A control method for intelligent mobile service robot based on head gesture recognition

Abstract

The invention provides a control method for an intelligent mobile service robot based on head posture recognition. The method uses a constrained local model algorithm to detect facial feature points, and then uses the geometric relationship of facial feature points to estimate the current head posture, so that patients can Use the head to control the movement of the intelligent mobile service robot. Its specific implementation includes the following steps: training head posture samples to obtain facial feature point shape templates and local templates of feature points; using Kinect to collect head images and find facial positions; detecting facial feature points by constrained local model algorithms; Use the detected feature points to estimate the current head pose; then send different control commands to the mobile service robot according to different head pose parameters. By using the method of the invention, the handicapped persons can also stably control the intelligent mobile service robot by using the head, which facilitates their travel.

Description

technical field [0001] The invention relates to the field of machine vision control, in particular to a control method for an intelligent mobile service robot based on head posture recognition, which is applied to an interactive system of an intelligent mobile service robot and is suitable for realizing a model based on local constraints on the interactive system of an intelligent mobile service robot (CLM) head pose real-time estimation and based on this to realize the control of the intelligent mobile service robot. Background technique [0002] With the development of society, the problem of population aging has become a common challenge faced by developed countries and some developing countries. The forecast shows that in the 20 years from 2015 to 2035, the proportion of China's elderly population will double to 20. %, many of them are old people with inconvenient activities due to diseases, and the number of disabled people with various disabilities caused by various di...

AI Technical Summary

Problems solved by technology

[0004] 1. The method of wearing a motion sensor. This method installs an accelerometer and a gyroscope sensor on the head of the test subject, and judges the head posture by receiving the sensor motion data. This method has high precision, but requires wearing a device, and the user experience is poor

[0005] 2. The LED calibration method, by wearing a headgear with a certain number of LEDs, using the camera to capture the image of the headgear, and then judging the current position according to the position of the LEDs, similarly, this method has high precision, but still needs to wear a device

[0006] 3. The machine vision method, which collects head images through the camera and judges the head posture through algorithms, is the current mainstream method, with zero contact with the test object and a better user experience. Estimating head pose (machine vision method), such as traditional template matching algorithm, mainstream random forest classifier, geometric relationship method of facial feature points, etc. The traditional CLM algorithm simply uses two-dimensional images, which has the problem of interference with different lighting conditions

Images