Browsing by Author "Subashini, P"

Now showing 1 - 20 of 85
  • No Thumbnail Available
    Item
    Analysis of Different Propagation Model for IPSec-LANMAR Routing Protocol to Secure Network Layer for MANET in Emergency Area Environment
    (2011) Subashini, P
    Mobile ad hoc networks (MANETs) consist of a collection of wireless mobile nodes, which dynamically exchange data among themselves without the reliance of a fixed base station. It have potential use in a wide variety of disparate situations such as responses to hurricane, tsunami, earthquake, emergency relief, terrorism and military operations. Inthiswork, investigations on the behavior of IPSec-LANMAR routing protocol with propagation models that take into account two main characteristics of the wireless channel - path loss and shadowing is presented. The choice of radio propagation models with IPSec-LANMAR also has a strong impact on the performance of a protocol because the propagation model determines the number of lodes within one collision domain, an important input for contention and interference. This, in turn, has a direct effect on a node’s ability to transmit a packet to another node and it provide security services for both routing information and data message at network layer. The simulation result shows that Free space propagation model with IPSec-LANMAR routing protocol outperforms compared to Two ray propagation model and Shadowing model in emergency area environment and the experiments are carried out using the simulator Qualnet version 5 also the performance of propagation models with IPSec-LANMAR for MANETs is compared based on different parameter metrics.
  • No Thumbnail Available
    Item
    Android Based Ship Recognition System
    (2013-05) Poovizhi, M; Subashini, P
  • No Thumbnail Available
    Item
    Android Based Ship Tracking System
    (2013-05) Saranya, S; Subashini, P
  • No Thumbnail Available
    Item
    ANODR-ECC Key Management protocol with TELNET to secure Application and Network layer for Mobile Adhoc Networks
    (2012) Padmavathi, G; Subashini, P
    A mobile ad hoc network (MANETs) is a self-organizing network that consists of mobile nodes that are connected through wireless media. A number of unique features, such as lack o f infrastructural or central administrative supports, dynamic network topologies, open communication channels, and limited device capabilities and bandwidths, have made secure, reliable and efficient routing operations in MANET a challenging task. The ultimate goal o f the security solutions for MANET is to provide security services, such as authentication, confidentiality, integrity, anonymity, and availability to mobile users. To achieve the goals, the security solution need for entire protocol stack.. The proposed protocol ANODRECC with Telnet provide application layer security and it ensures route anonymity and location privacy and is robust against eavesdropping attack. For route anonymity, it prevents strong adversaries from tracing a packet flow back 'to its source or destination; for location privacy, it ensures that adversaries cannot discover the real identities of local transmitters. The simulation is done using network simulator qualnet 5.0 for different number o f mobile nodes. The proposed model has exposed improved results in terms o f Average throughput. Average end to end delay. Average packet delivery ratio and Average jitter.
  • No Thumbnail Available
    Item
  • No Thumbnail Available
    Item
    Automatic Identification of Algal Community from Microscopic Images
    (2013) Subashini, P
    A good understanding of the population dynamics of algal communities is crucial in several ecological and pollution studies of freshwater and oceanic systems. This paper reviews the subsequent introduction to the automatic identification of the algal communities using image processing techniques from microscope images. The diverse techniques of image preprocessing, segmentation, feature extraction and recognition are considered one by one and their parameters are summarized. Automatic identification and classification of algal community are very difficult due to various factors such as change in size and shape with climatic changes, various growth periods, and the presence of other microbes. Therefore, the significance, uniqueness, and various approaches are discussed and the analyses image processing methods are evaluated. Algal identification and associated problems in water organisms have been projected as ^..allenges in image processing application. Various image processing approaches based on textures, shapes, and an object boundary, as well as some segmentation methods like, edge detection and color segmentations, are highlighted. Finally, artificial neural networks and some machine learning algorithms were used to classify and identifying the algae. Further, some of the benefits and drawbacks of schemes are examined.
