Guntūr, India
Guntūr, India

Time filter

Source Type

Venkatalakshmi K.,UCET
Asian Journal of Information Technology | Year: 2016

Image segmentation is a complex task which helps us to extract information for analysis a digital image. Millions of methods are available for image segmentation. Out of that image thresholding is a simple, efficient and frequently adopted method for image segmentation. Thresholding basically divide a digital image into two regions; foreground and background based on the intensity value of the pixels. The key point in image thresholding is on the optimum value of threshold of the digital image. It is an important and crucial task to select the optimum threshold. A false choice of threshold will lead to poor results in image segmentation. Generally optimization algorithms are used to select the optimum threshold value. Artificial Bee Colony (ABC) algorithm is one of the optimization algorithms which are the replica of natural behaviour of honey bees to find abundant nectar amount. This study describes an approach to segment an 8 bit human lung image using artificial bee colony algorithm based thresholding method. The proposed method proves that the uniformity factor in the image segmentation is good relative to other conventional methods. © Medwell Journals, 2016.

Suresh G.V.,UCET | Suresh Babu K.,VVIT | Karunakar K.,UCET | Vijaya Kumar N.,UCET
Journal of Theoretical and Applied Information Technology | Year: 2011

Data uncertainty is common in real-world applications due to various causes, including imprecise measurement, network latency, out-dated sources and sampling errors. These kinds of uncertainty have to be handled cautiously, or else the mining results could be unreliable or even wrong. We propose that when data mining is performed on uncertain data, data uncertainty has to be considered in order to obtain high quality data mining results. We present a Probabilistic Neural Network model which is suitable for classification problems. This model constitutes an adaptation of the classical RBF network where the outputs represent the class conditional distributions. Since the network outputs correspond to probability densities functions, training process is treated as maximum likelihood problem and an Expectation- Maximization (EM) algorithm is proposed for adjusting the network parameters. Experimental results show that proposed model exhibits superior classification performance on uncertain data. © 2005-2011 JATIT & LLS.

Kadhiravan D.,UCET | Gunavathy,UCET | Bharathi,UCET | Sumitha,UCET
International Journal of Applied Engineering Research | Year: 2014

This paper examines the problem of assigning channels to mobile users in the frequency domain. Such assignments must be made so that interference between calls is minimized, while demands for channels are satisfied. Recently, various approaches have been proposed for channel allocation based on intelligent techniques such as multi-agent technology and genetic algorithm. These approaches constitute heuristic solutions to resource management problem. This paper presents Ant colony optimization Algorithm, for minimizing the use of available channels in wireless communication networks, which is subjected to a number of constraints. This algorithm is proposed to reuse the available channels more efficiently by updating the pheromone. © Research India Publications.

Arunprasath S.,UCET | Chandrasekar S.,UCET | Venkatalakshmi K.,UCET | MercyShalinie S.,TCE
2010 IEEE International Conference on Communication Control and Computing Technologies, ICCCCT 2010 | Year: 2010

An attempt has been made in the paper to find globally optimal cluster centers for remote-sensed images with the proposed Rapid Genetic k-Means algorithm. The idea is to avoid the expensive crossover or fitness to produce valid clusters in pure GA and to improve the convergence time. The drawback of using pure GA in the problem is the usage of an expensive crossover or fitness to produce valid clusters (Non-empty clusters). To circumvent the disadvantage of GA, hybridization of GA with k-Means as Genetic k-Means (GKA) is already proposed[GKA, Fast,Flash]. The Genetic k-Means Algorithm always finds the globally optimal cluster centers but the drawback is the usage of an expensive fitness function which involves σ truncation. The Rapid GKA alleviates the problem by using a simple fitness function with an incremental factor. A k-Means operator, one-step of k-Means algorithm, used in GKA as a search operator is adopted in this paper. In Rapid GKA the mutation involves less computation than the mutation involved in GKA and Fast GKA(FGKA). In order to avoid the invalid clusters formed during the iterations the empty clusters are converted into singleton cluster by adding a randomly selected data item until none of the cluster is empty. The results show that the proposed algorithm converges to the global optimum in fewer numbers of generations than conventional GA and also found to consume less computational complexity than GKA and FGKA. It proves to be an effective clustering algorithm for remote sensed images. ©2010 IEEE.

Karthikeyan M.,UCET | Venkatalakshmi K.,UCET
2012 3rd International Conference on Computing, Communication and Networking Technologies, ICCCNT 2012 | Year: 2012

Efficient consumption of energy of sensor node in Wireless Sensor Networks (WSN's) is one of the noticeable challenges nowadays. We can prolong the lifetime of WSN by well-organized clustering of nodes. In this work we suggest PSO incorporated cuckoo search optimization algorithm for clustering in energy aware way and compared it with cuckoo search algorithm. Through efficient clustering of WSN we are reducing the total communication distance as well as providing more probability for higher energy node to be cluster head. The proposed algorithm provides improved lifetime than LEACH, SEP and cuckoo search algorithms. © 2012 IEEE.

