K. N. Balasubramanya Murthy

A smart automatic thumbnail cropping based on attention driven regions of interest extraction

Most of the Content-Based Image Retrieval (CBIR) solutions treat each image as a whole. But the fact that often a user will be searching for a part of the image i.e., a region in the image with obvious semantic meaning as opposed to the entire picture has led to view an image as a set of Regions of Interest (ROIs) rather than viewing it as a whole. To provide a quick scanning of large number of images thumbnail representation of the images are used in most of the image retrieval and browsing systems. Thumbnails generated by shrinking the original image often render the ROIs illegible. In this paper, we present an intelligent automatic cropping technique, prior to shrinking, based on an efficient, unsupervised visual attention driven ROI detection framework that can make ROIs of an image more recognizable. Experiments show that the thumbnails resulting from the proposed technique will efficiently increase the CBIR performance and is a valid approach to carry further research.

DBAM: Novel User Based Bandwidth Allocation Mechanism in WiMAX

IEEE 802.16e classifies packets into 5 different service classes, namely, UGS, rtPS, ertPS, nrtPS and BE. Bandwidth is allocated to SS based on the need of the service class. In this paper we propose a user based bandwidth allocation mechanism called Differentiated Bandwidth Allocation Mechanism (DBAM). WiMAX users shall be divided into 3 different priority levels as high-priority users, regular users and low-priority users. Bandwidth shall be allocated to the service flows of these service class based on the priority of the users. This method of bandwidth allocation provides a differentiated QoS to users for the same service class. Simulation results revel that when the number of Subscriber Stations (SS) is 10, high-priority and Regular users are able to transmit 25% more ertPS data compared to low-priority users and when the number of SS is 14, high-priority user transmits 30% more data compared to regular user.

Towards Realizing the Secured Multilateral Co-Operative Computing Architectural Framework

Innovative approaches for securing the computing systems have intense inference for our understanding of technological, societal, economical, and political phenomena in making guiding principles, policies, rules and security implementations. The model presented here is based on cooperative and collaborative multilateral relationship among the shared business community partners. The sole responsibility for the securing and maintenance of their own virtualized dedicated boxes with jointly hosted data centers distributed geographically is equally, vested with service consumers and vendors. Inspired by multilateral techniques used in army and health care applications, this model is a conceptual and empirical tool aimed, rather depicting a particular set of observed situations or making predictions. It is aimed is to develop our understanding of the fundamental mechanisms driving the security implementations of existing methods and provide the MCF - multilateral collaborative co-operative framework. This MCF demonstrates an interrelated multilayered virtualized architectural framework for computing utility using virtualization platform. We try to demonstrate qualitatively and empirically the proposed architecture works well for a wide range of workloads and devices belonging to multi tenants with varied security needs. This work is compared with currently existing virtualization platform security framework and avail the novelty of the proposed ontology and framework.

Expediting Automated Face Recognition Using the Novel ORB 2 -IPR Framework

Face Recognition (FR) is at the forefront of distinctly unresolved challenges in the domain of Computer Vision, due to the sharp accuracy and performance drops it undergoes, when there are pronounced variations in parameters such as illumination, pose, background clutter and so on between the input and database faces. In this paper, we attempt to expedite the performance of automated FR with real-time images, using a novel framework called ORB2-IPR (ORB based Bag of Interest Points using RANSAC), which exhaustively learns a vocabulary of highly discriminative facial interest points from the facial database images (which can be referred to, and compared directly, instead of following the conventional time-intensive approach of comparing a given input face with each database face separately) by employing the cost-effective ORB (Oriented Fast Rotated Brief) descriptor (instead of the commonly employed SIFT and SURF descriptors), followed by the application of RANSAC (Random Sample Consensus) as a post-processing step to remove noise in the form of outliers, in order to improve the accuracy of the system. We will conclusively demonstrate that our technique is capable of rendering superior performance than the state-of-the-art methodologies using extensive mathematical arguments and by carrying out ample experimentations on the benchmark ORL, Face 95 and LFW databases.

A New Approach for Analyzing the Performance of the WiMAX Networks based on QoS Traffic Prediction Routing Protocol using Gene Expression Programming

WiMAX is one of the broadband wireless access technologies, which provides the efficient QoS to the large number of users. The multimedia applications such as real time and non-real time services are gaining importance in the WiMAX network. To support such applications, there is a need to propose an efficient QoS traffic prediction routing protocol for the WiMAX networks. To address this, the authors are using Gene Expression Programming technique. They have generated datasets for CBR based traffic and file transfer applications. Here, they focus to develop the mathematical expressions for throughput of the network in terms of bandwidth, average end-to-end delay and average jitter for CBR based traffic and file transfer applications, so that they can analyze and predict the QoS traffic of the network. The simulation results show that the model values and the target values match with better approximation. Further, sensitivity analysis has been carried out for both CBR based traffic and file transfer applications.

