A gesture recognition algorithm for smart gloves

This paper proposes a gesture recognition algorithm for smart gloves. The main innovations of the method are: (a) When the slope method is used to fit the curve, the two curve equations are assumed, and then the error size of the two curves is judged, and the curve where the smaller error is located is the fitted curve. (b) Jaccard coefficient is introduced into the gesture recognition algorithm, and by comparing the trend of two gestures in the same position segmentation interval, it judges whether they are the same gesture. Compared with the previous trend determination condition
^[1]
the determination condition is simpler and the output result is more intuitive. The experiment was conducted by this algorithm, and better experimental results were achieved.