The growth of microblogging sites such as Biomedical, biomedical, defect, or bug databases makes it difficult for web users to share and express their context identification of sequential key phrases and their categories on text clustering applications. In the traditional document classification and clustering models, the features associated with TREC texts are more complex to analyze. Finding relevant feature-based key phrase patterns in the large collection of unstructured documents is becoming increasingly difficult, as the repository's size increases. The purpose of this study is to develop and implement a new hierarchical document clustering framework on a large TREC data repository. A document feature selection and clustered model are used to identify and extract MeSH related documents from TREC biomedical clinical benchmark datasets. Efficiencies of the proposed model are indicated in terms of computational memory, accuracy, and error rate, as demonstrated by experimental results.

Health care is a major research domain needed instantaneous solutions. Due to the digitalization of data in each and every domain it is becoming tedious to store and analysis. So, the demand of proficient algorithms for health care data analysis is also increasing. Predictive analytics is the major demand from the health care community to the computing researches in order to predict and reduce the potential health catastrophes. Parallel research attempts are made to predict the possibilities of the disease on the different health care domains at various regions. However, those attempts are limited and not remarkable to achieve the desired outcomes. Recently, in the field of data analytics; Machine Learning techniques became popular in generating optimized solutions with effective data processing capabilities. Henceforth, this research work considers the heart disease analysis using machine learning techniques to determine the disease severity levels. Experiments are made on UCI heart disease dataset and our results shows 92% accuracy the heart severity detection.

Remote sensing image classification is very much essential for many socio, economic and environmental applications in the society. They aid in agriculture monitoring, urban planning, forest monitoring, etc. Classification of a remote sensing image is still a challenging problem because of its multifold problems. A new algorithm LCDFOSCA (Linear Contact Distribution First Order Statistics Classification Algorithm) is proposed in this paper to extract the texture features from a Color remote sensing image. This algorithm uses linear contact distributions, mathematical morphology, and first-order statistics to extract the texture features. Later k-means is used to cluster these feature vectors and then classify the image. This algorithm is implemented on NRSC ‘Tirupathi’ area 2.5m, 1m color images and on Google Earth images. The algorithm is evaluated with various measures like the dice coefficient, segmentation accuracy, etc and obtained promising results.

Sclerosis is a disease that triggers mainly due to damage of nerve cells in the brain and spinal cord. Various impairments are observed with this disease. Analyzing this type of images is needed for the medical research field for early stage identification. So, the present paper uses Bivariate Gaussian Mixture distribution for analyzing the noisy sclerosis images. For this, the present paper uses neural network for classification. The proposed method is evaluated with various images of brain web repository and the results show the efficiency of the proposed method. 

Wavelet Transform (WT) has widely been used in signal processing. WT breaks a signal into its wavelets that are scaled and shifted versions of given signal. Thus wavelets are able represent graphical objects. The irregular shape and compact support of wavelets made them ideal for analyzing non-stationary signals. They are useful in analysis in both temporal and frequency domains. In contract, the Fourier transform provides information in frequency domain lacking in information in time domain. Thus wavelets became popular for signal processing and image processing applications. Nevertheless, wavelets suffer from a drawback as they cannot effectively represent images at different angles and different scales. To overcome this problem, of late, Curvelet Transform (CT) came into existence. CT is nothing but the higher dimensional generalization of WT which can effectively represent images at different angles and different scales. In this paper we proposed a CT method that is used to represent textures and classify them. The methodology used in this paper has an underlying approach that exploits statistical features of curvelets that resulted in curvelet decomposition. We built a prototype application using MATLAB to demonstrate proof of the concept. 

Recently, artificial neural networks are fund to be efficiently used in clustering algorithms. So, the present paper focuses on the development of a novel clustering method based on artificial neural networks. The present paper uses an enhancement filter to enhance the segments in the input image. After this, the various sub images are generated and features are computed for each sub and edge image. Finally, the Self Organizing Map (SOM) is used for clustering process. The proposed novel method is evaluated with a database of 795 leaf images. Further various Probability Distributed Functions (PDFs) are used to evaluate the efficacy of the proposed method. The performance measures of the proposed method indicate the efficiency of the extended clustering method with SOM.

The reliability of a software depends on the quality. So, the software growth models require efficient quality assessment procedure. It can be estimated by various parameters. The current paper proposes a novel approach for assessment of quality based on the Generalized Weighted Laplacian (GWL) method. The proposed method evaluates various parameters for detection and removal time. The Mean Value Function (MVF) is then calculated and the quality of the software is estimated, based on the detection of failures. The proposed method is evaluated on process CMMI level 5 project data and the experimental results shows the efficiency of the proposed method.