An analysis and comparison of different routing algorithms in WiMAX networks

World-wide interoperability for Microwave Access (WiMAX) is a wireless metropolitan area networks (WMANs). Routing and multimedia application plays an important role in the WiMAX networks. In this study, the main focus is on analyzing and comparing the different routing algorithms such as AODV, DYMO, DSR, Bellman-Ford and Fish-Eye for the WiMAX networks. We have considered few QoS parameters such as total data received, average end-to-end delay, average jitter and throughput. Here, we have considered the CBR application. From the obtained result DSR routing algorithm outperforms high throughput and total data received. Bellman-Ford and Fish-Eye provides less average end-to-end delay. Bellman-Ford routing algorithm also provides less average jitter.

Face Recognition Using the Novel Fuzzy-GIST Mechanism

Face Recognition (FR) is one of the most thriving fields of contemporary research, and despite its universal application in authentication and verification systems, ensuring its effectiveness in unconstrained scenarios has predominantly remained an on-going challenge in Computer Vision, because FR systems experience considerable loss in performance, when there exists significant variation between the test and database faces in terms of attributes such as Pose, Camera Angle, Illumination and so on. The potency of FR systems markedly declines in the presence of noise in a given face and furthermore, the performance is also determined to a large degree by the Feature Extraction technique that is employed. Hence in this paper, we propose a novel mechanism known as Fuzzy-GIST, that can proficiently perform FR by adeptly handling real-time images (which contain the aforementioned unconstrained attributes) in low-powered portable devices by employing Fuzzy Filters to eliminate extraneous noise in the facial image, prior to feature extraction using the computationally less demanding GIST descriptor. Backed by relevant mathematical defense, we will establish the efficacy of our proposed system by conducting detailed experimentations on the ORL and IIT-K databases.

Dominant feature based convolutional neural network for faces in videos

Standard face recognition modules are fabricated for general-purpose applications while few have been designed with speed in mind. This paper proposes an efficient architecture for face recognition in which two self-contained Convolutional Neural Networks (CNNs) are used to detect and recognize faces in regions containing a dense grouping of Features from Accelerated Segment Test (FAST). This configuration proves to be practical for videos as it is selective in its analysis of an input frame. City surveillance and public safety is a critical issue in smart cities and the deployment of Smart Video Surveillance systems is the need of the hour. Typically, the problem at hand will be person identification which is the association of a biometric trait with a particular human being. FAST key points can be generated and analyzed in near real-time and that data can be used to extract and process faces in the background. The CNNs were trained using a combination of datasets of labelled faces, videos and trivial objects. The results obtained upon analyzing the performance of the system on the ChokePoint dataset proved very insightful. This configuration leads to a very effective face recognition system.

Location area planning problem in WiMAX networks using nature inspired techniques: performance study

Worldwide Interoperability for Microwave Access (WiMAX) is a broadband wireless technology that provides an efficient service to mobile stations (MS). Whenever there is a need to establish communication and provide service to MSs, the network has to track the location of the MSs through the base station. Tracking the location of the MSs is a very difficult and complex problem in the WiMAX network. This paper discusses a location area planning problem, which can be solved by partitioning the WiMAX network into location areas, so that the cost per call arrival is minimum. Finding the optimal number of location areas and the corresponding configuration of the partitioned network is an NP-complete problem. In this study, we use nature inspired techniques, namely, genetic algorithm (GA), artificial bee colony (ABC) and artificial immune system (AIS) to find an optimal solution to the location area planning problem. The performance is analysed and compared for all these nature inspired techniques. In this study, we also compare these techniques to gauge their suitability for solving the location area planning problem. From the obtained results, we conclude that ABC gives better optimal solution and AIS takes less computational time to locate the optimal solution.

RISA: Rotation Illumination Scale and Affine Invariant Face Recognition

Face Recognition (FR) has been on the forefront of research efforts for the past two decades. In spite of considerable strides, it still suffers from the curse of false matches in the presence of variations in terms of parameters such as affine, scale, rotation and illumination. Since, real world images inherently consists of such variations, an effective FR system, should handle such variations deftly. Hence, in this paper, we propose a robust, yet simple and cost effective technique for overcoming some of the aforementioned challenges. The first stage of the proposed system deals with illumination variations by performing logarithm transform on the input face images. Further, the Non-subsampled Contourlet Transform (NSCT) is used to decompose the logarithm transformed facial images into low frequency and high frequency components. Subsequently, histogram equalization is carried out on the low frequency components. Finally, we employ Affine Scale Invariant Feature Transform (ASIFT) to find corresponding points that are translation and scale invariant. We will demonstrate by carrying out extensive experimentations on the benchmark datasets: ORL, Grimace, Face95 and Yale, that the proposed technique is more robust and yields comparable efficacy to most of the contemporary approaches.