Probabilistic Detection Method Research Articles

Probabilistic Detection Methods for Acoustic Surveillance Using Audio Histograms

Acoustic surveillance is gaining importance given the pervasive nature of multimedia sensors being deployed in all environments. In this paper, novel probabilistic detection methods using audio histograms are proposed for acoustic event detection in a multimedia surveillance environment. The proposed detection methods use audio histograms to classify events in a well-defined acoustic space. The proposed methods belong to the category of novelty detection methods, since audio data corresponding to the event is not used in the training process. These methods hence alleviate the problem of collecting large amount of audio data for training statistical models. These methods are also computationally efficient since a conventional audio feature set like the Mel frequency cepstral coefficients in tandem with audio histograms are used to perform acoustic event detection. Experiments on acoustic event detection are conducted on the SUSAS database available from Linguistic data consortium. The performance is measured in terms of false detection rate and true detection rate. Receiver operating characteristics curves are obtained for the proposed probabilistic detection methods to evaluate their performance. The proposed probabilistic detection methods perform significantly better than the acoustic event detection methods available in literature. A cell phone-based alert system for an assisted living environment is also discussed as a future scope of the proposed method. The performance evaluation is presented as number of successful cell phone transactions. The results are motivating enough for the system to be used in practice.

A probabilistic method for the detection and genotyping of small indels from population-scale sequence data

High-throughput sequencing technologies have made population-scale studies of human genetic variation possible. Accurate and comprehensive detection of DNA sequence variants is crucial for the success of these studies. Small insertions and deletions represent the second most frequent class of variation in the human genome after single nucleotide polymorphisms (SNPs). Although several alignment tools for the gapped alignment of sequence reads to a reference genome are available, computational methods for discriminating indels from sequencing errors and genotyping indels directly from sequence reads are needed. We describe a probabilistic method for the accurate detection and genotyping of short indels from population-scale sequence data. In this approach, aligned sequence reads from a population of individuals are used to automatically account for context-specific sequencing errors associated with indels. We applied this approach to population sequence datasets from the 1000 Genomes exon pilot project generated using the Roche 454 and Illumina sequencing platforms, and were able to detect a significantly greater number of indels than reported previously. Comparison to indels identified in the 1000 Genomes pilot project demonstrated the sensitivity of our method. The consistency in the number of indels and the fraction of indels whose length is a multiple of three across different human populations and two different sequencing platforms indicated that our method has a low false discovery rate. Finally, the method represents a general approach for the detection and genotyping of small-scale DNA sequence variants for population-scale sequencing projects. A program implementing this method is available at http://polymorphism.scripps.edu/~vbansal/software/piCALL/

A probabilistic method for detection and location of very high resistive earth faults

Abstract A probabilistic method is presented for the detection and location of very high resistive, permanent single-phase earth faults in medium voltage, radial operated distribution networks (20 kV). The systems considered have an unearthed or a compensated neutral and the fault resistances covered are in the range 5–220 kΩ. The algorithms of the method are based on the change of the neutral-voltage and zero-sequence currents, measured at the MV substation and also at the distribution line locations. The feasibility of the method was verified by an analysis based on system data recorded during field tests in real power systems.