Research Article

Reduce Surface Electromyography Channels for Gesture Recognition by Multitask Sparse Representation and Minimum Redundancy Maximum Relevance

Figure 4

The classification accuracies using three channels selected by different methods (channels ② and ③ and channels ② and ⑧ are selected by MTSR, channels ① and ⑤ are selected by mRMR, and channels ⑤ and ⑧ are jointly selected by the two methods).