SPSA Based Feature Relevance Estimation For Video Retrieval

With the availability of a huge amount of video data on various sources, efficient video retrieval tools are increasingly in demand. Video being a multi-modal data, the perceptions of ``relevance'' between the user provided query video (in case of Query-By-Example type of video search) and retrieved video clips are subjective in nature. We present an efficient video retrieval method that takes user's feedback on the relevance of retrieved videos and iteratively reformulates the input query feature vectors (QFV) for improved video retrieval. The QFV reformulation is done by a simple, but powerful feature weight optimization method based on Simultaneous Perturbation Stochastic Approximation (SPSA) technique. A video retrieval system with video indexing, searching and relevance feedback (RF) phases is built for demonstrating the performance of the proposed method. The query and database videos are indexed using the conventional video features like color, texture, etc. However, we use the comprehensive and novel methods of feature representations, and a spatio-temporal distance measure to retrieve the top M videos that are similar to the query. In feedback phase, the user activated iterative on the previously retrieved videos is used to reformulate the QFV weights (measure of importance) that reflect the user's preference, automatically. It is our observation that a few iterations of such feedback are generally sufficient for retrieving the desired video clips. The novel application of SPSA based RF for user-oriented feature weights optimization makes the proposed method to be distinct from the existing ones. The experimental results show that the proposed RF based video retrieval exhibit good performance.

Bayesian and Decision Tree Approaches for Pattern Recognition Including Feature Measurement Costs

The minimum cost classifier when general cost functionsare associated with the tasks of feature measurement and classification is formulated as a decision graph which does not reject class labels at intermediate stages. Noting its complexities, a heuristic procedure to simplify this scheme to a binary decision tree is presented. The optimizationof the binary tree in this context is carried out using ynamicprogramming. This technique is applied to the voiced-unvoiced-silence classification in speech processing.

Feature selection and the concept of immediate neighborhood in the context of clustering techniques

The concept of feature selection in a nonparametric unsupervised learning environment is practically undeveloped because no true measure for the effectiveness of a feature exists in such an environment. The lack of a feature selection phase preceding the clustering process seriously affects the reliability of such learning. New concepts such as significant features, level of significance of features, and immediate neighborhood are introduced which result in meeting implicitly the need for feature slection in the context of clustering techniques.

Feature selection and the concept of immediate neighborhood in the context of clustering techniques

The concept of feature selection in a nonparametric unsupervised learning environment is practically undeveloped because no true measure for the effectiveness of a feature exists in such an environment. The lack of a feature selection phase preceding the clustering process seriously affects the reliability of such learning. New concepts such as significant features, level of significance of features, and immediate neighborhood are introduced which result in meeting implicitly the need for feature slection in the context of clustering techniques.

Clinical and immunological studies on persons exposed toParthenium hysterophorus L

Studies on 300 persons subjected by occupational hazard to the allergenic weed, Parthenium hysterophorus L. for periods ranging from 3 to 12 months revealed that 4% of them developed contact dermatitis of the exposed parts of the body, while 56% of them got sensitized to the weed without apparently exhibiting any dermatitis. None of them suffered from allergic manifestations like rhinitis or bronchial asthma during the period of study which extended for 2 years.

Feature extraction for fingerprint classification

This paper presents two algorithms for smoothing and feature extraction for fingerprint classification. Deutsch's(2) Thinning algorithm (rectangular array) is used for thinning the digitized fingerprint (binary version). A simple algorithm is also suggested for classifying the fingerprints. Experimental results obtained using such algorithms are presented.

A glimpse into the clinical proteome of human malaria parasites Plasmodium falciparum and Plasmodium vivax

Malaria causes a worldwide annual mortality of about a million people.Rapidly evolving drug-resistant species of the parasite have created a pressing need for the identification of new drug targets and vaccine candidates. By developing fractionation protocols to enrich parasites from low-parasitemia patient samples, we have carried out the first ever proteomics analysis of clinical isolates of early stages of Plasmodium falciparum (Pf) and P. vivax. Patient-derived malarial parasites were directly processed and analyzed using shotgun proteomics approach using high-sensitivity MS for protein identification. Our study revealed about 100 parasite-coded gene products that included many known drug targets such as Pf hypoxanthine guanine phosphoribosyl transferase, Pf L-lactate dehydrogenase, and Plasmepsins. In addition,our study reports the expression of several parasite proteins in clinical ring stages that have never been reported in the ring stages of the laboratory-cultivated parasite strain. This proof-of-principle study represents a noteworthy step forward in our understanding of pathways elaborated by the parasite within the malaria patient and will pave the way towards identification of new drug and vaccine targets that can aid malaria therapy.

