Method and device for detecting user head-lowering posture

Abstract

The invention discloses a method and a device for detecting a user head-lowering posture, and is applied to intelligent handheld devices. The method comprises the following steps: judging whether the user head-lowering posture is abnormal or not; if the user head-lowering posture is abnormal, reminding the user of the abnormity of the head-lowering posture to request the user to adjust the head-lowering posture and open a timer; performing voice announcement on the user if the timer is overtime and the user does not adjust the head-lowering posture; setting the intelligent handheld device into a non-working mode after the voice announcement is ended.

Description

technical field [0001] The invention relates to the technical field of sensors, in particular to a method and device for detecting a user's head-down posture. Background technique [0002] In recent years, smart handheld devices have flourished and gradually become an indispensable part of life. Every day, people spend more fragmentary time staring at the screens of mobile phones and tablet computers. More and more young people become members of the head-down group, and the younger age trend of this group is becoming more and more obvious. [0003] Now "head bowing syndrome" is not only a problem for adults, but also affects the healthy growth of teenagers and children. Staring at the mobile phone screen for a long time will cause excessive and strong visual stimulation, causing visual fatigue and vision loss; if you keep your head down and play with your mobile phone for a long time, it is easy to cause dislocation of the cervical spine joints, and at the same time, the so...

AI Technical Summary

Problems solved by technology

Staring at the mobile phone screen for a long time will cause excessive and strong visual stimulation, causing visual fatigue and vision loss; if you keep your head down and play with your mobile phone for a long time, it is easy to cause dislocation of the cervical spine joints, and at the same time, the soft tissues such as the muscles of the neck, shoulders and back are constantly in tension. Cervical spondylosis due to excessive fatigue due to poor posture; these bad postures will cause great damage to the cervical spine, and this harm is dozens of times higher than watching a computer; therefore, the prevention of bowing people is imminent

[0004] At present, there are some solutions for the relevant research on the gesture detection of handheld device users, such as: fixing the camera on a certain device, and performing gesture detection based on image recognition technology, and parameters such as the angle of the camera and the distance between the camera and the user need to be pre-installed. In this way, the user's head posture can be correctly estimated, but for mobile terminals such as smartphones, the distance between the man and the machine and the angle of the camera are unknown and change in real time, so this technology cannot be implemented on mobile devices; through The front camera captures the length and width of the face in the image to roughly estimate the distance from the face to the phone, but when the face is very close to the camera or in a dark environment, the camera cannot capture a complete face image, so it cannot correctly estimate the distance between the face and the phone. distance from the phone; adding two front-facing cameras for calibration not only increases the hardware cost of the phone, but also requires processing two channels of video image information at the same time, resulting in a relatively large amount of calculation for the algorithm, so it needs to be processed on the background server. Running on mobile devices requires a lot of CPU and power, which seriously affects the smoothness of system operation and greatly reduces the normal use time of smart handheld devices

[0005] The above-mentioned technology based on face image processing has defects in posture detection of smart handheld device users, and can only estimate the normal distance from the face to the mobile phone, but cannot estimate the tilt angle of the user's head and the short human-machine distance.

Images