Contributions to unsupervised and supervised learning with applications in digital image processing

311 p. : il.

Query-by-Example Spoken Term Detection ALBAYZIN 2012 evaluation: overview, systems, results, and discussion

Query-by-Example Spoken Term Detection (QbE STD) aims at retrieving data from a speech data repository given an acoustic query containing the term of interest as input. Nowadays, it has been receiving much interest due to the high volume of information stored in audio or audiovisual format. QbE STD differs from automatic speech recognition (ASR) and keyword spotting (KWS)/spoken term detection (STD) since ASR is interested in all the terms/words that appear in the speech signal and KWS/STD relies on a textual transcription of the search term to retrieve the speech data. This paper presents the systems submitted to the ALBAYZIN 2012 QbE STD evaluation held as a part of ALBAYZIN 2012 evaluation campaign within the context of the IberSPEECH 2012 Conference(a). The evaluation consists of retrieving the speech files that contain the input queries, indicating their start and end timestamps within the appropriate speech file. Evaluation is conducted on a Spanish spontaneous speech database containing a set of talks from MAVIR workshops(b), which amount at about 7 h of speech in total. We present the database metric systems submitted along with all results and some discussion. Four different research groups took part in the evaluation. Evaluation results show the difficulty of this task and the limited performance indicates there is still a lot of room for improvement. The best result is achieved by a dynamic time warping-based search over Gaussian posteriorgrams/posterior phoneme probabilities. This paper also compares the systems aiming at establishing the best technique dealing with that difficult task and looking for defining promising directions for this relatively novel task.

Calpain-Mediated Processing of Adenylate Cyclase Toxin Generates a Cytosolic Soluble Catalytically Active N-Terminal Domain

Bordetella pertussis, the whooping cough pathogen, secretes several virulence factors among which adenylate cyclase toxin (ACT) is essential for establishment of the disease in the respiratory tract. ACT weakens host defenses by suppressing important bactericidal activities of the phagocytic cells. Up to now, it was believed that cell intoxication by ACT was a consequence of the accumulation of abnormally high levels of cAMP, generated exclusively beneath the host plasma membrane by the toxin N-terminal catalytic adenylate cyclase (AC) domain, upon its direct translocation across the lipid bilayer. Here we show that host calpain, a calcium-dependent Cys-protease, is activated into the phagocytes by a toxin-triggered calcium rise, resulting in the proteolytic cleavage of the toxin N-terminal domain that releases a catalytically active "soluble AC''. The calpain-mediated ACT processing allows trafficking of the "soluble AC'' domain into subcellular organella. At least two strategic advantages arise from this singular toxin cleavage, enhancing the specificity of action, and simultaneously preventing an indiscriminate activation of cAMP effectors throughout the cell. The present study provides novel insights into the toxin mechanism of action, as the calpain-mediated toxin processing would confer ACT the capacity for a space- and time-coordinated production of different cAMP "pools'', which would play different roles in the cell pathophysiology.

NOAA/MLML Radiometric Data Acquisition and Processing

This report is a detailed description of data processing of NOAA/MLML spectroradiometry data. It introduces the MLML_DBASE programs, describes the assembly of diverse data fues, and describes general algorithms and how individual routines are used. Definitions of data structures are presented in Appendices. [PDF contains 48 pages]

Processing NOAA Spectroradiometric Data

This report outlines the NOAA spectroradiometer data processing system implemented by the MLML_DBASE programs. This is done by presenting the algorithms and graphs showing the effects of each step in the algorithms. [PDF contains 32 pages]

The significance of fish handling, preservation and processing in the development of Nigeria inland fishery with special reference to Kainji Lake

The traditional approach to fish handling, preservation and processing technology in inland fishery is critically examined using the experience in Kainji Lake as a model. The need to uplift the fishermen technology is emphasized with the ultimate expectations of improvement in fish quality

Present status and potentials of inshore and offshore pelagic fisheries in the capture and processing industries of Nigeria

Apart from activities of some foreign-based vessels, commercial exploitation of pelagic fishery resources in Nigeria has been limited to inland and inshore waters. Estimated potential for the inshore pelagic fishery is 70,000-90,000 tonnes while the small pelagic resources in the near offshore as well as tuna and tuna-like fishes further offshore have potentials of about 10,000 metric tonnes each. Despite the abundance of tuna within and adjoining the Nigerian EEZ, and its importance in the international market, only foreign-flagged vessels take advantage. In addition, the inshore pelagic fisheries in Nigeria have for long remained underexploited. The most common processing method has remained the age-old traditional smoke-drying, which is inadequate resulting in colossal waste through denaturation and incessant infestations by insects and moulds among other causes. The use of modern smoking techniques coupled with effective distribution systems can undoubtedly reduce waste. However, these are often not within the reach of most artisanal processors. It is proposed that the organised private sector should invest on simple but proven processing equipment such as smoking kilns. The inshore pelagic fish species and other small fishes can sustain cottage canning industries sited in fishing villages/settlements while larger canning factories should be based on offshore resources. Modalities for successful investments are highlighted, while a major consideration is given to joint ventures