Currently clustering techniques play a vital role in object recognition process. The clustering techniques are found to be efficient with neural networks. So, the present paper proposed a novel method for clustering the input objects with Self-Organizing Map (SOM). The proposed method considers the input object as a random closed set. The random set can be efficiently described with various features viz., volume fractions, co-variance and contact distributions etc. In the proposed method, the input object is described efficiently with spherical contact distribution. The proposed method is experimented with the leaf data set with 795 images. The performance of the proposed method is evaluated with various topologies of SOM and is measured with four measures viz., FNR, FPR, TPR and TNR. The results indicate the efficiency of the proposed method.   

One of the popular descriptor for texture classification is the local binary pattern (LBP). LBP and its variants derives local texture features effectively. This paper integrates the significant local features derived from uniform LBPs(ULBP) and threshold based conversion factor non-uniform (NULBP) with complete textons. This integrated approach represents the complete local structural features of the image.   The ULBPs are proposed to overcome the wide histograms of LBP. The ULBP contains fundamental aspects of local features.  The LBP is more prone to noise and this may transform ULBP into NULBP and this degrades the overall classification rate. To addresses this, this paper initially transforms back, the ULBPs that are converted in to NULBPs due to noise using a threshold based conversion factor and derives noise resistant fundamental texture (NRFT) image. In the literature texton co-occurrence matrix(TCM) and multi texton histogram (MTH) are derived on a 2x2 window. The main disadvantage of the above texton groups is they fail in representing complete textons. In this paper we have integrated our earlier approach “complete texton matrix (CTM)” [16] on NRFT images. This paper computes the gray level co-occurrence matrix (GLCM) features on the proposed NRFCTM (noise resistant fundamental complete texton matrix) and the features are given to machine learning classifiers for a precise classification. The proposed method is tested on the popular databases of texture classification and classification results are compared with existing methods.

This work presents an enhancement in accuracy for brain disorder detection using optimal unification. The strategy for detection of segments and brain regions causing medical conditions are described. This work demonstrates the application of multilateral filter and applied watershed method with EM-GM method. The most popular existing techniques of brain tumor detection are not optimal compared to this combination of Watershed and EM-GM technique with the proposed optimal unification technique. The result is optimally unified and achieved high accuracy. The multilateral filter enhances the image edges for better segmentation using signal amplitude moderation of the pixel. In the unification process, the optimal sets of segments are divided and finest merged results are considered with the brain regions detected with anomalies. Henceforth the number of possible medical investigations will be reduced.

At present trillion of bytes of information is being created by projects particularly in web. To accomplish the best choice for business benefits, access to that information in a very much arranged and intuitive way is dependably a fantasy of business administrators and chiefs. Information warehouse is the main feasible arrangement that can bring the fantasy into reality. The upgrade of future attempts to settle on choices relies on upon the accessibility of right data that depends on nature of information basic. The quality information must be created by cleaning information preceding stacking into information distribution center following the information gathered from diverse sources will be grimy. Once the information have been pre-prepared and purified then it produces exact results on applying the information mining question. There are numerous cases where the data is sparse in nature. To get accurate results with sparse data is hard. In this paper the main goal is to fill the missing values in acquired data which is sparse in nature. Precisely caution must be taken to choose minimum number of text pieces to fill the holes for which we have used Jaccard Dissimilarity function for clustering the data which is frequent in nature.

Texture deals with the visual properties of an image. Texture analysis plays a dominant role for image segmentation. In texture segmentation, model based methods are superior to model free methods with respect to segmentation methods. This paper addresses the application of multivariate generalized Gaussian mixture probability model for segmenting the texture of an image integrating with hierarchical clustering. Here the feature vector associated with the texture is derived through DCT coefficients of the image blocks. The model parameters are estimated using EM algorithm. The initialization of model parameters is done through hierarchical clustering algorithm and moment method of estimation. The texture segmentation algorithm is developed using component maximum likelihood under Bayesian frame. The performance of the proposed algorithm is carried through experimentation on five image textures selected randomly from the Brodatz texture database. The texture segmentation performance measures such as GCE, PRI and VOI have revealed that this method outperform over the existing methods of texture segmentation using Gaussian mixture model. This is also supported by computing confusion matrix, accuracy, specificity, sensitivity and F-measure.

An overview of state of art in computerized object recognition techniques regarding digital images is revised. Advantages of shape based techniques are discussed. Importance of "Fourier Descriptor" (FD) for the shape based object representation is described. A survey for the available shape signature assignment methods with Fourier descriptors is presented. Details for the design of shape signature containing the crucial information of corners of the object are depicted. A novel shape signature is designed basing on the Farthest Point Angle (FPA) which corresponds to the contour point. FPA signature considers the computation of the angle between the line drawn from each contour point and the line drawn from the farthest corner point. Histogram for each 15o angle conceiving the information of the object is constructed. FPA signature is evaluated for three standard databases; viz., two in Kimia {K-99, K-216} and one in MPEG CE-1 Set B. The performance of the present FPA method estimated through recognition rate, time and degree of matching and is found to be higher.

