Showing papers by "KCG College of Technology published in 2012"

An empirical study on prediction of heart disease using classification data mining techniques

Abstract: In this research paper, the use of pattern recognition and data mining techniques into risk prediction models in the clinical domain of cardiovascular medicine is proposed. The data is to be modelled and classified by using classification data mining technique. Some of the limitations of the conventional medical scoring systems are that there is a presence of intrinsic linear combinations of variables in the input set and hence they are not adept at modelling nonlinear complex interactions in medical domains. This limitation is handled in this research by use of classification models which can implicitly detect complex nonlinear relationships between dependent and independent variables as well as the ability to detect all possible interactions between predictor variables.

Robust approach for palm (Roi) extraction in palmprint recognition system

Abstract: Palmprint based biometric system is potentially a good choice for biometric applications due to its richness in amount of features. Palmprint contains geometric features, line features, point features, texture features, and statistical features. One of the important stages in the system is pre-processing which contains some operations such as filtering, Morphological operations, Region of Interest (ROI) extraction, normalization, etc. In this paper, we propose a simple methodology for localizing the palm and extracting the ROI. Although pre-processing has been discussed in different papers incidentally, but no definite precise approach has been introduced. We pay more attention towards more essential stage of palm Localization and ROI Extraction.

Enabling visually impaired to read messages from modern gadgets

Abstract: This paper introduces a design of a new independent electronic device which will enable a visually challenged person to read messages from cell phones to emails. These devices operate at a very low voltage and are attached wirelessly through Bluetooth to the modern gadgets. The visually challenged people are more sensitive to touch at the finger tip when compared to the normal individuals. Taking advantage of this extra sensitiveness of the visually challenged person, these devices does the conversion of alphabets into vibrations in a definite format. The vibration is generated from the small DC vibrator motors placed in 3×2 matrix formats. Despite of technological development these persons are unable to utilize the opportunity to share messages through modern gadgets without help from normal persons. This paper is an initiative to reduce the difficulties faced by the visually challenged persons to read messages from modern gadgets and to enable them to read messages with ease and comfort autonomously.

Classification of color objects like fruits using probability density function (PDF)

Abstract: Fruits like apples are valued based on their appearance (i.e. color, sizes, shapes, presence of surface defects) and hence classified into different grades. Grading process helps in achieving better standards and quality of fruits. Of the many available color models, HSI model provides a highly effective color evaluation particularly for analyzing biological products. Human assessment furnishes only qualitative data and such inspection is time consuming and cost-intensive. Machine vision systems with specialized image processing software provide a solution that may satisfy the demand. The analysis was carried out on images of 187 apple fruits, shows that classification done based on median of PDF. In order to avoid the mismatch in grading the same it has been classified further using Histogram Intersection, which determines the closeness between two images i.e. 1 if two images are similar and 0 if they are dissimilar.

Tunable energy transfer distributed feedback dye laser using pyronin B and crystal violet dye mixture

Abstract: Using pyronin B and crystal violet dye mixture, the characteristics of energy transfer distributed feedback dye laser (ETDFDL) pumped by 532 nm Nd:YAG laser were investigated. The characteristics of donor DFDL and acceptor DFDL as well as the dependence of their pulse widths and output powers on pump power and acceptor concentrations were studied. The output energy of DFDL was measured at the emission peaks of donor and acceptor dyes for different pump powers and acceptor concentrations, and tunability was observed from 570 to 645 nm using prism dye cell arrangement.

An efficient multiband antenna for satellite and 4G communication systems

Abstract: In this paper, an efficient and compact multi-band micro strip patch antenna for both the 4G WiMAX (3.4GHz) and for satellite applications (4.9GHz) is developed. The three resonant modes of the proposed antenna are 3.4 GHz, 4.4 GHz, 4.9 GHz. By proper adjustment of the U shaped slotted patch antenna enables the operation of multi band with a -10dB return loss, a omni directional radiation pattern and a good antenna gain with high efficiency. Design simulation and optimization processes are carried out with the aid of the Advanced Design System (ADS) electromagnetic simulator that uses the full-wave Method of Moment (MoM) numerical technique [3]. The losses that are caused by both the substrate and metallization are taken into account during the simulation process. Antenna compactness was our target during the design process to fix the antenna in the new 4G communication mobile equipments. The proposed antenna is suitable for wireless communication applications requiring a small antenna.

