Film segmentation and indexing using autoassociative neural networks

In this paper, Autoassociative Neural Network (AANN) models are explored for segmentation and indexing the films (movies) using audio features. A two-stage method is proposed for segmenting the film into sequence of scenes, and then indexing them appropriately. In the first stage, music and speech plus music segments of the film are separated, and music segments are labelled as title and fighting scenes based on their position. At the second stage, speech plus music segments are classified into normal, emotional, comedy and song scenes. In this work, Mel frequency cepstral coefficients (MFCCs), zero crossing rate and intensity are used as audio features for segmentation and indexing the films. The proposed segmentation and indexing method is evaluated on manual segmented Hindi films. From the evaluation results, it is observed that title, fighting and song scenes are segmented and indexed without any errors, and most of the errors are observed in discriminating the comedy and normal scenes. Performance of the proposed AANN models used for segmentation and indexing of the films, is also compared with hidden Markov models, Gaussian mixture models and support vector machines.     

Pitch synchronous and glottal closure based speech analysis for language recognition

This paper explores pitch synchronous and glottal closure (GC) based spectral features for analyzing the language specific information present in speech. For determining pitch cycles (for pitch synchronous analysis) and GC regions, instants of significant excitation (ISE) are used. The ISE correspond to the instants of glottal closure (epochs) in the case of voiced speech, and some random excitations like onset of burst in the case of nonvoiced speech. For analyzing the language specific information in the proposed features, Indian language speech database (IITKGP-MLILSC) is used. Gaussian mixture models are used to capture the language specific information from the proposed features. Proposed pitch synchronous and glottal closure spectral features are evaluated using language recognition studies. The evaluation results indicate that language recognition performance is better with pitch synchronous and GC based spectral features compared to conventional spectral features derived through block processing. GC based spectral features are found to be more robust against degradations due to background noise. Performance of proposed features is also analyzed on standard Oregon Graduate Institute Multi-Language Telephone-based Speech (OGI-MLTS) database.     

Optimal weight tuning method for unit selection cost functions in syllable based text-to-speech synthesis

This paper proposes a method for tuning the weights of unit selection cost functions in syllable based text-to-speech (TTS) synthesis system. In this work, unit selection cost functions, namely target cost and concatenation cost, are designed appropriate to syllables. The method tunes the weights in such a way that perceptual preference patterns are appropriately considered while selecting the units. The method uses genetic algorithm to derive the optimal weights. Fitness function is designed to map perceptual preference patterns into weights of unit selection cost functions. The effectiveness of proposed method is evaluated by both subjective and objective measures. From the results, it is observed that the derived optimal weights can synthesize good quality speech compared to manually tuned weights.     

A graphical surface navigation technique for multiobjective programming

A methodology is presented to explore the tradeoffs associated with the numerous solutions typically obtained through multiobjective optimization procedures. The method is a graphical surface navigation technique based on a spring equilibrium analogy. The procedure is cast in the framework of an interactive graphical session, which permits a user easily to move from one noninferior point to another on the objective response surface. A zooming capability is also incorporated which facilitates refinement of the noninferior set approximation over particular areas of interest and as well, provides for the capability to examine the exact noninferior set. A three-dimensional and a five-dimensional problem are used to illustrate the method.     

A PCR-Based Assay for the Detection and Differentiation of Potential Fumonisin-Producing Fusarium verticillioides Isolated from Indian Maize Kernels

One hundred and three Fusarium isolates from maize samples collected from different districts of Karnataka state, India, were analyzed with genus-specific, species-specific, and potential fumonisin specific oligonucleotide primers. One set of genus-specific primers ITS F and ITS R based on a highly conserved ITS region of the genus Fusarium were used to differentiate Fusarium species from closely related genera. All the Fusarium species tested scored positive with the ITS pair of primers. Detection and identification of Fusarium verticillioides species was done by using a newly designed reverse primer VERT-R (5′- CGA CTC ACG GCC AGG AAA CC ?3′) based on an intergenic spacer sequence (IGS) combined with an already designed forward primer VERTF-1 (5′-GCG GGA ATT CAA AAG TGG CC -3′) published previously. Out of 103 Fusarium species tested, 83 isolates of F. verticillioides scored positive for VERTF-1/ VERT-R species-specific pair of primers. Further to discriminate potential fumonisin-producing and nonproducing strains of F. verticillioides, the VERTF-1/VERTF-2 set of primers [VERTF-1 (5′-GCG GGA ATT CAA AAG TGG CC -3′) and VERTF-2 (5′-GAG GGC GCG AAA CGG ATC GG -3′)] were used. 64 isolates of F. verticillioides scored positive for VERTF-1/ VERTF-2 pair of primers. In total, three primers, one forward primer VERTF-1 and two reverse primers VERT-R and VERTF-2, were used for the confirmation of F. verticillioides up to the species level and the second pair of primers were used to confirm the potential for fumonisin production. The developed PCR assay should provide a powerful tool for the detection and differentiation of potential fumonisin-producing F. verticillioides strains in a population.     

