An adaptive lectern tilt angle adjustment method and processing terminal

Abstract

The application discloses a self-adaptive podium tilt angle adjusting method and processing terminal, and belongs to the method which comprises the following steps: step 1: obtaining a current posture index r of a speaker in front of a podium; step 2: calculating a podium tilt angle alpha corresponding to the current posture index r by a corresponding formula, wherein the corresponding formula comprises parameters A, B and C, A represents a change amplitude adjusting coefficient of the podium tilt angle, C represents a desktop tilt angle when the speaker stands straight without bowing or lifting the head, B represents a response speed of the podium tilt angle after the response to the posture change of the speaker, the current posture index r of the speaker in front of the podium is measured in real time, the podium tilt angle alpha corresponding to the current posture index r is obtained according to the current posture index r, and then the podium is adjusted to the podium tilt angle alpha, and the current podium tilt angle adjustment is completed. The application can well adaptively adjust the podium tilt angle under different postures, and the comfort and experience are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of podium adjustment control, and particularly relates to a self-adaptive podium tilt angle adjustment method and a processing terminal. BACKGROUND

[0002] At present, the tilt angle of many podiums is fixed, that is, the tabletop tilt angle of the podium is fixed. The tabletop of the podium is usually used to place a display screen, books and other carriers for providing information display to a speaker. The fixed tilt angle is inconvenient for speakers of different heights, and the comfort and experience of the speaker are poor. Therefore, it is desirable to adjust the tilt angle according to the height of the speaker to better meet the needs of speakers of different heights.

[0003] In the prior art, some podiums can adjust the tilt angle. However, most of them are manually adjusted, which cannot meet the automatic adjustment of the speaker in different postures. Although a small part of them can automatically adjust the tilt angle, they still cannot quickly and accurately respond to the tilt angle adjustment according to the posture of the speaker. SUMMARY

[0004] In view of the deficiencies of the prior art, the purpose of the present application is to provide a self-adaptive podium tilt angle adjustment method and a processing terminal, which can solve the problems described in the background.

[0005] The technical solution for achieving the purpose of the present application is as follows: a self-adaptive podium tilt angle adjustment method, comprising the following steps:

[0006] Step 1: obtaining a current posture index r of a speaker in front of a podium;

[0007] Step 2: substituting the current posture index r into formula ① to calculate the tilt angle α of the podium:

[0008] α=A*(1-exp(-B*r))+C---①

[0009] In the formula, the parameters A, B and C are adjustable parameters, A represents the change amplitude adjustment coefficient of the tilt angle of the podium, C represents the tabletop tilt angle when the included angle is 0, that is, the speaker stands straight without bending forward or looking up, and B represents the response speed of the tilt angle of the podium after responding to the posture change of the speaker. The greater B is, the faster the response speed is, and the smaller B is, the slower the response speed is. 0

[0010] The current posture index r of the speaker in front of the podium is measured in real time, the tilt angle α of the podium corresponding to the current posture index r is calculated, and then the tilt angle α of the podium is adjusted to complete the current tilt angle adjustment of the podium.

[0011] Further, in step 1, the current posture index of the podium speaker is detected by the camera, the posture index includes head key point H, neck key point N and spine key point C, a vector HN representing the direction of head key point H and neck key point N is constructed, and a vector NC representing the direction of neck key point N and spine key point C is constructed, the included angle θ between the vector HN and the vector NC is calculated, and the included angle θ is converted into radian, and the radian is the current posture index r.

[0012] Further, the shoulder key point S is used to replace the spine key point C, and correspondingly, the vector NC is replaced by the vector NS.

[0013] Further, the included angle θ is calculated by the following arccosine function formula:

[0014] θ=arccos((HN·NC) / (|HN|·|NC|))

[0015] Further, a mapping relationship between the height of the speaker and the parameters A and C is established in advance, so as to adjust the corresponding podium inclination angle α according to different heights, so as to adapt to the inclination angle adjustment of different speaker heights.

[0016] Further, the parameter B is positively correlated with the radian r, that is, the larger the radian r is, the larger the parameter B is, and vice versa, the smaller the radian r is, the smaller the parameter B is.

