Towards Integration of Domain Knowledge-Guided Feature Engineering and Deep Feature Learning in Surface Electromyography-Based Hand Movement Recognition

<div>A schematic diagram of (a) DWT and (b) DWPT with the wavelet level of 3, where L↓2 denotes the half-band low-pass filter and H↓2 denotes the half-band high-pass filter. CAA denotes the approximate representation (scaling coefficients) of CA, CDA denotes the detailed representation (wavelet coefficients) of CA, and so on.</div>

Computational Intelligence and Neuroscience

fig1

Figure 1

Figure 1: Towards Integration of Domain Knowledge-Guided Feature Engineering and Deep Feature Learning in Surface Electromyography-Based Hand Movement Recognition 

Figure 1 | Towards Integration of Domain Knowledge-Guided Feature Engineering and Deep Feature Learning in Surface Electromyography-Based Hand Movement Recognition 