Real-Time and Automatic System for Performance Evaluation of Karate Skills Using Motion Capture Sensors and Continuous Wavelet Transform

Algorithm 1

Building the standard model.

(1)	for p in players do
(2)	for j in joints do
(3)	for a in angles do
(4)	peaks = CWT(T(p, j, a))
(5)	inverted_T =
(6)	valleys = CWT (T(p, j, a))
(7)	main_points.add (peaks)
(8)	main_points.add (valleys)
(9)	end
(10)	end
(11)	end
(12)	for p in players do
(13)	for j in joints do
(14)	for a in angles do
(15)	T_normalized (p, j, a) = normalize (main_points, T(p, j, a))
(16)	end
(17)	end
(18)	end
(19)	min = 361
(20)	max = −361
(21)	mean = 0
(22)	for k = 0 to T_normalized[0].size() do
(23)	min [k] = min (T_normalized[k][0:T_normalized.size()])
(24)	max [k] = max (T_normalized[k][0:T_normalized.size()])
(25)	mean [k] = mean (T_normalized[i][0:T_normalized.size()])
(26)	std [k] = STD (T_normalized[i][0:T_normalized.size()])
(27)	end