[0017] Further, the parameters A, B and C are calculated by a sliding window method, and the sliding window method includes the following steps:

[0018] Step S1: Collect a sequence of included angles θ of a plurality of speakers in a period of time, and screen the current included angle θ sequence according to a preset sliding window size n, so as to continuously collect new posture data and discard the oldest posture data;

[0019] Step S2: Calculate the average value ave and the standard deviation SD of the included angle θ sequence based on the current included angle θ sequence, the average value can represent the overall posture of the speaker in the period of time, and the standard deviation represents the posture change amplitude of the speaker in the period of time;

[0020] Step S3: If the average value ave is greater than or equal to the first preset threshold, it is determined that the average value ave is large, and the value of the parameter A is increased to adapt to the current posture of the speaker being more bent; if the average value ave is less than the first preset threshold, it is determined that the average value ave is small, and the value of the parameter A is reduced to adapt to the current posture of the speaker being more straight,

[0021] If the standard deviation SD is greater than or equal to the second preset threshold and is considered large, then the value of parameter B is increased to improve the sensitivity of the response, allowing the podium to respond more quickly to changes in the speaker's posture. If the standard deviation SD is less than the second preset threshold and is considered small, then the value of parameter B is decreased to reduce the sensitivity of the response, causing the podium to adjust its tilt angle less frequently when the speaker's posture changes are relatively stable.

[0022] The parameter C is determined by taking into account both the mean (ave) and the standard deviation (SD).

[0023] Furthermore, it also includes filtering outliers in attitude indicators, including filtering out the angle θ sequence values ​​in attitude indicators, and using a median filtering method to filter out outliers in the angle θ sequence values.

[0024] Furthermore, a median filter with a window size of k is selected to filter the angle θ sequence values. For each data point of the current angle θ, its adjacent k-1 data points are taken, and then these k-1 data points are sorted. The median value after sorting is used as the filtered data point.

[0025] A processing terminal, comprising:

[0026] Memory, used to store program instructions;

[0027] A processor is configured to run the program instructions to perform the steps of the adaptive lectern tilt angle adjustment method.

[0028] The beneficial effects of this invention are as follows: This invention can effectively and adaptively adjust the tilt angle of the podium under different postures. Moreover, it can dynamically and adaptively adjust the tilt angle of the podium according to the actual usage scenario and the needs of different speakers, as well as the dynamic changes in the posture of different speakers or the same speaker. It can also avoid drastic changes in the tilt angle of the podium due to small changes in posture, thus adapting to the speaker and improving comfort and experience. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the process of the present invention;

[0030] Figure 2 This is a schematic diagram of the processing terminal. Detailed Implementation

[0031] The present invention will now be further described with reference to the accompanying drawings and specific embodiments:

[0032] like Figure 1 As shown, an adaptive lectern tilt angle adjustment method includes the following steps:

[0033] Step 1: Detect the current posture index of the podium speaker through the camera, the posture index includes head key point H, neck key point N and spine key point C, construct the vector HN representing the direction of head key point H and neck key point N and the vector NC representing the direction of neck key point N and spine key point C, and calculate the included angle θ between the vector HN and the vector NC.

[0034] Among them, the shoulder key point S can be used instead of the spine key point C, and correspondingly, the vector NC is replaced by the vector NS.

[0035] The included angle θ between the vector HN and the vector NC can be obtained by using the arccos calculation formula of the inverse cosine function, and the calculation formula of the inverse cosine function arccos is as follows:

[0036] θ=arccos((HN·NC) / (|HN|·|NC|))

[0037] The inverse cosine function calculation formula is to solve the included angle between two vectors, HN·NC represents the dot product of two vectors, and |HN| and |NC| represent the corresponding module length of each vector, that is, the absolute value.

[0038] Step 2: Convert the dimensional included angle θ to dimensionless radian r, that is, r=θ*π / 180°, for example, assuming θ=30°, then r=30°*π / 180°=0.52 radian. Then substitute the radian r into formula ① to calculate the podium tilt angle α:

[0039] α=A*(1-exp(-B·r))+C------①

[0040] In the formula, the parameters A, B and C are adjustable parameters, which can also be constants, A represents the change amplitude adjustment coefficient of the podium tilt angle, C represents the desk tilt angle when the included angle is 0, that is, the speaker stands straight without bending or lifting, and B represents the response speed of the podium tilt angle after responding to the posture change of the speaker, the greater B is, the faster the response speed is, the smaller B is, the slower the response speed is, and 0<B<1.

