Table 1 Data collection and preprocessing specifications.

From: Intelligent optimization of track and field teaching using machine learning and wearable sensors

Data type

Primary source

Sampling rate

Preprocessing methods

Acceleration

Wearable IMUs

200 Hz

Butterworth filtering, gravity compensation, anatomical alignment

Angular velocity

Wearable IMUs

200 Hz

Drift correction, anatomical reference transformation

Video

Multi-angle HD cameras

120 fps

Background subtraction, human pose estimation, trajectory extraction

Force/pressure

Instrumented surfaces

1000 Hz

Baseline calibration, impulse identification, spatial normalization

Performance metrics

Electronic timing systems

Event-based

Statistical validation, outlier detection

Environmental

Weather station

0.1 Hz

Temporal interpolation, contextual annotation