Many diversified applications do exist in science & technology, which make use of the primary theory of a recognition phenomenon as one of its solutions. Recognition scenario is incorporated with a set of decisions and the action according to the decision purely relies on the quality of extracted information on utmost applications. Thus, the quality decision making absolutely reckons on processing momentum and precision which are entirely coupled with recognition methodology. In this article, a latest rule is formulated based on the degree of correlation to characterize the generalized recognition constraint and the application is explored with respect to image based information extraction. Machine learning based perception called feed forward architecture of Artificial Neural Network has been applied to attain the expected eminence of elucidation. The proposed method furnishes extraordinary advantages such as less memory requirements, extremely high level security for storing data, exceptional speed and gentle implementation approach.

In the present day real time applications of visual object tracking in surveillance, it has become extremely complex, time consuming and tricky to do the tracking when there are occlusions are present for small duration or for longer time and also when it is done in outdoor environments. In these conditions, the target to be tracked can be lost for few seconds and that should be tracked as soon as possible. As from the literature it is observed that particle filter can be able to track the target robustly in different kinds of background conditions, and it’s robust to partial occlusion. However, this tracking cannot recover from large proportion of occlusion and complete occlusion, to avoid this condition, we proposed two new algorithms (modified kalman and modified particle filter) for fast tracking of objects in the presence of occlusions. We considered the complete occlusion of tracking object and the main objective is how fast the system is able to track the object after the occlusion is crossed. From the experimental results, it is observed that the proposed algorithms have shown good improvement in results compared to the traditional methods.

This paper gives the survey of the existing developments of Visual object target tracking using particle filter from the last decade and discusses the advantage and disadvantages of various particle filters. A variety of different approaches and algorithms have been proposed in literature. At present most of the work in Visual Object Target Tracking is focusing on using particle filter. The particle filters has the advantage that they deal with nonlinear models and non-Gaussian innovations, and they focus sequentially on the higher density regions of the state space, mostly parallelizable and easy to implement, so it gives a robust tracking framework, as it models the uncertainty and showing good improvement in the recognition performance compared to the kalman filter and other filters like Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF).Various features and classifiers that are used with particle filter are given in this survey.

In the context of Acoustic Echo Cancellation (AEC), sparseness level of acoustic impulse response (AIR) varies greatly in mobile environments. The modified sparseness-controlled Improved PNLMS (MSC-IPNLMS) algorithm proposed in this paper, exploits the sparseness measure of AIR using l1, l2 and l∞ norms. The MSC-IPNLMS is the modified version of SC-IPNLMS which uses sparseness measure based on l1 and l2 norms. Sparseness measure using l1, l2 and l∞ norms is the good representation of both sparse and dense impulse response, where as the measure which uses l1 and l2 norms is the good representation of sparse impulse response only. The MSC-IPNLMS is based on IPNLMS which allocates adaptation step size gain in proportion to the magnitude of estimated filter weights. By estimating the sparseness of the AIR, the proposed MSC-IPNLMS algorithm assigns the gains for each step size such that the proportionate term of the IPNLMS will be given higher weighting for sparse impulse responses. For a less sparse impulse response, a higher weighting will be given to the NLMS term. Simulation results, with input as white Gaussian noise (WGN), show the improved performance over the SC-IPNLMS algorithm in both sparse and dense AIR.

Probability distributions formulate the basic framework for developing several segmentation algorithms. Among the various segmentation algorithms, skin colour segmentation is one of the most important algorithms for human computer interaction. Due to various random factors influencing the colour space, there does not exist a unique algorithm which serve the purpose of all images. In this paper a novel and new skin colour segmentation algorithms is proposed based on bivariate Pearson type II mixture model since the hue and saturation values always lies between 0 and 1. The bivariate feature vector of the human image is to be modeled with a Pearson type II mixture (bivariate Beta mixture) model. Using the EM Algorithm the model parameters are estimated. The segmentation algorithm is developed under Bayesian frame. Through experimentation the proposed skin colour segmentation algorithm performs better with respect to segmentation quality metrics such as PRI, VOI and GCE. The ROC curves plotted for the system also revealed that the proposed algorithm can segment the skin colour more effectively than the algorithm with Gaussian mixture model for some images.

In this paper, we introduce a face recognition algorithm based on doubly truncated multivariate Gaussian mixture model with DCT under logarithm domain. In face recognition, the face image is subject to the variation of illumination. The effect of illumination cannot be avoided by mere consideration of DCT coefficients as feature vector. The illumination effect can be minimized by utilizing DCT coefficients under logarithm domain and discarding sum of the DCT coefficients which represents the illumination in the face image. Here, it is assumed that the DCT coefficients under logarithm domain after adjusting the illumination follow a doubly truncated multivariate Gaussian mixture model. The truncation on the feature vector has a significant influence in improving the recognition rate of the system using EM algorithm with K-means or hierarchical clustering, the model parameters are estimated. A face recognition system is developed under Bayesian frame using maximum likelihood. The performance of the system is demonstrated by using the databases namely, JNTUK and Yale and comparing it’s performance with the face recognition system based on GMM. It is observed that the proposed face recognition system outperforms the existing systems.