[0041] In formula ①, the current posture of the speaker in front of the podium can be measured in real time, the corresponding podium tilt angle α is obtained according to the current posture and the included angle θ, and then the podium is adjusted to the podium tilt angle α, so that the tilt angle of the corresponding podium can be dynamically adjusted according to the posture of the speaker, so as to adaptively adjust the tilt angle of the same speaker during the speech process, so as to meet the different heights of different speakers.

[0042] In an optional embodiment, by pre-establishing the mapping relationship between the height of the speaker and the parameter A and the parameter C, the corresponding lectern inclination angle a can be adjusted according to different heights even in the same posture (for example, the degree (angle) of bowing is consistent), so that the inclination angle adjustment can be adapted to different heights of the speaker.

[0043] In an optional embodiment, the parameter B is positively correlated with the radian r, that is, the larger the radian r is, the larger the parameter B is, and vice versa, the smaller the radian r is, the smaller the parameter B is.

[0044] In an optional embodiment, the parameter A, the parameter B and the parameter C are calculated by a sliding window method, which includes the following steps:

[0045] Step S1: Collect a sequence of angles θ of a plurality of speakers in a period of time, and screen the current angle θ sequence according to a preset sliding window size n, so as to continuously collect new posture data and discard the oldest posture data.

[0046] Step S2: Calculate the average value ave and the standard deviation SD of the angle θ sequence based on the current angle θ sequence, the average value can represent the overall posture of the speaker in the period of time, and the standard deviation can represent the posture change amplitude of the speaker in the period of time.

[0047] Step S3: If the average value ave is greater than or equal to the first preset threshold value, it is determined that the average value ave is large, and the value of the parameter A is increased, that is, the value of the parameter A corresponding to the current posture is larger than that of the parameter A corresponding to the last posture, so as to adapt to the posture of the speaker that is more bent; if the average value ave is less than the first preset threshold value, it is determined that the average value ave is small, and the value of the parameter A is reduced, that is, the value of the parameter A corresponding to the current posture is smaller than that of the parameter A corresponding to the last posture, so as to adapt to the posture of the speaker that is more straight.

[0048] If the standard deviation SD is greater than or equal to the second preset threshold value, it is determined that the standard deviation is large, and the value of the parameter B is increased, that is, the value of the parameter B corresponding to the current posture is larger than that of the parameter B corresponding to the last posture, so as to improve the sensitivity of the response, so that the lectern can respond to the posture change of the speaker more quickly; if the standard deviation SD is less than the second preset threshold value, it is determined that the standard deviation is small, and the value of the parameter B is reduced, that is, the value of the parameter B corresponding to the current posture is smaller than that of the parameter B corresponding to the last posture, so as to reduce the sensitivity of the response, so that the lectern will not frequently adjust the inclination angle when the posture change of the speaker is stable.

[0049] Wherein, the parameter C can be determined by comprehensively considering the average value ave and the standard deviation SD, so as to ensure that a good inclination angle can be provided under different postures, so as to provide a comfortable reading angle for the speaker to view.

[0050] In another optional embodiment, the abnormal values of the posture index are filtered, that is, the noise is filtered, including filtering the angle θ sequence values in the posture index. Wherein, the median filtering method can be used to filter the abnormal values in the angle θ sequence values. For example, a median filter with a window size of k is selected to filter the angle θ sequence values. For each current angle θ data point, take its adjacent k-1 data points, then sort the k-1 data points, and the middle value after sorting is taken as the filtered data point.

[0051] The present application can well adaptively adjust the podium inclination angle under different postures, and can dynamically and adaptively adjust the podium inclination angle according to the actual use scene, the needs of different speakers, or the posture dynamic change process of the same speaker, and can avoid the dramatic change of the podium inclination caused by the small change of the posture, so as to adapt to the speaker, and improve the comfort and experience.

[0052] As shown in Figure 2 The present application also provides a processing terminal 100, which comprises:

[0053] A memory 101 for storing program instructions;

[0054] A processor 102 for running the program instructions to execute the steps of the adaptive podium inclination angle adjustment method.

[0055] The embodiments disclosed in the specification are only an example of the one-sided features of the present application, and the protection scope of the present application is not limited to this embodiment. Any other functionally equivalent embodiments fall within the protection scope of the present application. For those skilled in the art, other various corresponding changes and deformations can be made according to the above described technical solutions and concepts, and all these changes and deformations should belong to the protection scope of the claims of the present application.