  • No Thumbnail Available
    Item
    Automatic Noise Identification in Images using Statistical Features
    (2011) Subashini, P
    Noises are unwanted information which will be normally present in an image. It should be removed in such a way that the important information present in an image is preserved. De-noising an image is very active research area in image processing. There are several algorithms for de-noise but each algorithm has its own assumptions, advantages and limitations. The method proposed in this paper uses a simple pattern classification for noise identification. The significant idea is to obtain the noise samples and to extract their statistical feature for identifying the noise type. Simple image filters are used to get the noise samples and noise identification is achieved by using few statistical features. The method is capable of accurately classifying the type of noise besides the type of images used.
  • No Thumbnail Available
    Item
    A Bayes fusion method based ensemble classification approach for Brown cloud application
    (2014) Subashini, P
    Classification is a recurrent task o f determining a target function that maps each attribute set to one o f the predefined class labels. Ensemble fusion is one o f the suitable classifier model fusion techniques which combine the multiple classifiers to perform high classification accuracy than individual classifiers. The main objective o f this paper is to combine base classifiers using ensemble fusion methods namely Decision Template, Dempster- Shafer and Bayes to compare the accuracy o f the each fusion methods on the brown cloud dataset. The base classifiers like KNN, MLP and SVM have been considered in ensemble classification in which each classifier with fo u r different function parameters. From the experimental study it is proved, that the Bayes fusion method performs better classification accuracy o f 95% than Decision Template o f 80%, Dempster-Shaferof 85%, in a Brown Cloud image dataset.
  • No Thumbnail Available
    Item
    Bidding System With Multibrand And Product Selection
    (2015-03) Karthiga, G; Subashini, P
  • No Thumbnail Available
    Item
    Bio-Inspired Routing Protocol for Mobile AD HOC Networks
    (2012-04) Sivasankari, N; Subashini, P
  • No Thumbnail Available
    Item
    Building and AI Snake Identifier Using Fastai
    (2021-05) Thenmozhi, M; Subashini, P
  • No Thumbnail Available
    Item
    CCMP-AES Model with SNAuth-SPMAODV Routing Potocol to Secure Link and Network Layer for Mobile Adhoc Networks in Military Scenario
    (2012) Subashini, P
    Mobile Adhoc network is a special kind o f wireless networks. It is a collection o f mobile nodes without having aid of established infrastructure. Mobile Adhoc network are vulnerable to attacks compared to wired networks due to limited physical security, volatile network topologies, power-constrained operations, intrinsic requirement of mutual trust among all nodes. During deployment, security emerges as a central requirement due to many attacks that affects the performance o f the ad hoc networks. Particularly Denial o f Service attack is one such severe attack against ad hoc routing protocols which is a challenging one to defend against in military communication environments. The proposed model combines SNAuth-SPMAODV with CCMP-AES mode to defend against Denial o f Service attack and it also provides confidentiality and authentication of packets in both routing and link layers of MANET. The primary focus o f this work is to provide security mechanisms while transmitting data frames in a node to node manner. The security protocol CCMP-AES working in data link layer keeps data frame from eavesdropping, interception, alteration, or dropping from unauthorized party along the route from the source to the destination. The simulation is done for different number o f mobile nodes using network simulator Qualnet 5.0. The proposed model has shown better results in terms of total bytes received, packet delivery ratio, throughput. End to End delay and Average jitter.
  • No Thumbnail Available
    Item
    Cell Nuclei segmentation in Pap smear images using Optimized Binarization technique with Adaptive Wiener Filter
    (2013) Subashini, P
    Cervical cancer is the second leading reason o f death among women in India. The most common screening technique is Pap smear test which is used to detect abnormal growth o f cervical cells at an early stage. The accuracy rate o f cervical cancer diagnosis using Pap smear images depends on the segmentation o f the cell nuclei. This paper proposes the optimized binarizatidh technique with Adaptive Wiener Filter (AWF) for the cell nuclei segmentation fn ^ ^o key steps. First, the Adaptive Wiener Filter is used for the noise removal as well as to preserve the details o f the Pap smear images. Followed by, the threshold value is being obtained from the sure shrinkage method. Ant Colony Optimization and Particle swarm optimization for the exact segmentation o f cell nuclei from Pap smear images. Due to the limitations in the segmentation techniques, the loss o f cell nuclei gets a raise which affects the quality and efficiency o f cervical cancer detection. To know about the loss in the number o f cell nuclei during segmentation step, the number o f nuclei is counted from the segmented images and cell count results are compared with each other. From the results, it is found that the Adaptive Wiener Filter in combination with PSO based threshold segmentation performs well in terms o f MSE, cell nuclei count, sensitivity and specificity.