Jegatha Deborah L.,UCET | Baskaran R.,Anna University | Kannan A.,Anna University | Vijayakumar P.,UCET
Malaysian Journal of Computer Science | Year: 2013

Performances of the students in learning a programming course is not same, since learning to program is greatly influenced by two dominating factors namely self-efficacy and mental efforts. Prior research efforts have shown that high self-efficacy can have an increased effect of being a trained programmer, especially in an intelligent agent based pair programming system. The main objective of this work is to increase the self-efficacy of the students by providing prior-learning experiences. This experience is facilitated by recommendation agents that provide suitable E-Learning programming course contents based on identifying their individual learning styles which can be used as a factor of prior self-learning computing experience. This helps in increasing the programming abilities when learning in an agent-based pair programming environment subsequently. Moreover, the proposed system analyzes the educational effects of the students learning using pair programming agents based on increased self-efficacy.

Ali S.H.,French Atomic Energy Commission | Ghani U.,UET | Latif A.,UCET | Ijaz N.,UET | Pasha G.A.,UET
Life Science Journal | Year: 2013

The experiments in an open-channel flume with modeled vegetated weir-like structures have been used to understand how the flow is affected by them. Laser Doppler velocimeter measurements of the water flow velocities over trapezoidal vegetated and non-vegetated weir-like structures (dike, groyne) have been made. The measurements extended to a distance of 20 weir heights downstream of the weir crest, it also included the flow separation zone behind the weir crest. The Reynolds's number was of the order of 104 in the flume, and imperfect flow conditions with Froude number<0.4 above the weir crest were considered. Two discharge values were considered in the experimental work. A comparison between the flow characteristics of vegetated and non vegetated weir-like structure was made. The variables investigated included longitudinal and vertical velocity components. Reynolds shearing stresses have also been investigated. The measured mean and turbulent velocities provided more detailed insight about the flow behind vegetated weirs. Strong vortices and turbulent intensities in region especially downstream of vegetated weir crest showed that the flow in the region near bed (on downstream slope of weir the recirculation zone is the main contributor) and at the top of the modeled vegetation is very unstable and leads to the formation of the coherent structures and it is the area of significant mass and momentum exchange. The results indicated that regaining of the logarithmic velocity profile behind the vegetated weir-like structures are delayed due to the presence of vegetation.

Ali S.,French Atomic Energy Commission | Ghani U.,UET Taxila | Latif A.,UCET
Life Science Journal | Year: 2013

The present paper presents results from an experimental work in an open channel flow. The open channel contains a weir-like obstruction with different leeward slopes. Two discharge values have been used under subcritical flow conditions. The objective of the present study is to investigate the flow behavior behind a vegetated obstacle. The characteristics explored included the turbulent kinetic energy and recirculation zone behind the vegetated obstacles. It was observed from this work that the TKE has higher values in recirculation regions. On the other hand it was also found to be of high intensity in the vegetated zones of the flow. However TKE was maximum and positive close to the bed at a section at the end of the weir crest and it was negative below the top of the vegetation dowels. As far as recirculation region was concerned, it was observed that the vegetation had no effect on the recirculation zone behind the vegetated weir. In case of weir with mild downstream slope (1:7), the flow separation zone vanished and the energy head loss in this case decreased due to the decrease in form drag of the weir.

Ghani U.,UET Taxila | Ali S.,French Atomic Energy Commission | Latif A.,UCET
Life Science Journal | Year: 2013

This paper presents numerical modeling of an open channel with heterogeneous bed strips. The bed formation comprises of checker-board like configuration. At any location along the channel, one half of the bed width was rough and rest half was smooth. The rough side was comprised of gravels. An attempt has been made to investigate how many patches of bed configuration will be required so that flow investigation can be made under periodic boundary condition. Simulation over a length of four patches with periodic boundary condition at inlet/outlet was performed for this purpose. A three dimensional Computational Fluid Dynamics (CFD) numerical model FLUENT was used in this work. The results have been presented in the form of primary velocity contours overlaid by the secondary velocity vectors. These results were calculated at different critical locations along the patches to investigate the flow development. It was observed that the flow patterns in the third and fourth patches are of the same style as that observed in the initial two patches i.e. the developing velocity contours and secondary velocity vectors happened twice in four patches. It can therefore be concluded that two patches are sufficient for any kind of numerical study in these types of bed formations under periodic boundary condition.

Loading UCET collaborators
Loading UCET collaborators