S. Indu

Optimal sensor placement for surveillance of large spaces

Visual sensor network design facilitates applications such as intelligent rooms, video surveillance, automatic multi-camera tracking, activity recognition etc. These applications require an efficient visual sensor layout which provides a minimum level of image quality or image resolution. This paper addresses the practical problem of optimally placing the multiple PTZ cameras to ensure maximum coverage of user defined priority areas with optimum values of parameters like pan, tilt, zoom and the locations of the cameras. The proposed algorithm works offline and does not require camera calibration. We mapped this problem as an optimization problem using Genetic Algorithm, by defining, coverage matrix as a set of sensor parameters and the space model parameters like priority areas, obstacles and feasible locations of the sensors, and by modelling discrete spaces using probabilistic frame work. We minimized the probability of occlusion due to randomly moving objects by covering each priority area using multiple cameras. The proposed method will be applicable for surveillance of large spaces with discrete priority areas like a hall with more than one entrance or many events happening at different locations in a hall eg.Casino. As we are optimizing the parameters like pan, tilt, zoom and even the locations of the cameras, the coverage provided by this approach will assure good resolution, which improves the QOS of the visual sensor network.

Design of high speed Vedic multiplier using multiplexer based adder

Real time applications such as controlling environmental conditions demand quick response of the processor for processing the acquired signals. Multiplier is an important feature of signal processing. Vedic Mathematics provides principles of high speed multiplication. Motivated by this, a high speed Vedic multiplier using multiplexer based adder is proposed in this paper. Proposed design is simulated using ModelSim and synthesized using Xilinx ISE 14.7. When compared with existing Vedic multipliers, proposed design shows a significant improvement in speed.

Fish Species Classification Using Graph Embedding Discriminant Analysis

Fish Species classification is of great utility to marine biologists for the understanding of underwater ecology and fish behavior as well as to keep a log of endangered species, which assists in fisheries management. Traditional methods being either tedious or too computationally intensive lead to the requirement of an automated method of analysis and counting. Generally, the classification of underwater images is challenging due to difficulties in camera calibrations, which lead to distortion, background noise, occlusion and image quality to name a few. This paper proposes a novel method based on an improved image-set matching approach, which makes use of Graph-Embedding Discriminant Analysis. In contrast to the state-of-the-art methods, which operate on single input images, our method makes, use of explicit image set matching which renders it robust computationally. In comparison with the previous classification methods, this method results in considerable discrimination accuracy improvements.

Big data analysis using Hadoop cluster

Industries keep a check on all statistics of their business and process this data using various data mining techniques to measure profit trends, revenue, growing markets and interesting opportunities to invest. These statistical records keep on increasing and increase very fast. Unfortunately, as the data grows it becomes a tedious task to process such a large data set and extract meaningful information. Also if the data generated is in various formats, its processing possesses new challenges. Owing to its size, big data is stored in Hadoop Distributed File System (HDFS). In this standard architecture, all the Data Nodes function parallel but functioning of a single Data Node is still in sequential fashion. This paper proposes to execute tasks assigned to a single Data Node in parallel instead of executing them sequentially. We propose to use a bunch of streaming multi-processors (SMs) for each single Data Node. An SM can have various processors and memory and all SMs run in parallel and independently. We process big data which may be coming from different sources in different formats to run parallelly on a Hadoop cluster, use the proposed technique and yield desired results efficiently. We have applied proposed methodology to the raw data of an industrial firm, for doing intelligent business, with a final objective of finding profit generated for the firm and its trends throughout a year. We have done analysis over a yearlong data as trends generally repeat after a year.

Digitization of Historic Inscription Images using Cumulants based Simultaneous Blind Source Extraction

In this paper a novel method to address the problem of enhancement and binarization of historic inscription images is presented. Inscription images in general have no distinction between the text layer and background layer due to absence of color difference and possess highly correlated signals and noise. The proposed technique provides a suitable method to separate the text layer from the historic inscription images by considering the problem as blind source separation which aims to calculate the independent components from a linear mixture of source signals, by maximizing a contrast function based on higher order cumulants. Further, the results are compared with existing ICA based techniques like NGFICA and Fast-ICA.

Road traffic model using distributed camera network

Traffic monitoring/prediction using a distributed camera network is presented in this paper. The activities on each road link are monitored and features are derived to identify the pattern. Then it is learnt, classified, predicted and communicated to neighboring road links. We used GMM-EM based classification and HMM based prediction. Optimum path is determined by assigning proportional weights to the predicted states of the connected road links. The proposed method is neither based on tracking nor on vehicle detection. Apart from this the method is flexible, adaptive, robust and computationally light. Unlike the existing methods it does not assume or draws analogies of traffic moving as particles, neither does it impose restriction on road conditions or road tributaries and distributaries. The model is validated using traffic simulator and tested on real road network.

Novel method for manuscript and inscription text extraction

Computer vision involves image processing which has the use of various algorithms for text extraction as an integral part. This paper delves into the use of filtering based thresholding on HSV plane which serves as a common tool for text extraction in manuscripts as well as inscriptions. The basic and simple technique proposed has provided positive results for various images as compared to the other commonly used complex algorithms. The paper also involves a comparison with various other algorithms which is the nick, niblack, Messalodi, Otsu, Gllavata, Kim, Lee, Color Clustering Technique and FastICA helping us understand the advantages that this algorithm presents.

Secure route discovery in AODV in presence of blackhole attack

In recent years, there has been tremendous increase in mobile adhoc networks applications ranging from military and rescue operations to collaborative and distributed computing, therefore secure data transmission has become one of the critical issue in MANET. A mobile adhoc network is genrally established without relying on centralized and dedicated servers. Attackers exploit the loopholes of route discovery process to carry out their malicious intent as it is an inevitable process in reactive protocols. Blackhole is one such popular attack that sends forged routing information to fool source node and drops all the data packets after introducing itself in the route between source node to destination. In this paper, damage caused to AODV in presence of blackhole node has been evaluated and solution to defend against blackhole attack has been proposed and simulated on Ns-2 to prove its efficiency and reliability. The packet processing technique of normal AODV is enhanced to detect routing misbehavior and alert other nodes using default AODV control message, HELLO messages to reduce additional overhead.

Enhancement and Retrieval of Historic Inscription Images

In this paper we have presented a technique for enhancement and retrieval of historic inscription images. Inscription images in general have no distinction between the text layer and background layer due to absence of color difference and possess highly correlated signals and noise; pertaining to which retrieval of such images using search based on feature matching returns inaccurate results. Hence, there is a need to first enhance the readability and then binarize the images to create a digital database for retrieval. Our technique provides a suitable method for the same, by separating the text layer from the non-text layer using the proposed cumulants based Blind Source Extraction(BSE) method, and store them in a digital library with their corresponding historic information. These images are retrieved from database using image search based on Bag-of-Words(BoW) method.

Camera and Light Source Placement: A Multi-Objective Approach

Optimal placement of visual sensors along with good lighting conditions is indispensable for the successful execution of surveillance applications. Limited field-of-view, depth-of-field, occlusion due to presence of different objects in the scene form the major constraints for visual sensor placement. While over/under exposed objects, shadowing and light rays directly incident on the camera lens are some of the constraints for light source placement. Because of the nature of the constraints and complexity of the problem, the placement problem is considered to be a multi-objective global optimization problem. The paper outlines the camera and light source location optimization problem with multiple objective functions. Multi-Objective Genetic Algorithm is used to maximize the camera coverage with optimum illumination of the sensing space.