Venue : Vel Tech University, Chennai, India

MARCH 7-8, 2015

Home Paper Submission Program Committee Accepted Papers Contact Us

Accepted Papers


  • Automated Tuberculosis Screening Using Zeihl-Neelson Method
    Lavanya Govindan,Rajalakshmi Engineering College,Chennai, India
    ABSTRACT
    In 2013, according to WHO report India is the highest burden disease for Tuberculosis. Earlier diagnosis of TB may save many patient's life and their neighbor. But for earlier diagnosis of TB disease practice in India is still challenging problem. Ziehl-Neelson(ZN) staining, a microscopic method is most common and conventional way for diagnosis of Tuberculosis bacteria. In this proposed work, the segmentation of tuberculosis bacilli and classification of TB-positive and negative cases are performed depending upon the features of bacilli. The algorithm is implemented in MATLAB, and Graphical User Interface (GUI) created using Java wrapper and integrate them into Java code.
  • COLOR IMAGE COMPRESSION BY USING HYBRID DISCRETE WAVELET & COSINE TRANSFORMS
    Vijayakumar K ,Kalyan chakravarthi P and Suresh D,GMR IT ,Rajam,Andhrapradesh,India
    ABSTRACT
    The need for an efficient technique for compression of Images ever increasing because the raw images need large amounts of disk space seems to be a big disadvantage during transmission & storage. Even though there are so many compression techniques like JPEG & JPEG-2000 already present. A better technique which is faster, memory efficient and simple surely suits their requirements of the user. We took the disadvantages in JPEG & JPEG-2000 to overcome that in this paper we proposed the Lossless method of colour image compression using hybrid discrete wavelet & discrete cosine transforms. This technique is simple in implementation and utilizes less memory. A software algorithm has been developed and implemented to compress and the given colour image using hybrid discrete wavelet & discrete cosine transform in a MATLAB platform.
  • DETECTION OF ABNORMAL BLOOD VESSELS IN DIABETIC RETINOPATHY BASED ON BRIGHTNESS VARIATIONS IN SDOCT RETINAL IMAGES
    D.N.Archana,Rajalakshmi Engineering College, Chennai, India
    ABSTRACT
    Diabetic Retinopathy accounts for nearly 5% of the world's 39 million people who suffer from blindness. The number of people at the risk of diabetic retinopathy is on the rise. Currently there are more than 382 million people have diabetes worldwide. Diabetic retinopathy is the most common diabetic eye disease, which occurs due to damage in the blood vessels of the retina. Sometimes of these vessels swell and leak fluid or even close off completely; abnormal new blood vessels grow on the surface of the retina. At the early stage of diabetic retinopathy that is mild non proliferative diabetic retinopathy, there will be formation of micro aneurysms, hard exudates, hemorrhages, soft woolen spots. At the proliferative stage abnormal blood vessels begin to grow and this leads to loss of vision. In this proposed work retinal layer is segmented and the abnormal blood vessels are segmented at the early stage of DR and the thickness of abnormal layer is compared with that of normal layer using MATLAB software image processing tool. RNFL Thickening shows the presence of blood vessels, shadows at the boundary of RNFL layer shows the location of the blood vessels.
  • CBIR BASED DOCUMENT AND IMAGE IN DIGITAL LIBRARY
    D.Geetha Priya,Rajalakshmi Engineering College, Chennai, India
    ABSTRACT
    The current trends in the applications of visual Search techniques into digital library Image Retrieval techniques in the field of Digital Libraries. Content Based Image Retrieval (CBIR) is a technique uses visual contents to search images from large scale image database in an active area of research for the past decade. Presented an algorithm for retrieving images with respect to database consisting of color, shape and texture features. By applying CBIR technique the content based image properties extract the color, shape and texture from the image. The algorithm uses the shape information in an image along with its 2D information, color features extracted for red, green, blue values. A linear approximation procedure that can capture the depth information using the idea of shape from shading has been used. Find the similarity matching between the images with the content properties. Applying the Relevance Feedback algorithm to rank the images in decreasing order. Most similar images are ranked top in the database then less similar images are placed next in the database. If user didn't get needed images then user can give the feedback to iterate the images to get the similar images.
