Hand function rehabilitation quantitative evaluation method based on 'finger adduction and abduction' actions of human hand

The invention relates to a hand function rehabilitation quantitative evaluation method based on 'finger adduction and abduction' actions of a human hand. Based upon computer vision and optical motioncapture technologies, and in combination and computer intelligent computing, information of skeleton and various articulation points of the human hand can be acquired in real time. Information of an affected hand and a healthy hand of a patient can be computed on the basis of a current rehabilitation evaluation standard, and an evaluated value is applied to quantitative description of wrist back stretching action of the patient in an entire rehabilitation training process. Meanwhile, in combination with such information as speed, acceleration and the like of hand motions, doctors can be assisted in disease diagnosis. According to the quantitative evaluation method provided by the invention, a hand function recovery degree of the patient can be objectively and effectively reflected, so as to bring about benefits fo