Persistence of pesticides in water,sediment and fish from fish farms in Kolleru Lake,India

Kolleru Lake is a freshwater lake in south‐eastern India with a high potential for aquaculture. It is located between 61°32′ and 61°47′N latitude and 81°15′ and 81°27′E longitude. Various chemicals such as pesticides are used to control diseases and ectoparasites in intensive aquaculture farms in India. The main aim of this paper was to determine selected pesticides in water, sediment and fish from 12 fish farms located in Kolleru Lake in order to assess the quality of fish for human consumption and to improve the further risk assessment and management of such chemicals in aquaculture and also to evaluate the impact of aquaculture chemicals on the Kolleru Lake environment. The chemicals used in intensive aquaculture farms were documented. Water, sediment and fish samples were collected from the fish farms during the production period and analysed for the pesticides α‐BHC, γ‐BHC, malathion, chloropyrifos, isodrin, endosulfan, dieldrin and p,p′‐DDT using gas chromatography. © 2002 Society of Chemical Industry     

Crystal structure of Ba(Mg1/3Ta2/3)O3 calcinated at 1400°C

A new phase of Ba(Mg_1/3Ta_2/3)O₃ has been obtained by calcination at 1400°C. It belongs to the cubicPm3m witha = 4·08840(4)Å and the crystal structure has been found to be the ideal perovskite by the Rietveld method from powder X-ray diffraction data.     

Sustained release optimized formulation of anastrozole-loaded chitosan microspheres: in vitro and in vivo evaluation

The purpose of this study was to develop sustained release formulation of anastrozole-loaded chitosan microspheres for treatment of breast cancer. Chitosan microspheres cross-linked with two different cross-linking agents viz, tripolyphosphate (TPP) and glutaraldehyde (GA) were prepared using single emulsion (w/o) method. A reverse phase HPLC method was developed and used for quantification of drug in microspheres and rat plasma. Influence of cross-linking agents on the properties of chitosan microspheres was extensively investigated. Formulations were characterized for encapsulation efficiency (EE), compatibility of drug with excipients, particle size, surface morphology, swelling capacity, erosion and drug release profile in phosphate buffer pH 7.4. EE varied from 30.4 ± 1.2 to 69.2 ± 3.2% and mean particle size distribution ranged from 72.5 ± 0.5 to 157.9 ± 1.5 μm. SEM analysis revealed smooth and spherical nature of microspheres. TPP microspheres exhibited higher swelling capacity, percentage erosion and drug release compared to GA microspheres. Release of anastrozole (ANS) was rapid up to 4 h followed by slow release status. FTIR analysis revealed no chemical interaction between drug and polymer. DSC analysis indicated ANS trapped in the microspheres existed in amorphous form in polymer matrix. The highest correlation coefficients (R ²) were obtained for Higuchi model, suggesting a diffusion controlled mechanism. There was significant difference in the pharmacokinetic parameters (AUC_0−∞, Kel and t_1/2) when ANS was formulated in the form of microspheres compared to pure drug. This may be attributed to slow release rate of ANS from chitosan microspheres and was detectable in rat plasma up to 48 h which correlates well with the in vitro release data.     

Speaker identification under background noise using features extracted from steady vowel regions

In this paper, we are exploring features extracted from steady vowel segments for improving the performance of speaker identification system under background noise. Steady vowel regions are produced by periodic impulse‐like excitation and they contain relatively high signal energy. Hence, speaker specific information present in steady vowel regions may be less affected by the noise. In this work, steady vowel regions are determined by using the knowledge of accurate vowel onset points and epochs. Speaker identification studies are carried out using TIMIT database for white and vehicle noises. Universal background model–Gaussian mixture model‐based modeling is explored for developing speaker models. Significant improvement in the performance of speaker identification is observed by using features extracted from steady vowel region in presence of noisy environments. Copyright © 2012 John Wiley & Sons, Ltd.     

Efficient design of an authenticated key agreement protocol for dew-assisted IoT systems

The Journal of Supercomputing - Real-time communication is a significant aspect of Internet of Things (IoT). IoT-enabled devices requires the immediate adoption of the highly distributed and...