No abstract

The precise measurement and analysis of human movements is an essential step in biomechanical research used in sports or medicine. Measurement systems used for motion tracking should be non-invasive, safe to use, widely customisable and cost-efficient. In this study, complete design, development and evaluation of a high-speed optical motion tracking and analysis system is described. The system aims to analyse movements for sports and medical applications. The novelty of the proposed system is its design, which is based on visible light light-emitting diode (LED) markers, rather than infrared markers that are commonly used, and a pair of high-speed digital cameras. Calibration procedures and a super-resolution marker model are introduced, ensuring sub-pixel marker centre detection which results in higher three-dimensional reconstruction accuracy. Evaluation of the system included an accuracy test of the proposed system on static and moving objects with known dimensions, followed by analysis of kinematic data obtained in dynamic conditions while measuring human gait. The evaluation results are presented, and conclusions about system performance with possible improvements are discussed.

No abstract

No abstract

Predicting future trends in the stock market from time-series data is a challenging task due to its high non-linear nature caused by the complexity involved in the trading process. This paper emphasizes the importance of time-series data filtering when neural network models are used for stock market direction forecasting. Performances of three different neural network models are compared on raw data, processed data with simple moving average, and data filtered with discrete wavelet transformation. Applying wavelet transformation on input financial data as a processing step shows better results than the use of raw financial data or simple moving average. Also, among tested neural network models, the better results are obtained by using long short-term neural network then by using other neural network models.