Claims

1. A method for adjusting the tilt angle of an adaptive lectern, characterized in that, Includes the following steps: Step 1: Obtain the current posture index r representing the speaker at the podium; Step 2: Substitute the current posture index r into formula ① to calculate the podium tilt angle α: α=A*(1-exp(-B·r))+C------① In the formula, parameters A, B, and C are all adjustable parameters. A represents the adjustment coefficient for the change range of the podium tilt angle; C represents the tabletop tilt angle when the included angle is 0, that is, when the speaker is standing upright without looking down or up; B represents the response speed of the podium to change its tilt angle in response to changes in the speaker's posture. The larger B is, the faster the response speed; the smaller B is, the slower the response speed; 0 < B < 1. The speaker's current posture index 'r' is measured in real time. Based on the corresponding podium tilt angle 'α', the podium is adjusted to the specified tilt angle 'α', completing the current podium tilt angle adjustment. In step 1, the current posture indicators of the speaker on the podium are detected by a camera. These indicators include head keypoint H, neck keypoint N, and spine keypoint C. Vectors HN representing the head keypoint H and the direction of the neck, and NC representing the direction of the neck keypoint N and the direction of the spine keypoint C, are constructed. The angle θ between vectors HN and NC is calculated and converted into radians. This radian is the current posture indicator r. Parameters A, B, and C are calculated using the sliding window method, which includes the following steps: Step S1: Collect the angle θ sequence values ​​of several speakers over a period of time, and filter the current angle θ sequence values ​​by sliding the window of the preset size n, so as to continuously collect new posture data and discard the oldest posture data. Step S2: Calculate the mean ave and standard deviation SD of the included angle θ sequence values ​​based on the current included angle θ sequence values. The mean value can represent the overall posture of the speaker during this period, and the standard deviation can represent the range of changes in the speaker's posture during this period. Step S3: If the average value ave is greater than or equal to the first preset threshold and is therefore considered large, then increase the value of parameter A to accommodate the speaker's current more hunched posture; if the average value ave is less than the first preset threshold and is therefore considered small, then decrease the value of parameter A to accommodate the speaker's current more upright posture. If the standard deviation SD is greater than or equal to the second preset threshold and is considered large, then the value of parameter B is increased to improve the sensitivity of the response, allowing the podium to respond more quickly to changes in the speaker's posture. If the standard deviation SD is less than the second preset threshold and is considered small, then the value of parameter B is decreased to reduce the sensitivity of the response, causing the podium to adjust its tilt angle less frequently when the speaker's posture changes are relatively stable. The parameter C is determined by taking into account both the mean (ave) and the standard deviation (SD).

2. The adaptive lectern tilt angle adjustment method according to claim 1, characterized in that, The shoulder key point S is used to replace the spine key point C, and the vector NC is replaced by the vector NS.

3. The adaptive lectern tilt angle adjustment method according to claim 1, characterized in that, The included angle θ is calculated using the following inverse cosine function formula: θ=arccos((HN·NC) / (|HN|·|NC|)).

4. The adaptive lectern tilt angle adjustment method according to claim 1, characterized in that, A mapping relationship between the speaker's height and parameters A and C is established in advance so that the corresponding lectern tilt angle α can be adjusted according to different heights, thereby adapting to the tilt angle adjustment for different speaker heights.

5. The adaptive lectern tilt angle adjustment method according to claim 4, characterized in that, The parameter B is positively correlated with the radian r. That is, the larger the radian r is, the larger the parameter B is, and vice versa.

6. The adaptive lectern tilt angle adjustment method according to claim 1, characterized in that, It also includes filtering outliers in attitude indicators, including filtering out the angle θ sequence values ​​in attitude indicators, and using median filtering to filter out outliers in the angle θ sequence values.

7. The adaptive lectern tilt angle adjustment method according to claim 6, characterized in that, A median filter with a window size of k is selected to filter the angle θ sequence values. For each data point of the current angle θ, its k-1 neighboring data points are taken, and then these k-1 data points are sorted. The median value after sorting is used as the filtered data point.

8. A processing terminal, characterized in that, It includes: Memory, used to store program instructions; A processor for running the program instructions to perform the steps of the adaptive lectern tilt angle adjustment method as described in any one of claims 1-7.

Images