K. Satya Prasad

Multiwavelet Based Texture Features for Content Based Image Retrieval

Content based image Retrieval has become one of the most active research areas in the past few years .CBIR system using multiwavelet based features with high retrieval rate and less computational complexity is proposed in this paper. Multiwavelets offer simultaneous orthogonality, symmetry and short support. This property made it a powerful tool for feature extraction of images in the database. A comparative study is done between multiwavelet feature vectors and wavelet feature vectors for 999 different texture images in a databank. Result analysis show that extracting texture information using multiwavelet provides significantly improved results in terms of retrieval accuracy, computational complexity and storage space of feature vectors as compared to wavelet based CBIR. Euclidean Distance and Canberra Distance are used as similarity measure in the proposed CBIR system.

A new method of target tracking by EKF using bearing and elevation measurements for underwater environment

Underwater moving object detection/tracking is critical in various applications such as exploration of natural undersea resources, acquiring of accurate scientific data to maintain regular surveillance of missions, navigation and tactical surveillance. Real time object detection/tracking which tends to obstacle avoidance is possible with an autonomous underwater vehicle (AUV) fitted with sensor(sonar). To bring these applications into effective use, there is a need to evaluate various solutions for the safe navigation of AUV in the significant underwater environment. Convergence time becomes a problem and plays an increasingly important role in safe navigation of AUV applications. To achieve this, several methods, i.e. Kalman Filter (KF), Extended Kalman Filter (EKF) and Particle Filter (PF) have been investigated, although all these methods have their own limitations. In this paper, a new method has been developed wherein tracking algorithm using EKF has been extended to the Bearing and Elevation only Tracking (BEOT) method. By using Monte Carlo approach, the performance of this algorithm has been analyzed. Consequently, the time of convergence has been calculated and accordingly the results have been plotted. Innovative approach for target tracking for an autonomous underwater vehicle (AUV) has been analyzed.Convergence issues pertaining to the new designed algorithm, wherein EKF has been extended.Mathematical analysis is performed for the updated measurements of bearing and elevation data.By examining the case of single sensor/observer bearing and elevation only tracking (BEOT) problem as the inaccuracies can be handled effectively.

A Robust Watermarking Technique based on Nonsubsampled Contourlet Transform and SVD

The paper proposes a novel robust watermarking technique based on newly introduced Nonsubsampled contourlet transform(NSCT) and singular value decomposition(SVD) for multimedia copyright protection. The NSCT can give the asymptotic optimal representation of the edges and contours in image by virtue of the characteristics of good multi resolution shift invariance and multi directionality. After decomposing the host image into sub bands, we choose the low frequency directional sub band and apply singular value decomposition. The singular values of the original image are then modified by the singular values of nonsubsampled contourlet transformed visual grayscale logo watermark image. This hybrid approach improves the performance of the watermarking technique compared to earlier techniques. Experimental results shows that the hybrid technique is resilient to various linear and non linear filtering ,JPEG compression, JPEG2000 compression, Histogram equalization, Grayscale inversion, Contrast adjustment, gamma correction, alpha mean ,cropping ,Gaussian noise, scaling etc.

Linear Discriminant Analysis F-Ratio for Optimization of TESPAR & MFCC Features for Speaker Recongnition

This paper deals with implementing an efficient optimization technique for designing an Automatic Speaker Recognition (ASR) System, which uses average F-ratio score of TESPAR(Time Encoded Signal Processing And Recognition) and MFCC(Mel frequency Cepstral Coefficients) features, to yield high recognition accuracy even in adverse noisy conditions. A new ranking scheme is also proposed in order to stabilize the rank of features in various noise levels by taking Arithmetic Mean of the F-Ratio scores obtained from various levels of Signal to Noise Ratio (SNR). The result is presented for a Text-Dependent ASR system with 20 speaker database. An RBF (Radial Basis Function) Neural Network is used for Recognition purpose. Also a comparative study has been performed for recognition accuracies of optimized MFCC and TESPAR features and we conclude that new proposed average F-Ratio technique has resulted in better accuracy compared to simple F-ratio in noisy environment and also we came to know that TESPAR features are more redundant compared to MFCC.

