Operatives training programmes in India

During the last few years great changes have taken place in the fishing industry as a result of which production of fish in the world has increased enormously. From an insignificant trade employing tools and methods of primitive nature, fishing in many countries has become an important industry utilizing complex modern vessels equipped with electronic equipment and operating in the high seas with highly mechanized fishing gear.

Competitive training in hidden Markov models

The use of hidden Markov models is placed in a connectionist framework, and an alternative approach to improving their ability to discriminate between classes is described. Using a network style of training, a measure of discrimination based on the a posteriori probability of state occupation is proposed, and the theory for its optimization using error back-propagation and gradient ascent is presented. The method is shown to be numerically well behaved, and results are presented which demonstrate that when using a simple threshold test on the probability of state occupation, the proposed optimization scheme leads to improved recognition performance.

MMI training for continuous phoneme recognition on the TIMIT database

This paper reports our experiences with a phoneme recognition system for the TIMIT database which uses multiple mixture continuous density monophone HMMs trained using MMI. A comprehensive set of results are presented comparing the ML and MMI training criteria for both diagonal and full covariance models. These results using simple monophone HMMs show clear performance gains achieved by MMI training, and are comparable to the best reported by others including those which use context-dependent models. In addition, the paper discusses a number of performance and implementation issues which are crucial to successful MMI training.

Wadge Bank trawl fishery studies. Pt. 3. Nature and composition of the resident population

The demersal stock of the Wadge Bank is known to be made up of two major groups, namely the resident and migrant stocks (Sivalingam 1966b). It is necessary to analyse the two groups separately in order to correctly interpret the changes in relative abundance of the demersal fish stocks of the Bank. The object of this paper is to present the nature of the resident stock and discuss its changes in composition from 1945 to 1960. That of the migrant population will be presented later. The significance of this analysis has been discussed earlier (Sivalingam 1966a).

Wadge Bank trawl fishery studies. Pt. 5. Rational exploitation of the resident demersal stock

The general history of the trawl fishery of the Wadge Bank off Cape Comorin, South India (Fig. 1), the nature and composition of its demersal fish population, and the present state of its fishery has been given by various authors (Malpas 1926, Pearson and Malpas 1926, Sivalingam and Medcof 1957, Medcof 1963, Mendis 1965a, 1965b, Sivalingam 1966a, 1966b, 1969a and 1969b). It has been shown earlier (Sivalingam and Medcof 1957, Sivalingam 1969a) that the Wadge Bank stock is made up of two groups. The resident stock which is present on the fishing grounds throughout the year and the migrant stock that appears on the fishing grounds only during the southwest monsoon months. The object of this paper is to discuss the effect of fishing on the resident stock between 1945 and 1962 and based on the information available; assess the maximum sustained yield of the resident stock. The "Bigfish" of the resident stock is the mainstay of the Wadge Bank trawl fishery (Sivalingam 1969a) and it will be shown that this stock has been overexploited from 1953 to 1957. The first sign of recovery was evident in 1960 and continued till 1962. The data since 1962 are not available to the author for analysis. It has been reported by Mendis (1965b) that considerable expansion of the trawler fleet was anticipated in 1966, and if so, the history of the fishery from 1953 to 1957 may be repeated. The assessment presented in this paper should form a rational basis for management programs.

Foot contact detection for sprint training

We introduce a new algorithm to automatically identify the time and pixel location of foot contact events in high speed video of sprinters. We use this information to autonomously synchronise and overlay multiple recorded performances to provide feedback to athletes and coaches during their training sessions. The algorithm exploits the variation in speed of different parts of the body during sprinting. We use an array of foreground accumulators to identify short-term static pixels and a temporal analysis of the associated static regions to identify foot contacts. We evaluated the technique using 13 videos of three sprinters. It successfully identifed 55 of the 56 contacts, with a mean localisation error of 1.39±1.05 pixels. Some videos were also seen to produce additional, spurious contacts. We present heuristics to help identify the true contacts. © 2011 Springer-Verlag Berlin Heidelberg.