Time filter

Source Type

Hyderabad, India

Kavitha Ch.,Alstom | Rao M.B.,Alstom | Rao B.P.,JNTUK | Govardhan A.,JNTUHCE
Communications in Computer and Information Science | Year: 2011

In this paper we propose a new and efficient technique to retrieve images based on multi-resolution color and texture features of image sub-blocks. Firstly the image is divided into sub blocks of equal size in two resolutions. The size of the sub-block is fixed in two resolutions. Color of each sub-block is extracted by quantifying the HSV color space into non-equal intervals and the color feature is represented by cumulative histogram. Similarly the texture of the sub-block is extracted based on edge oriented gray tone spatial dependency matrix (GTSDM). An integrated matching scheme based on Most Similar minimum cost (MSMC) principle is used to compare the query and target image. The adjacency matrix of a bipartite graph is formed using the sub-blocks of query and target image. This matrix is used for matching the images. The experimental results show that the proposed method has achieved highest retrieval performance. © 2011 Springer-Verlag. Source

Rao M.B.,Alstom | Kavitha Ch.,Alstom | Rao B.P.,JNTUK | Govardhan A.,JNTUHCE
Communications in Computer and Information Science | Year: 2011

There is a great need of developing an efficient content based image retrieval system because of the availability of large image databases. A new image retrieval technique to retrieve the images using four features called dominant color (DC), scan pattern co-occurrence matrix of a motif (SPCMM), scan pattern internal pixel difference (SPIPD) and shape is proposed. Shape information is captured in terms of edge images computed using Gradient Vector Flow fields. Invariant moments are then used to record the shape features. Experimental results show that the proposed image retrieval is more efficient in retrieving the user- interested images. © 2011 Springer-Verlag. Source

Naik P.P.S.,JNTUK | Gopal T.V.,JNTUHCE
2015 International Conference on Communication and Signal Processing, ICCSP 2015 | Year: 2015

Segmentation is a process of partitioning the image into several objects. It plays a vital role in many fields such as satellite, remote sensing, object identification, face tracking and most importantly medical applications. Here in this paper, we here supposed to propose a novel image segmentation using iterative partitioning mean shift clustering algorithm, which overcomes the drawbacks of conventional clustering algorithms and provides a good segmented images. Simulation performance shows that the proposed scheme has performed superior to the existing clustering methods. © 2015 IEEE. Source

Fatima S.K.,Jawaharlal Nehru Technological University | Sreenivasa Rao D.,JNTUHCE
International Journal of Applied Engineering Research | Year: 2015

Nowadays Mobile Adhoc Network (MANET) is becoming more popular due to its mobile and ease of deployment nature. However due to wireless and dynamic nature of network topology, it make them more exposed to various types of attacks. The main issue is to assure secure network services. In order to overcome this issue, a Secure Cluster based Architecture for MANET with Threshold Signature and certificate Revocation is proposed. In this technique, a secure cluster is formed based on the trust value. The node with high trust value is considered as the Cluster Head. In order to increase the security, the selected Cluster Head CH is verified by using Threshold Signature. Also, a certification revocation technique is implemented to stop the participation of any attackers in further activities. © Research India Publications. Source

Madhavi K.R.,JNTUA | Babu A.V.,JNTUHCE | Rao A.A.,JNTUACE | Raju S.V.N.,JNTUH College of Engineering
ACM International Conference Proceeding Series | Year: 2012

Identification of useful clusters in large datasets has attracted considerable interest in clustering process. Since data in the World Wide Web is increasing exponentially that affects on clustering accuracy and decision making, change in the concept between every cluster occurs named concept drift. This newly added time based data must be assigned/labeled into generated clusters at our hand. To say that the data labeling was performed well, the clusters must be efficient. Selecting initial cluster center (centroid) is the key factor that has high affection in generating effective clusters. The existing clustering methods selects centroid randomly. Different centroids results in different clusters. To avoid this random selection, we are proposing methods in selecting the centroid by analyzing the properties of data since the data with different properties exists in real world. Our previous work was concentrated in the identification centroid for the functions of single variable and two variable functions. This paper proposes methods in finding optimal cluster centroid for the multi-variable functions and then apply any existing clustering algorithm to generate clusters by using suitable distance measure. © 2012 ACM. Source

Discover hidden collaborations