The most popular systems for automatic sign language recognition are based on vision. They are user-friendly, but very sensitive to changes in regards to recording conditions. This article presents a description of the construction of a more robust system -an accelerometer glove -as well as its application in the recognition of sign language gestures. The basic data regarding inertial motion sensors and the design of the gesture acquisition system as well as project proposals are presented. The evaluation of the solution presents the results of the gesture recognition attempt by using a selected set of sign language gestures with a described method based on HMM and Parallel HMM approaches. The proposed usage of Parallel HMM for sensor-fusion modeling reduced the equal error rate by more than 60%, while preserving 99.75% recognition accuracy.

Maximal heart rate (HRmax) is associated mostly with age, but age alone explains the variance in HRmax to a limited degree and may not be adequate to predict HRmax in certain groups. The present study was carried out on 3374 healthy Caucasian, Polish men and women, clients of a sports clinic, mostly sportspeople, with a mean age of 36.57 years, body mass 74.54 kg, maximum oxygen uptake (VO2max, ml∗kg–1∗min–1) 50.07. Cardiopulmonary exercise tests (CPET) were carried out on treadmills or cycle ergometers to evaluate HRmax and VO2max. Linear, multiple linear, stepwise, Ridge and LASSO regression modeling were applied to establish the relationship between HRmax, age, fitness level, VO2max, body mass, age, testing modality and body mass index (BMI). Mean HRmax predictions calculated with 5 previously published formulae were evaluated in subgroups created according to all variables. HRmax was univariately explained by a 202.5–0.53∗age formula (R2 = 19.18). The weak relationship may be explained by the similar age with small standard deviation (SD). Multiple linear regression, stepwise and LASSO yielded an R2 of 0.224, while Ridge yielded R2 0.20. Previously published formulae were less precise in the more outlying groups of the studied population, overestimating HRmax in older age groups and underestimating in younger. The 202.5–0.53∗age formula developed in the present study was the best in the studied population, yielding lowest mean errors in most groups, suggesting it could be used in more active individuals. Tanaka’s formula offers the second best overall prediction, while the 220-age formula yields remarkably high mean errors of up to 9 bpm. In conclusion, adding the studied variables in multiple regression models improves the accuracy of prediction only slightly over age alone and is unlikely to be useful in clinical practice.

No abstract

This paper presents characteristics of model-based optimization methods utilized within the Generalized Self-Adapting Particle Swarm Optimization (GA– PSO) – a hybrid global optimization framework proposed by the authors. GAPSO has been designed as a generalization of a Particle Swarm Optimization (PSO) algorithm on the foundations of a large degree of independence of individual particles. GAPSO serves as a platform for studying optimization algorithms in the context of the following research hypothesis: (1) it is possible to improve the performance of an optimization algorithm through utilization of more function samples than standard PSO sample-based memory, (2) combining specialized sampling methods (i.e. PSO, Differential Evolution, model-based optimization) will result in a better algorithm performance than using each of them separately. The inclusion of model-based enhancements resulted in the necessity of extending the GAPSO framework by means of an external samples memory - this enhanced model is referred to as M-GAPSO in the paper.We investigate the features of two model-based optimizers: one utilizing a quadratic function and the other one utilizing a polynomial function. We analyze the conditions under which those model-based approaches provide an effective sampling strategy. Proposed model-based optimizers are evaluated on the functions from the COCO BBOB benchmark set.

No abstract