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

<div>A schematic drawing demonstrating the progressive fusion module in our proposed PFNet. The boxes marked with “FC” denote the fully connected layers, and the numbers before “FC” denote their number of hidden units. The 3-stage progressive fusion process is highlighted by yellow boxes, and the 1st and 2nd subnetworks for feature-level fusion are marked with blue and red lines, respectively.</div>

Computational Intelligence and Neuroscience

fig3

Figure 3

Figure 3: Towards Integration of Domain Knowledge-Guided Feature Engineering and Deep Feature Learning in Surface Electromyography-Based Hand Movement Recognition 

Figure 3 | Towards Integration of Domain Knowledge-Guided Feature Engineering and Deep Feature Learning in Surface Electromyography-Based Hand Movement Recognition 