WO2018068403A1

WO2018068403A1 - Gesture recognition method and apparatus based on touch control remote controller

Info

Publication number: WO2018068403A1
Application number: PCT/CN2016/112399
Authority: WO
Inventors: 杨斌
Original assignee: 深圳Tcl新技术有限公司
Priority date: 2016-10-14
Filing date: 2016-12-27
Publication date: 2018-04-19
Also published as: CN106658123A

Abstract

A gesture recognition method and apparatus based on a touch control remote controller. The method comprises the following steps: during a touch operation process, performing smoothing processing on a touch motion track according to point location information, collected in real time, on the touch motion track, and calculating a touch distance according to the touch motion track after the smoothing processing so as to determine whether the touch distance reaches a threshold distance (S10); and when the touch distance reaches the threshold distance, performing gesture recognition according to the touch motion track after the smoothing processing, and outputting a gesture recognition result (S20). Gesture recognition is performed during a touch motion process, so that during the process of a user sliding a touch pad, a smart television synchronously responds to a touch action of a touch control remote controller, and the reaction is rapid.

Description

Gesture recognition method and device based on touch remote controller

Technical field

The present invention relates to the field of remote control technologies, and in particular, to a gesture recognition method and apparatus based on a touch remote controller.

Background technique

With the continuous development of science and technology, electronic technology has also developed rapidly, and there are more and more types of electronic products. People also enjoy the convenience brought by the development of science and technology. With the popularization and application of touch technology, more and more users are using touch devices, making human-computer interaction more human and intelligent. For example, the remote control that controls the smart TV has gradually replaced the traditional push button remote control with a touch remote controller.

However, in order to improve the accuracy of the touch gesture recognition, the existing touch-type remote controller often performs software algorithm processing when the user's finger raises the end sliding motion, and then sends a control signal indicating the recognized gesture motion to the control signal. Smart TV.

Since the touch remote control transmission time is generally 20 milliseconds, and the finger operation time varies from 180 milliseconds to 800 milliseconds, it takes at least 200 milliseconds from the user sliding the touchpad to the smart TV interface to respond to the operation command, if the finger manipulation is compared. Slow, the response time will be longer. The user must feel that the touch action of the smart TV in response to the remote control is slow, which seriously affects the user experience.

The above content is only used to assist in understanding the technical solutions of the present invention, and does not constitute an admission that the above is prior art.

Summary of the invention

The main purpose of the present invention is to provide a gesture recognition method and device based on a touch remote controller, which aims to solve the problem that the existing touch remote controller needs to wait for the user's finger to lift and end the sliding motion, and then perform the gesture motion recognition to make the smart television Responding to the problem that the remote control's touch action is slow.

To achieve the above objective, the present invention provides a gesture recognition method based on a touch remote controller, which includes the following steps:

Step A: During the touch operation, the touch motion track is smoothed according to the point information on the touch motion track collected in real time, and the touch distance is calculated according to the smoothed touch motion track to determine whether the touch distance reaches the threshold. distance;

Step B: When the touch distance reaches a threshold distance, perform gesture recognition according to the smoothed touch motion track, and output a gesture recognition result.

Preferably, the smoothing process includes linear smoothing, function fitting smoothing, and exponential smoothing.

Preferably, the step A comprises:

Step A1: When collecting the point information of the second point, determine a point from the line connecting the first point and the second point as a smoothed point according to a preset rule, and smooth the processed The point as the starting point for smoothing the trajectory;

Step A2: Determine a point from the line between the subsequently collected point and the adjacent previous point as a smoothed point according to a preset rule, and use the smoothed point as an intermediate point of the smoothed processing track;

Step A3: Calculate the distance from the intermediate point to the starting point, and use the distance as the touch distance;

In step A4, it is determined whether the touch distance reaches the threshold distance. When the touch distance does not reach the threshold distance, the process returns to step A2. When the touch distance reaches the threshold distance, the process proceeds to step B.

Preferably, after the step A3, the method further includes:

Step A5: When the distance between the intermediate point and the starting point is less than the threshold distance, predict the smoothing of the next point on the trajectory according to the intermediate point and the adjacent previous smoothed point;

Step A6. Calculate the distance from the predicted point to the starting point, and use the distance as the touch distance;

Step A 7. Determine whether the touch distance reaches the threshold distance. When the touch distance does not reach the threshold distance, return to step A2. When the touch distance reaches the threshold distance, proceed to step B.

