A human sitting posture detection method based on depth camera

Abstract

The invention relates to a method for detecting the sitting posture of a human body based on a depth camera, and relates to the field of image processing. The purpose of the invention is achieved by obtaining the depth image and color image of the back area of ​​the human body, obtaining the foreground area of ​​the image by combining the depth image and the color image, and using part of the foreground area to calculate whether the sitting posture of the human body is correct. The invention is a non-contact sitting posture monitoring method based on machine vision. The data of the back of the human body is collected by means of a depth camera, and images can be acquired in real time through the sitting posture monitoring algorithm, and the sitting posture monitoring can be completed automatically. The interactive software reminds the user to correct the sitting posture in time. The sitting posture of a person is closely related to the state of the spine. Therefore, collecting data on the back for sitting posture analysis is the most accurate, and at the same time, it can avoid the privacy problems caused by collecting human facial information.

Description

technical field [0001] The invention relates to the field of human body sitting posture monitoring, in particular to a non-contact human body sitting posture monitoring system. Background technique [0002] In recent years, people have been in a sitting position for more and more time, and many people spend more than 6 hours a day sitting and standing. Sitting for too long and sitting incorrectly for a long time will cause fatigue in the cervical and lumbar spine. In the long run, it will lead to cervical and lumbar diseases. According to statistics, more and more people are troubled by cervical spondylosis and lumbar disease. [0003] At present, the monitoring of sitting posture at home and abroad is mainly divided into two ways. The first is the contact-type sitting posture detection method, which uses sensors, pressure sensors, gyroscopes, etc. to collect sensor signals when the human body is sitting and standing, and obtain the sitting posture of the person after sig...

AI Technical Summary

Problems solved by technology

[0004] The contact sitting posture detection method needs to use a combination of multiple sensors. The sensors are distributed on the back and buttocks of the human body. After processing the raw data of the sensors through an external circuit, it is then transmitted to the host for sitting posture detection. The overall structure is relatively complicated.

The detection method of machine vision uses the camera to collect data, and then transfers the data to the computer for algorithm processing. It does not require external circuits, so the overall structure is relatively simple, but the collection of facial information may easily cause users to worry about privacy issues.

Images