METHOD AND DEVICE FOR CORRECTING POSTURE

Abstract

A device for reducing the health stress on a user is disclosed. The device has the following features: a) a sensor for detecting the posture of the user; b) a neural network that is trained for the following activities: i. categorizing the posture of the user in relation to categories of posture; ii. classifying the posture in relation to a degree of stress on the body of the user due to his posture; and c) an output unit for outputting feedback to the user in order to reduce the health stress.

Claims

1. A device for reducing the health stress on a user, comprising: a) a sensor for measuring the posture of the user; b) a neural network that is trained for the following activities: i. recognizing the posture of the user; ii. evaluating the posture in relation to the stress on the body of the user due to the user's posture; and c) an output unit for outputting feedback to the user in order to reduce the health stress.

2. The device according to claim 1, wherein, when evaluating the posture of the user, the neural network is trained to take into account the length of time the user assumes the posture.

3. The device according to claim 1, wherein the sensor has a depth sensor and/or a camera.

4. The device according to claim 1, wherein the neural network is embodied as a convolutional neural network.

5. The device according to claim 1, wherein the sensor has at least 10,000 pixels and/or measuring devices for measuring depth points, being more than 50,000 pixels and/or measuring devices for measuring depth points.

6. The device according to claim 1, wherein the output unit has a display.

7. The device according to claim 1, wherein the device is embodied to determine the posture of the user at a distance from the user, being at a distance of 5 cm to 4 m, the device being embodied to be arranged mechanically independent of the body of the user.

8. The device according to claim 1, wherein the device is embodied to measure distances of at least one meter from objects.

9. The device according to claim 1, wherein the device is embodied to output said feedback after various lengths of time in which the user assumes a posture, the length of time for outputting said feedback being shorter the higher the stress is on the body of the user due to the user's posture.

10. The device according to claim 1, wherein the device is embodied to provide said feedback with an ergonomic evaluation of a posture of the user.

11. The device according to claim 1, wherein the device is embodied to detect and evaluate a movement behavior of the user.

12. The device according to claim 11, wherein the device is embodied to output said feedback after various lengths of time in which the user exhibits the movement behavior, the length of time said feedback is output being shorter the higher the stress on the body of the user is due to the user's movement behavior.

13. The device according to claim 1, wherein the device is embodied to detect and evaluate a sitting posture and/or a standing posture of the user.

14. The device according to claim 1, wherein the device is embodied to detect that a user is staying within a predetermined distance of the device, and, to take into account the duration of this stay in order to determine the stress on the body of the user.

15. A method for reducing the health stress on a user with the device according to claim 1, comprising the steps of: i. measuring the posture of the user by using the sensor; ii. detecting the posture of the user by using the trained neural network; iii. evaluating the posture of the user in relation to the stress on the body of the user due to the user's posture by using the trained neural network; and iv. outputting feedback to the user to reduce the health stress by using an output unit.

16. The method according to claim 15, wherein the measuring the posture of the user by using the sensor and detecting the posture of the user by using the trained neural network at a distance from the user, is at the distance of 5 cm to 4 m, wherein the sensor and the neural network are arranged mechanically independent of the body of the user.

17. The method according to claim 15, wherein the evaluation of the posture of the user takes into account a length of time the user assumes the posture.

18. The method according to claim 15, wherein the user is given an alternative posture to the user's posture and/or in that the user is given a movement exercise and/or in that the user is provided a recommended drinking amount and/or in that the user is provided a recommendation to take a break and/or in that a standing posture is recommended to the user.

19. The method according to claim 15, wherein the user is regularly provided feedback regarding the user's posture.

20. The method according to claim 15, wherein the detection and evaluation is of the movement behavior of the user.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0044] FIG. 1 shows a schematic view of a device for reducing the health stress on a user due to his posture; and

[0045] FIG. 2 shows a schematic representation of a method for reducing this stress.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0046] The device 10 shown in FIG. 1 for improving the posture 12 of a user 14 sitting at a table T has a sensor 16 for measuring this posture 12. The sensor 16 transmits the measurement data to a trained neural network 18 which, from the measurement data from the sensor 16, assigns the posture 12 to a category of postures 12 and determines the length of time the user 14 assumes the posture 12. After determining the posture 12 and the associated length of time, the neural network determines the stress on the body 20 of the user 14. An output unit 22 then provides the user 14 feedback, for example in the form of an exercise or an alternative posture 12 for reducing the health stress. The output unit 22 is embodied in particular as a display 24.

[0047] FIG. 2 schematically illustrates a method 100 for reducing the stress on the body 20 of the user 14 due to his posture 12 (see FIG. 1). In a first step 102, the posture 12 of a user 14 is measured using a sensor 16. In a second step 104, a trained neural network 18 detects the posture 12 of the user 14. In a third step 106, the trained neural network 18 determines the length of time the user 14 assumes the posture 12. In a fourth step 108, the trained neural network 18 evaluates the posture 12 of the user 14 in relation to the stress on the body 20 of the user 14 due to his posture 12. In a fifth step 110, an output unit 22 provides feedback to the user 14 for reducing the health stress.

[0048] Taking all the figures in the drawings together, the invention relates to a device 10 for reducing the health stress on a user 14 and has the following features: [0049] A sensor 16 for detecting the posture 12 of the user 14; [0050] A neural network 18 which is trained for the following activities: [0051] i. Categorizing the posture 12 of the user 14 in relation to categories of the posture 12; [0052] ii. Classifying the posture 12 in relation to a degree of stress on the body 20 of the user 14 due to his posture 12; [0053] An output unit 22 for outputting feedback to the user 14 to reduce the degree of stress.

LIST OF REFERENCE SIGNS

[0054] 10 Device [0055] 12 Posture [0056] 14 User [0057] 16 Sensor [0058] 18 Trained neural network [0059] 20 Body [0060] 22 Output unit [0061] 24 Display [0062] 100 Method [0063] 102-110 Method steps