  • A Novel Approach of Morphological Based Watershed Algorithm using Distance Transform to Image Segmentation
    Debasree Mitra, Aparupa Sarkar ,Debosree Cjakarborty ,Avijit Das and Nimmi Roy,JIS College of Engineering, Kalyani,Nadia,India
    ABSTRACT
    All the different types of the watershed algorithm are not equally well suited for hardware implementation. In this paper we will discuss about Morphological Based Watershed Algorithm using Distance Transform to Image Segmentation. The proposed algorithm will detect a detailed and an accurate image.
  • Scrambling based robust image watermarking in DWT-DCT fused domain
    T. Meenpal, Aditi Jain and Ankita Jain,NIT Raipur,India
    ABSTRACT
    In this paper, a new robust digital image watermarking algorithm based on joint Discrete Wavelet Transform (DWT)-Discrete Cosine Transform (DCT) is proposed. The original watermark image is scrambled by Arnold transformation. Original image is decomposed into various sub-bands by applying a 3-level DWT. Unlike the existing wavelet based watermarking techniques, the proposed scheme concentrates on marking the horizontal and vertical detail sub-bands of 3rd level DWT of the cover image. It leads to achieving a better balance between fidelity and robustness. DCT of each selected DWT sub-band is computed. The scrambled watermark is embedded in the DCT middle frequency coefficients by spread spectrum technique generating different PN-sequence for '0' and '1'. To minimize the effect of intentional/un-intentional noise, the watermarked image is first preprocessed by sharpening and Laplacian of Gaussian filters. Similar approach as the embedding process is then used to extract the scrambled watermark bits. Inverse of the scrambling operation is applied later to recover the original watermark. The proposed algorithm has the advantages of robustness as well as security due to scrambling operation.
  • Towards Interesting Rare Itemset Mining using Tree Structure
    Urvi Bhatt and Pratik Patel ,Gujarat Technological University,Gujarat, India
    ABSTRACT
    Pattern mining methods describe valuable and advantageous items from a large amount of records stored in corporate datasets and repositories. While mining, literature has been almost singularly focused on frequent itemset but in many applications rare ones are of higher interest. Example of such application can be a medical dataset where rare amalgamation of prodrome plays a vital role for the physicians. As rare items contain worthwhile information, researchers are making efforts to examine effective methodologies to extract the same. In this paper, we make an effort to analyze the complete set of rare items for finding the most off all rare association rules from the dataset. Proposed approach uses Maximum constraint model for extracting rare items. A new approach is efficient to mine rare association rules which can be defined as the rules containing the rare items. Based on the study of relevant data structures of the mining space, our approach utilizes the tree structure to ascertain the rare items. Finally we have tried to demonstrate that, this approach is more virtuous and robust than the existing algorithms.
  • REDUCTION OF AZIMUTH UNCERTAINITIES IN SAR IMAGES USING SELECTIVE RESTORATION
    S.Kalaivani and Mr.P.S.Ramesh,Arunai Engineering College, Thiruvannamalai, India
    ABSTRACT
    A framework is proposed for reduction of azimuth uncertainty space borne strip map synthetic aperture radar (SAR) images. In this paper, the azimuth uncertainty in SAR images are identified by using a local average SAR image, system parameter, and a distinct metric derived from azimuth antenna pattern. The distinct metric helps isolate targets lying at locations of uncertainty. The method for restoration of uncertainty regions is choosed on the basis of size of uncertainty regions. A compressive imaging technique is engaged to bring back isolated ambiguity regions (smaller regions of interrelated pixels), clustered regions (relatively bigger regions of interrelated pixels) are filled by using exemplar-based in painting. The recreation results on a real Terra SAR-X data set established that the proposed method can effectively remove azimuth uncertainties and enhance SAR image quality.
  • EFFECTUATION OF SALIENCY DETECTION IN MOTION FIELDS
