A new hand shape positioning algorithm based on Voronoi diagram

For the problem of inaccurate accuracy caused by deformation region of hand shape contour in traditional hand shape recognition algorithms, this paper proposes a new hand shape positioning algorithm based on Voronoi diagram. The algorithm first extracts the hand shape contour. And using the geometric information of the finger, this algorithm accurately extract the central axis of the fingers based on Voronoi diagram. This paper also identifies the optimal value of the threshold of normalized feature deviation and overall feature deviation, etc. Finally, the extracted finger width feature is used to recognize and match. Under the constraint of the false rejection rate and the error acceptance rate, recognition rate as high as 98. 952%. And the algorithm has low complexity and high accuracy and stability.