  • No Thumbnail Available
    Item
    Classification of Handwritten Ancient Tamil characters using Complex Extreme Learning Machine
    (2013) Subashini, P
    Classification is the problem o f identifying, to which set o f categories a new observation belongs, on the basis o f a training set whose category membership is known. The process o f handwritten script classification involves extraction o f some defined characteristics calledfeatures to classify an unknown handwritten character into one o f the known classes. Zernike moments and regional features are extracted from the Tamil characters and they are formed as feature vectors. Complex Extreme Learning Machine is used to classify the handwritten ancient Tamil characters. Complex Extreme Learning Machine is trained with feature vectors. From the experimental result it is observed that the classifier when trained by combining Zernike moments with regional features gives a highest classification accuracy o f82.63%.
  • No Thumbnail Available
    Item
  • No Thumbnail Available
    Item
    Decision Based Median Filter using Particle Swarm Optimization for Impulsive Noise
    (2013) Subashini, P
    Infrared image sensors and communication medium often introduce impulse noise in image acquisition and transmission. Most commonly available filters to remove impulsive noises are median fillers with different versions, but the most important drawbacks identified with them are low noise suppression and edge blurring. To preserve the sharp and valuable information present in the image, the filtering algorithms should preserve the information available in them. The proposed work consists o f particle swarm optimization based weight adaptation approach in the design of the filter. The filter weights are adapted and optimized directly to restore a corrupted pixel in a mean square sense. This proposed work results in replacement of noisy pixels by near originals along with its edge direction. The objective parameters used are Peak Signal to Noise Ratio, Mean Absolute Error, and Correlation. This proposed research work is designed at presenting a new filtering framework for impulse noise removal using Particle Swarm Optimization.
  • No Thumbnail Available
    Item
  • No Thumbnail Available
    Item
    Detection and Recognition of Objects in X – Ray Radiography Images
    (2010-05) Yamunavathi, T; Subashini, P
  • No Thumbnail Available
    Item
    DIFFERENTIAL EVOLUTION AND GENETIC ALGORITHM BASED FEATURE SUBSET SELECTION FOR RECOGNITION OF RIVER ICE TYPES
    (2014) Subashini, P
    One of the essential motivations for feature selection is to defeat the curse of dimensionality problem. Feature selection optimization is nothing but generating best feature subset with maximum relevance, which improves the result of classification accuracy in pattern recognition. In this research work, Differential Evolution and Genetic Algorithm, the two population based feature selection methods are compared. First, this paper presents Differential Evolution float number optimizer in the combinatorial optimization problem of feature selection. In order to build the solution generated by the Differential Evolution float-optimizer suitable for feature selection, roulette wheel structure is constructed and supplied with the probabilities of features distribution. To generate the most promising feature set during iterations these probabilities are constructed. Second, Genetic Algorithm minimizes the Joint Conditional Entropy between the input and output variables. Practical results indicate Differential Evolution feature selection method with ten features achieves 93% accuracy when compared with Genetic Algorithm method.
  • No Thumbnail Available
    Item
    DSSS with ISAKMP Key Management Protocol to Secure Physical Layer for Mobile Adhoc Network
    (2012) Padmavathi, G; Subashini, P
    The wireless and dynamic nature of mobile ad hoc networks (MANETs) leaves them more vulnerable to security attacks than their wired counterparts. The nodes act both as routers and as communication end points. This makes the physical layer more prone to security attacks. The MANET physical layer is challenging to DoS attack'and also some passive attacks. The physical layer protocol in MANETs is responsible for bit-level transmission between network nodes. The proposed model combines spread spectrum technology Direct Sequence Spread Spectrum (DSSS) with key management technique ISAKMP to defend against signal jamming denial-of-service attacks in physical layer of MANET.DSSS with ISAKMP is found to be a good security solution even with its known security problems. The simulation is done using network simulator qualnet 5.0 for different number of mobile nodes. The proposed model has shown improved results in terms of Average throughput. Average end to end delay, Average packet delivery ratio, and Average Jitter.