Gesture detection method based on acceleration compensation and intermetacarpal point detection

Abstract

The invention provides a gesture detection method based on acceleration compensation and intermetacarpal point detection. The method comprises the following steps of S1, image preprocessing: performing preprocessing on a gesture image captured by a camera; S2, calculation of an acceleration and an angle: calculating changes of the acceleration and the angle of a pixel point in the image; S3, illumination balance: compensating an illumination change by use of histogram equalization; S4, search of a gesture outline: searching a human hand outline by use of a connected region; and S5, gesture filtration: realizing filtration of a non-gesture part through positioning of a fingertip point and an intermetacarpal point. According to the scheme, acceleration data are collected and a tilt angle iscalculated by use of an acceleration sensor to eliminate the interference of shaking or movement of a head on extraction of a human hand coordinate; the gesture outline is search by use of the connected region, a fingertip point and an intermetacarpal point are searched to realize filtration of a non-gesture region, thereby reducing the interference brought by shaking of a body and improving the accuracy of gesture detection.

Description

technical field [0001] The invention relates to the field of human-computer interaction, in particular to a gesture detection method based on acceleration compensation and palm point detection. Background technique [0002] In order to help the disabled / elderly maintain communication and communication with the outside world, improve their independent living ability, and reduce the burden on the family and society, many scientists around the world have begun to explore new ways of human-computer interaction. The so-called interaction technology includes the interaction between human and actuators (such as robots, etc.) and the interaction between actuators and the environment. The significance of the former is that humans can implement planning and decision-making that are difficult for executive agencies in unknown or uncertain environments; the significance of the latter is that robots can complete tasks in harsh environments or remote environments that humans cannot reach....

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to overcome the deficiencies of the prior art, especially to solve the problem that in the existing computer vision-based gesture interaction technology, it is difficult to establish a general and efficient skin color model for skin colors with different colors, and it is difficult to solve the problem of lower detection rate caused by shaking of the human body area

Images