    Ms.S.Dhoulath and Ms.R.Jayachitra ,Arunai Engineering College, Thiruvannamalai, India
    ABSTRACT
    Automatically detecting people in videos is the first step in a wide range of tracking applications, which can be successively used in video surveillance systems, traffic monitoring, Object Detection and Action Recognition systems. Therefore detecting the outstanding feature of an image is considered to be more important. The outstanding feature of an image with respect to its neighborhood pixel is termed as "Saliency". In this proposed work, a Locally Adaptive Regression Kernel based object detection is employed for detection process. The proposed method is found to be simple and effective since the process does not require prior knowledge about objects in videos and also the model is parameter independent fast process.
  • Pattern Recognition & Matching using Refine Chain Code
    Pooja Dabi and Praveen Bhanodia ,Patel college of science &technology,Indore,India
    ABSTRACT
    Pattern recognition is a major problem in image understanding and computer vision, where identification of Shape is important. Object recognition is an extension of Shape recognition. To recognize an object, it is necessary to identify first Shape of object then extend it for object recognition. This paper focuses on recognizing a Shape and Shape matching technique based on their chain codes. The main advantage of chain code techniques is that it is invariant to translation, rotation and scaling. In this process we also describe number of objects in the image, number of corners in each Shape of the image and also describe types of Shape. The main goal of this paper is to matching two or more images with different Shape based on their chain code.
  • Sensorless Vector Control of BLDC Using Extended Kalman Filter
    Y.Lavanya and S.P.G.Bhavani ,Meenakshi College of Engineering, Chennai,India
    ABSTRACT
    This Paper mainly deals with the implementation of vector control technique using the brushless DC motor (BLDC). Generally tachogenerators, resolvers or incremental encoders are used to detect the speed. These sensors require careful mounting and alignment, and special attention is required with electrical noises. A speed sensor need additional space for mounting and maintenance and hence increases the cost and size of the drive system. These problems are eliminated by speed sensor less vector control by using Extended Kalman Filter and Back EMF method for position sensing. By using the EKF method and Back EMF method, the sensor less vector control of BLDC is implemented and its simulation using MATLAB/SIMULINK and hardware kit is implemented.
  • A hybrid neural network model for Active Noise Cancellation using TMS320C6713 DSK
    Shashi Kumar1, Rajeswari2, G. Ramanjulu3, NNSSRK Prasad41,2Department of ECE, Acharya Institute of Technology, Bangalore, India, 3,4Aeronautical Defense Agency (Ministry of Defense), Bangalore, India
    ABSTRACT
    This paper discusses the implementation and performance analysis of varied neural network algorithms for Active Noise Cancellation (ANC) using TMS320C6713 processor. A hybrid neural network model exploring the feed-forward and feedback neural network models has been proposed. The proposed system is suitable for high frequency noise signal with lower SNR levels like cockpit noise embedded low frequency audio signals. The experimental results indicate that the proposed hybrid model provides a better and stable control signal, with an improvement in SNR for Active Noise Cancellation when compare to feed-forward and feedback neural models.
  • Incentivize Cooperative Spectrum Sensing in Cognitive Radio using data fusion
    M.Divya and V.Saravanan, Arunai Engineering College, Thiruvannamalai, India
    ABSTRACT
    Cooperative spectrum sensing for cognitive radio network with different data fusion rules are proposed in this work. Centralized sensing is used to collect sensing information through central unit from other cognitive devices. The main objective of this work is to prevent the interference with Primary Users (PU) and identifies the available white spaces to enhance spectrum utilization. Energy detection technique is used to sense the existence of primary user (PU) signal. Cooperative spectrum sensing intensify the reliability of detecting primary users by data fusion rule. The performance of cooperative spectrum sensing is evaluated with the hard combination OR, AND and MAJORITY rules. The simulation results show that MAJORITY rule is near optimal for the desirable amount of false alarm and detection rates.