Prevalence of multidrug-resistant Staphylococcus aureus in diabetics clinical samples

Antibiotic resistance in 40 Staphylococcus aureus clinical isolates from 110 diabetic patients (36%) was evaluated. Of these, 32 (80%) of the isolates showed multidrug-resistance to more than eight antibiotics and 35% isolates were found to be methicillin resistant S. aureus (MRSA). All 40 S. aureus strains (100%) screened from diabetic clinical specimens were resistant to penicillin, 63% to ampicillin, 55% to streptomycin, 50% to tetracycline and 50% to gentamicin. Where as low resistance rate was observed to ciprofloxacin (20%) and rifampicin (8%). In contrast, all (100%) S. aureus strains recorded susceptibility to teicoplanin, which was followed by vancomycin (95%). Genotypical examination revealed that 80% of the aminoglycoside resistant S. aureus (ARSA) have aminoglycoside modifying enzyme (AME) coding genes; however, 20% of ARSA which showed non-AME mediated (adaptive) aminoglycoside resistance lacked these genes in their genome. In contrast all MRSA isolates possessed mecA, femA genetic determinants in their genome.

Aminoglycoside-Resistance Mechanisms in Multidrug-Resistant Staphylococcus aureus Clinical Isolates

Aminoglycoside resistance in six clinically isolated Staphylococcus aureus was evaluated. Genotypical examination revealed that three isolates (HLGR-10, HLGR-12, and MSSA-21) have aminoglycoside-modifying enzyme (AME) coding genes and another three (GRSA-2, GRSA-4, and GRSA-6) lacked these genes in their genome. Whereas isolates HLGR-10 and HLGR-14 possessed bifunctional AME coding gene aac(6′)-aph(2′′), and aph(3′)-III and showed high-level resistance to gentamycin and streptomycin, MSSA-21 possessed aph(3′)-III and exhibited low resistance to gentamycin, streptomycin, and kanamycin. The remaining three isolates (GRSA-2, GRSA-4, and GRSA-6) exhibited low resistance to all the aminoglycosides because they lack aminoglycoside-modifying enzyme coding genes in their genome. The transmission electron microscopy of the three isolates revealed changes in cell size, shape, and septa formation, supporting the view that the phenomenon of adaptive resistance is operative in these isolates.

Block-based feature detection and matching for mosaicing of camera-captured document images

In this paper, we present a new feature-based approach for mosaicing of camera-captured document images. A novel block-based scheme is employed to ensure that corners can be reliably detected over a wide range of images. 2-D discrete cosine transform is computed for image blocks defined around each of the detected corners and a small subset of the coefficients is used as a feature vector A 2-pass feature matching is performed to establish point correspondences from which the homography relating the input images could be computed. The algorithm is tested on a number of complex document images casually taken from a hand-held camera yielding convincing results.

Learning Feature Trajectories Using Gabor Filter Bank for Human Activity Segmentation and Recognition

We describe a novel method for human activity segmentation and interpretation in surveillance applications based on Gabor filter-bank features. A complex human activity is modeled as a sequence of elementary human actions like walking, running, jogging, boxing, hand-waving etc. Since human silhouette can be modeled by a set of rectangles, the elementary human actions can be modeled as a sequence of a set of rectangles with different orientations and scales. The activity segmentation is based on Gabor filter-bank features and normalized spectral clustering. The feature trajectories of an action category are learnt from training example videos using dynamic time warping. The combined segmentation and the recognition processes are very efficient as both the algorithms share the same framework and Gabor features computed for the former can be used for the later. We have also proposed a simple shadow detection technique to extract good silhouette which is necessary for good accuracy of an action recognition technique.

