ABOUT TWO THE LEAST-SQUARES METHOD MODIFICATIONS FOR LOST DATA RECOVERY IN VIDEOANALYSIS SYSTEM BASED ON ACCELEROMETER DATA

Авторы: Bobilev A.N., Bolotin Yu V., Voronov A.V., Kruchinin P.A.
Журнал: Russian Journal of Biomechanics
Том: 16
Номер: 1
Год издания: 2012
Издательство: Perm National Research Polytechnic University
Местоположение издательства: Perm, Russia
Первая страница: 80
Последняя страница: 91
Аннотация: In this study, we discuss the lost data recovery in videoanalysis of human movement based on measures of two-component accelerometer. The algorithm uses mathematical model of movement and consists of two stages: identification of unknown parameters and proper stage of recovery of lost information. Both problems are reduced to solution of overdetermined systems of linear equations. Two approaches to recovery problem solution are compared. The first one uses traditional non-recurrent procedure of the least-squares method The second one uses the procedure of suboptimal algorithm of smoothing based on Kalman filtering. These algorithms are applied to recovery of lost values of the angle between the shin of a person and the horizontal
Добавил в систему: Болотин Юрий Владимирович

	ИСТИНА	Войти в систему Регистрация
	ФНКЦ РР
	Главная Поиск Статистика О проекте Помощь

ИСТИНА