataset variables values and identifies correlations between them. The scope of this research is to analyze temporal rules generated to predict day to day temperature variation of a specific region Surat, India. To accomplish this, the framework is proposed for prediction of day temperature variation from seasons. From the experiments, achieved higher accuracy compare to other data mining technique and the rules which show how day variations are related. Also prepared the list of parameters which is less in number to help for the prediction instead of all parameters and thus it helps in the reduction of the dataset size.