A New SVD based Hybrid Color Image Watermarking for Copyright Protection using Contourlet Transform

The authors propose a new hybrid watermarking scheme for copyright protection of color images using contourlet transform and singular value decomposition. The host color image and color watermark images are decomposed into directional sub- bands using contourlet transform and then applied Singular value decomposition to mid frequency subband coefficients. The singular values of mid frequency subband coefficients of color watermark image are embedded into singular values of mid frequency sub-band coefficients of host color image in Red, Green and Blue color spaces simultaneously based on spread spectrum technique. The experimental results shows that the proposed hybrid watermarking scheme is robust against common image processing operations such as, JPEG, JPEG 2000 compression, cropping, Rotation, histogram equalization, low pass filtering ,median filtering, sharpening, shearing ,salt & Pepper noise, Gaussian noise, grayscale conversion etc. It has also been shown the variation of visual quality of watermarked image for different scaling factors. The comparative analysis reveals that the proposed watermarking scheme out performs the color image watermarking schemes reported recently.

Comparison of Fuzzy Contrast Enhancement Techniques

The aim of the image enhancement is to improve the interpretability or perception of the information in images for human viewers, or to provide ‘better’ input for other automated image processing techniques. It is an indispensable tool for researchers in wide verity of fields including art studies, medical imaging, forensics and atmospheric sciences. Most of images like satellite images, medical images and even real life photographs may suffer from poor contrast due to the inadequate or insufficient lighting during image acquiring. So it is necessary to enhance the contrast of an image. In this paper two enhancement techniques namely fuzzy rule based contrast enhancement, and contrast enhancement using intensification operator (INT) are presented for the low contrast grayscale images. In first technique fuzzy system response function is obtained by simple if-then rules, and in second technique the fuzzy contrast intensification operator is taken as a tool for the enhancement in the fuzzy property domain. Comparative analysis of these enhancement techniques is carried out by means of index of fuzziness (IOF) and processing time.

Optimization of TESPAR Features using Robust F-Ratio for Speaker Recognition

This paper deals with implementing an efficient optimization technique for designing an automatic speaker recognition (ASR) System, which uses average F-ratio score of TESPAR features, to yield high recognition accuracy even in adverse noisy conditions. A new ranking scheme is also proposed in order to stabilize the rank of features in various noise levels by taking arithmetic mean of the F-Ratio scores obtained from various levels of signal to noise ratio (SNR). The result is presented for a text-dependent ASR system with 20 speaker database. An RBF (radial basis function) neural network is used for recognition purpose

A Hybrid Swarm Optimization for Energy Efficient Clustering in Multi-hop Wireless Sensor Network

Wireless sensor network refers to distributed sets of embedded devices, all of them having processing units, wireless transmission interface and sensors or actuators. Data accumulation through effective network organizations helps nodes to be split into small sets known as clusters. This grouping of sensor nodes as clusters is known as clustering. All clusters have leaders known as cluster heads (CHs). Clustering networks for minimizing total distance is an NP-hard issue. For a particular network topology, it is hard to discover optimum quantity of cluster-heads as well as their positions. The current article suggests a hybrid differential evolution with multi objective bee swam optimization (MOBSO-DE) for efficient clustering. CH selection process is based on communication energy and factors like residual energy and energy constraint metric. Simulation shows that the new MOBSO-DE method outperformed LEACH and MOBSO for packet delivery ratio and network lifetime.

Energy Efficient Clustering in Multi-hop Wireless Sensor Networks Using Differential Evolutionary MOPSO

The primary challenge in organizing sensor networks is energy efficacy. This requisite for energy efficacy is because sensor nodes capacities are limited and replacing them is not viable. This restriction further decreases network lifetime. Node lifetime varies depending on the requisites expected of its battery. Hence, primary element in constructing sensor networks is resilience to deal with decreasing lifetime of all sensor nodes. Various network infrastructures as well as their routing protocols for reduction of power utilization as well as to prolong network lifetime are studied. After analysis, it is observed that network constructions that depend on clustering are the most effective methods in terms of power utilization. Clustering divides networks into inter-related clusters such that every cluster has several sensor nodes with a Cluster Head (CH) at its head. Sensor gathered information is transmitted to data processing centers through CH hierarchy in clustered environments. The current study utilizes Multi-Objective Particle Swarm Optimization (MOPSO)-Differential Evolution (DE) (MOPSO-DE) technique for optimizing clustering.

Fuzzy Edge Linking Process on Fuzzy Noise Filtered Image

Processing of images plays a vital role in many fields such as medical and scientific applications. During the transmission of images, effect of noise plays a key role. A fuzzy filter is presented for additive noise removal from color images. During the process of noise removal, some of the edges may be disappeared. This paper presents two independent fuzzy based edge linking algorithms which are capable of finding a set of edge points in an image and linking these edge points by thresholding. The first algorithm includes a set of 16 fuzzy templates, representing the edge profiles of different types. The second algorithm relies on the image gradient to locate breaks in uniform regions and is based on fuzzy if-then rules. Performance evaluation of these algorithms is known by calculating peak signal to noise ratio (PSNR).