On Guo and Nixon's Criterion for Feature Subset Selection: Assumptions, Implications,and Alternative Options

Guo and Nixon proposed a feature selection method based on maximizing I(x; Y),the multidimensional mutual information between feature vector x and class variable Y. Because computing I(x; Y) can be difficult in practice, Guo and Nixon proposed an approximation of I(x; Y) as the criterion for feature selection. We show that Guo and Nixon's criterion originates from approximating the joint probability distributions in I(x; Y) by second-order product distributions. We remark on the limitations of the approximation and discuss computationally attractive alternatives to compute I(x; Y).

Detection,amplification and sequence homology of sodC in clinical isolates of Salmonella sp

Background & objectives: Periplasmic copper and zinc superoxide dismutase (Cu,Zn-SOD or SodC) is an important component of the antioxidant shield which protects bacteria from the phagocytic oxidative burst. Cu,Zn-SODs protect Gram-negative bacteria against oxygen damage which have also been shown to contribute to the pathogenicity of these bacterial species. We report the presence of SodC in drug resistant Salmonella sp. isolated from patients suffering from enteric fever. Further sodC was amplified, cloned into Escherichia coli and the nucleotide sequence and amino acid sequence homology were compared with the standard strain Salmonella Typhimurium 14028. Methods: Salmonella enterica serovar Typhi (S. Typhi) and Salmonellaenterica serovar Paratyphi (S. Paratyphi) were isolated and identified from blood samples of the patients. The isolates were screened for the presence of Cu, Zn-SOD by PAGE using KCN as inhibitor of Cu,Zn-SOD. The gene (sodC) was amplified by PCR, cloned and sequenced. The nucleotide and amino acid sequences of sodC were compared using CLUSTAL X.Results: SodC was detected in 35 per cent of the Salmonella isolates. Amplification of the genomic DNA of S. Typhi and S. Paratyphi with sodC specific primers resulted in 519 and 515 bp amplicons respectively. Single mutational difference at position 489 was observed between thesodC of S. Typhi and S. Paratyphi while they differed at 6 positions with the sodC of S. Typhimurium 14028. The SodC amino acid sequences of the two isolates were homologous but 3 amino acid difference was observed with that of standard strain S. Typhimurium 14028.Interpretation & conclusions: The presence of SodC in pathogenic bacteria could be a novel candidate as phylogenetic marker.

A preliminary survey of dermatomycoses in Bangalore

Clinical and mycological investigations were made on 225 cases of suspected dermatomycoses. Of these, 102 were microscopically positive. But only 63 were culturally positive, and these are analysed here with regard to clinical patterns and aetiological species, age, sex and occupational incidence and susceptibility to griseofulvin in vitro. As in most other parts of India, Trichophyton rubrum was the dominant species. A high proportion of Epidermophyton floccosum was an unusual feature seen. Of the clinical types, tinea cruris was the most common. The isolates were sensitive to griseofulvin at low concentrations of 1 to 5 μg per ml of agar medium, E. floccosum being the most sensitive.

Immediate hypersensitivity to Parthenium hysterophorus. II. Clinical studies on the prevalence of parthenium rhinitis

The airborne pollen of the South American weed, Parthenium hysterophorus (American feverfew), accidentally introduced into India was found to be responsible for severe allergic rhinitis. A random clinical survey conducted on 2035 residents of Bangalore city with the aid of questionnaires and skin tests revealed that 7.1% of the study population was suffering from allergic rhinitis due to exposure to Parthenium pollen. Skin-prick tests performed on 1294 clinic patients suffering from nasobronchial allergy during the past 4 years have also shown that 42.5% were sensitive to Parthenium pollen. IgE and IgG antibodies specific for Parthenium pollen allergens were demonstrable in the sera of Parthenium-sensitive rhinitis patients. The specificity of these antibodies to Parthenium allergens was established by ELISA. A 7- to 11-fold higher stimulation was observed when lymphocytes from two Parthenium-sensitive rhinitis patients were treated in vitro with Parthenium pollen extract. To our knowledge, nowhere in the world has such a high incidence of allergic rhinitis due to a single pollen ever been reported.