Gesture Recognition Algorithm Based On Space Sequence Recursive Model

In the field of human-computer interaction,gesture recognition has been paid more and more attention with its natural,rich and direct manner in recent years.To improve the accuracy and robustness,gesture recognition algorithm are proposed based on compression distance of recurrence patterns extracted from the perspective of space sequence.The main contents are listed:(1)To extract the target hand gesture region under complex background and to effectively solve problems of skin-color interference and varying illumination,we propose a hand gesture segmentation method based on statistical histogram.Considering the influence of the arm region on the gesture recognition,the automatic wrist positioning method is put forward based on the key point.The experimental results shows that the gesture image for the wrist bending and arbitrary angles placed can accurately locate the position of the wrist.(2)A new hand gesture feature extraction method is proposed based on spatial sequence,which deals with the problems of the complicated feature extraction method and the finite of hand characteristics,utilizes the properties that the same gestures have the same palm edge sequence expansion in rotation,moving,zooming.First,we locate the coordinates of the starting point of the palm of the hand,and then build edge sequences of palms following the change of the spatial position from the start point(3)In order to overcome the problem of unequal length of the edge sequences and the trajectories sequence,a hand gesture recognition model is established based on compression distance of recurrence patterns.The experimental results show that the proposed algorithm achieves high robustness in the case of rotation,translation and scaling,with 97%recognition accuracy in static gesture;with 97.48%recognition accuracy on the data set of 120 dynamic gestures,which already overtaken the current popular hand gesture recognition algorithms.