Wearable Inertial Sensors for Daily Activity Analysis Based on Adam Optimization and the Maximum Entropy Markov Model

. 2020 May 20;22(5):579. doi: 10.3390/e22050579

Algorithm 1: Inertial Signal Features Computation

Input: Acc = Accelerometer (x,y,z), Gyr = Gyroscope (x,y,z) and SR = Sampling Rate (100 Hz)

Output: Multifused feature vectors

(u_{1}, u_{2}, u_{3} \dots \dots u_{n})

f e a t u r e v e c t o r s \leftarrow

[]

s a m p l e s i g n a l \leftarrow

GetSampleSignal()

Overlap

\leftarrow

GetOverlappingTime()

Procedure HAR(Acc,Gyr,SR)

M u l t i F u s e d V e c t o r \leftarrow

[]

DenoiseData

\leftarrow

MedianFilter(Acc,Gyr)

SampledData(DenoiseData,SR)

While exit condition not satisfied do

[m i n, m a x, m e a n, v a r i a n c e] \leftarrow

ExtractStatisticalFeatures(sample data)

[L B P F e a t u r e s] \leftarrow

ExtractLocalBinaryPatternFeatures(sample signal)

[W H T, C Z T, H T] \leftarrow

ExtractWaveletFeatures(sampledsignal)

M u l t i F u s e d V e c t o r \leftarrow

[ min, max, mean, variance, LBP, WHT, CZT, HT]

Return MultiFusedVector