Preventive approach to avoid intrusion in relational databases using response patterns

Abstract: A rash of well-publicized database hacking incidents brings security to the forefront of issues faced by the database community. Anomalous requests are processed by the Intrusion Detection System (IDS). The intrusion response component is the major responsible component in IDS for anomalous requests processing. Response component manages some of the policy for authenticating the user requests for further access. The system is proposed with the Database Response Policies with two issues Policy Matching and Policy Administration. This system is termed as Joint Threshold Administrative Module (JTAM). The principle used in JTAM is Separation of Duty. The major idea is used to prevent the malicious modification of the policies by database administrators (DBA's). The database modification proposed ought to be signed by k DBA's for complete modified policy. Cryptographic algorithms are used to encrypt the policy and the signatures. The policies are limited for the user's view and for this `pattern concept' is used. An Intrusion Prevention Protocol (IPP) is proposed which detects the users that attacks the system and revokes the access rights of that user.

Automated mobile physiological measurement system

Abstract: The convergence of an aging population and breakthrough technological advances has created endless opportunities for automated medical devices. These devices help ensure the future health of millions of people by providing advances in home health care, clinical activities and medical imaging. But these devices could only measure a single parameter and just display the data. This paper discusses the health monitoring system using microcontrollers and GSM technology. It is a combination of ubiquitous computing with mobile health technology to monitor the wellbeing of patients. The microcontroller analyses the data in real-time and determines whether the person needs external help. It interfaces with medical devices including a blood-pressure monitor, a heart rate monitor and an electronic thermometer and offer suggestions based on the readings. If the readings are abnormal then it sends messages to the care takers and in case of critical conditions it automatically gives call/message to the ambulance. It mainly focuses to improve the health of people and could ease the burden on the health-care system.

Design and development of a microcontroller based neuro fuzzy temperature controller

Abstract: The temperature control in plastic extrusion system is a crucial process, which is an important factor to produce high quality plastic products. The plastic extrusion model is a first order non linear system has multiple stages coupled with each other and comprises a long delay time, large time constant and undesirable overshoot. To acquire a solution to the problem of difficulty to control temperature in process industries, a new adaptive control strategy, embedded based neuro fuzzy controller is proposed. The software incorporates Matlab simulink and real time control algorithm has been developed using PIC 16F877A microcontroller. The proposed method hardware test results are compared with that of the simulated results. The results obtained conclude that the use of proposed controller improve the process in terms of time domain specification, set point tracking and rejection of disturbances more effectively and rapidly with optimum stability for the temperature control in plastic extrusion system.

Reliable routing strategy for data communication in vehicular adhoc networks

Abstract: VANET is a network formed by vehicles communicating among themselves for safety and comfort messages. In VANET, high speed of nodes and dynamic change in topology directly impacts the network that has to deliver the data. Routing is a major issue in case an exchange of data packet uses an invalid path. This work is going to overcome the dropping of data packet by selecting an optimal path. Path can be selected based on two parameters such as speed and vehicle's position. These parameters are used by ant colony optimization (Bio-inspired) procedures in order to provide routing decisions. In an existing work, nodes in cluster lead to resource consumption such as buffer where all the nodes in the cluster has maintained buffer for data delivery. The proposed scheme is being used for mitigating the resource consumption by keeping the buffer only at the cluster head and also it employs autonomous clustering procedures. Finally, the proposed algorithm SI-DYMO performance is going to be compared with the AODV with respect to data delivery ratio, delay, transmission cost and throughput.