Kinect based traffic police gesture recognition method

Abstract

The present invention relates to a Kinect based traffic police gesture recognition method, and belongs to the field of virtual simulation technologies. The method comprises: first, using Kinect to collect data, and preprocessing the data, to obtain police gesture data; then performing feature extraction on the gesture data, to obtain a movement trajectory T of each bone node; then calculating a weight w of each bone node according to the movement trajectory T of each bone node; extracting a movement trajectory S corresponding to each bone node in each pre-stored standard gesture; then separately calculating similarity d between the movement trajectory T of a bone node in a to-be-recognized gesture and the movement trajectory S corresponding to a bone node in the standard gesture by using a DTW based trajectory comparison algorithm; finally calculating similarity delta between the to-be-recognized gesture and each standard gesture according to a value d and a value w; and using a standard gesture corresponding to a minimum delta as a recognition result. Compared with the prior art, the Kinect based traffic police gesture recognition method has advantages of being low in costs, high in accuracy, and fast in operation speed.

Description

technical field [0001] The invention relates to a gesture recognition technology, in particular to a Kinect-based traffic police gesture recognition method, which belongs to the field of virtual simulation technology. Background technique [0002] At present, the research on the recognition technology of traffic police command gestures still stays at the stage of traditional image recognition methods. According to the different methods of gesture collection, it can be divided into two types. The first is to wear data clothes on traffic police or A way for motion capture devices such as data gloves to obtain gesture data. This method of recognition works well due to the high precision data that motion capture devices can obtain. However, this type of instrument is often expensive, and this method requires a certain number of sensors to be worn on the user's body, which reduces the user's comfort and even affects the quality of the action. The second method is to obtain traf...

AI Technical Summary

Problems solved by technology

This method can greatly reduce the restrictions on traffic police command gestures, and the naturalness and portability are better. However, this method also has its disadvantages, such as: a large amount of gesture data, complex processing methods, and the recognition efficiency and accuracy are not high enough.

Images