  • DESIGN AND IMPLEMENTATION OF TRUNCATED MULTIPLIER IN FIR FILTER
    R.MUTHAMMAL and SANDHYA.S, GKM COLLEGE OF ENGG AND TECHNOLOGY Chennai, India
    ABSTRACT
    Low-cost finite impulse response (FIR) designs are presented using the concept of faithfully rounded truncated multiplier. This multiplier design is usually considered where the maximum absolute error is no more than 1 unit of least position. And also this truncated multipliers offer significant improvement in area, delay and power. The proposed method jointly consider the deletion, reduction, truncation and rounding of partial product bits in order to minimize the number of full adders and half adders during tree reduction. In addition, the truncated multiplier design also has smaller delay due to the smaller bit width in the final carry-propagate adder. In previous papers truncation error is reduced by adding error compensation circuits in fixed width multiplier to get a precised output. But here, there is no need of error compensation circuits and the final output will be precised. The proposed filter using truncated multiplier will be designed using VerilogHDL and synthesis using ISE Simulator (ISIM) and simulate it using MODELSIM ALTERA 6.4a (Quartus II 9.2i).It achieves best area and power result when compared with previous FIR design approaches.
  • DETECTION OF UNTENANTED SPECTRUM IN COGNITIVE RADIO USING NON COOPERATIVE SPECTRUM SENSING
    R.Kavitha, Dr.V.Saravanan ,Arunai Engineering College, Thiruvannamalai, India
    ABSTRACT
    Cognitive radio (CR) technology is a new way to compensate the spectrum shortage problem of wireless environment. This work explores the implementation of Non Cooperative Spectrum Sensing using blind detection technique to detect whether the primary user (PU) signals is present or not in the tested channel. In this technique, estimated energy of the received signal is compared with threshold (predefined) value by using decision rule. Simulation results show that the probability of detection increases when signal to noise ratio increases and also probability of detection will increases in the decrease of probability of false alarm detection.
  • Design of low leakage Standard cells using gate length biasing in cadence virtuoso and ALU using power gating sleep transistor technique in soc encounter
    Priyanka Mehra and Mrs.P.Ramani ,SRM University,Chennai,India
    ABSTRACT
    The leakage current, power and area have become import parameters in circuit designing as the technology is scaling.Leakage power has become one of the most critical design concerns for the system level chip designer. While lowered supplies and hence lowered threshold voltage) and aggressive clock gating can achieve dynamic power reduction, these techniques increase the leakage power and, therefore, causes its share of total power to increase. The basic gates such as inverter, NAND, and NOR are important elements in digital circuits.Gate length biasing is a method to optimize the design by varying the gate length so as to decrease power dissipation. In the current technology, the leakage power is the major contributor to the total power consumption . Power gating and clock gating are technique which are used to reduce the leakage power by switching off the unused transistors and clock using sleep transistor technique
  • ADAPTIVE MODULATION AND CODING TECHNIQUE FOR EFFECTIVE COMMUNICATION IN LTE-A
    T.Latha and S.Priyadharsini ,Arunai Engineering College, Thiruvannamalai, India
    ABSTRACT
    To cooperate with increasing traffic due to high data rate video transmission revolutionary changes needed in the existing wireless networks. In 5th generation wireless networks mainly concentrate on user requirements of high data rate, speed, and higher coverage area .This paper mainly concentrate on physical layer operations. Modulation and coding operation performed in physical layer. To improve the performance of physical layer new technology introduced such that adaptive modulation and coding (AMC). Depending upon the CQI value modulation and coding adapted automatically so spectral efficiency will improved more and avoiding the wastage of resources but in the previous case fixed modulation scheme only used. Channel quality indicator (CQI) Measured based upon noise, receiver performance, interference, channel condition.

AIRCC Library



Courtesy

Technically Sponsored by