Gesture recognition system based on CNN-IndRNN and OpenBCI

Abstract

Surface electromyography (sEMG), as a key technology of non-invasive muscle computer interface, is an important method of human-computer interaction. We proposed a CNN-IndRNN (Convolutional Neural Network-Independent Recurrent Neural Network) hybrid algorithm to analyse sEMG signals and classify hand gestures. Ninapro’s dataset of 10 volunteers was used to develop the model, and by using only one time-domain feature (root mean square of sEMG), an average accuracy of 87.43% on 18 gestures is achieved. The proposed algorithm obtains a state-of-the-art classification performance with a significantly reduced model. In order to verify the robustness of the CNN-IndRNN model, a compact real¬time recognition system was constructed. The system was based on open-source hardware (OpenBCI) and a custom Python-based software. Results show that the 10-subject rock-paper-scissors gesture recognition accuracy reaches 99.1%.

A real-time basketball referee gesture recognition system based on multi-scale spatio-temporal features

“…Wu, N. et al proposed a hybrid CNN-IndRNNN (Convolutional Neural Network-Independent Recurrent Neural Network) algorithm to analyze the surface electromyographic signals and categorize the gestures. To test the model's robustness, a compact real-time recognition system was constructed, and the results indicated that the recognition accuracy reached 92% [13]. Tu, P and other scholars based on the markerless gesture segmentation algorithm based on palm neighborhood and the threshold detection algorithm based on palm contour to identify the operator's gestures and motion trajectories, which are transformed into the movements of the robotic arm and the model is verified by simulation experiments to have the excellent performances of simple operation, fast response speed and accuracy [14].…”