The invention discloses a finger
recognition algorithm, and relates to a human-computer
interaction interface, in particular to judgment of an operator and finger recognition in a multi-point touch interface and rich input primitives in the human-computer
interaction interface so as to solve the problems that an existing
recognition algorithm depends on specific equipment and is high in operation complexity, poor in real-time performance and the like. According to the
algorithm, fingers corresponding to touch points in a multi-point touch interface are recognized through information of natural gestures such as the relative orientation relation and the distance between the
finger touch points, and whether an operating hand is a left hand or a right hand is recognized according to positive and negative cross products of vectors between the
thumb touch point and the
index finger touch point and between the
thumb touch point and the
little finger touch point. According to the method, input primitives in the multi-point touch interface can be enriched, convenience is provided for people to realize'blind operation 'man-
machine interaction such as touch
typing text input and the like in the multi-point touch interface, convenience is provided for blind people to effectively utilize multi-point touch equipment and the like, and interaction performance is improved.