Preferably, the step B comprises:

Step B1, calculating an angle formed by the line connecting the starting point to the predicted point and the reference line, and using the angle as a linear motion angle;

Step B2, when the linear motion angle is within the first angular range, determining that the touch motion is an upward shift, and when the linear motion angle is within a second angular range, determining that the touch motion is a left shift And determining that the touch motion is a downward shift when the linear motion angle is within a third angular range, and determining that the touch motion is a right shift when the linear motion angle is within a fourth angular range.

Preferably, the method further includes:

The starting point on the smoothing processing track is the center of the circle, and the threshold distance is used as the radius. When the intermediate point or the predicted point exceeds the range of the circle, it indicates that the touch distance reaches the threshold distance.

Preferably, the method further includes:

According to the position of the line connecting the first point and the center of the circle beyond the circle in the circle, the direction of the touch motion is identified as left shift, right shift, up shift or down shift, and the gesture recognition result is output.

Preferably, the method further includes:

Step C: When the touch distance reaches the threshold distance, the number of points from the starting point to the predicted point is N, and the N+1th point collected on the touch motion track is taken as the first point of the next piece of the touch motion track.

In addition, in order to achieve the above object, the present invention further provides a gesture recognition device based on a touch remote controller, including:

The calculation module is configured to smooth the touch motion track according to the point information on the touch motion track collected in real time during the touch operation, and calculate the touch distance according to the smoothed touch motion track to determine whether the touch distance is Reaching the threshold distance;

The identification module is configured to perform gesture recognition according to the smoothed touch motion track when the touch distance reaches a threshold distance, and output a gesture recognition result.

Preferably, the calculation module comprises:

a smoothing unit, configured to determine a point from the line connecting the first point and the second point as a smoothed point according to a preset rule when collecting the point information of the second point, and smooth the point The processed point serves as the starting point for smoothing the trajectory; and

Determining a point from the subsequently acquired point and the adjacent previous point as a smoothed point according to a preset rule, and using the smoothed point as an intermediate point of the smoothed processing track;

a calculating unit, configured to calculate a distance from the intermediate point to the starting point, and use the distance as the touch distance;

The comparing unit is configured to determine whether the touch distance reaches a threshold distance.

Preferably, the calculation module further includes a prediction unit;

The prediction unit is configured to: when the distance between the intermediate point and the starting point is less than the threshold distance, predict the smoothing of the next point on the trajectory according to the intermediate point and the adjacent previous smoothed point;

The calculating unit is further configured to calculate a distance from the predicted point to the starting point, and use the distance as the touch distance;

The comparing unit is further configured to determine whether the touch distance reaches a threshold distance.

Preferably, the identification module comprises:

An angle unit, configured to calculate an angle formed by a line connecting the starting point to the predicted point and a reference line, and using the angle as a linear motion angle;

a determining unit, configured to: when the linear motion angle is within a first angular range, determine that the touch motion is an upward shift, and when the linear motion angle is within a second angular range, determine that the touch motion is Moving to the left, when the linear motion angle is within the third angular range, determining that the touch motion is a downward shift, and when the linear motion angle is within the fourth angular range, determining that the touch motion is a right shift .

Preferably, the identification module is further configured to use a starting point on the smoothing processing track as a center, and a threshold distance as a radius as a circle. When the intermediate point or the predicted point exceeds a range of the circle, the touch distance is reached. Threshold distance.

Preferably, the identification module is further configured to identify, according to a position of the line connecting the first point and the center of the circle beyond the circle, the direction of the touch motion is left shift, right shift, up shift or down Move and output the gesture recognition result.

Preferably, the touch remote controller-based gesture recognition device further includes:

a starting point module, when the touch distance reaches a threshold distance, the number of points from the starting point to the predicted point is N, and the N+1th point collected on the touch motion track is taken as the first point of the next piece of the touch motion track .

In the touch operation process, the touch motion track is smoothed according to the point information on the touch motion track collected in real time, and the touch distance is calculated according to the smoothed touch motion track to determine whether the touch distance reaches the threshold. a distance; when the touch distance reaches a threshold distance, gesture recognition is performed according to the smoothed touch motion track, and the gesture recognition result is output. In the process of the touch motion, when the smoothed touch motion track reaches the threshold distance, the gesture recognition is performed, so that the smart TV synchronizes the touch action of the touch remote controller in the process of sliding the touch panel, and the response is rapid.

DRAWINGS

1 is a schematic flow chart of a first embodiment of a gesture recognition method based on a touch remote controller according to the present invention;

2 is a schematic flow chart of the refinement of the first embodiment of step S10 of FIG. 1;

FIG. 3 is a schematic diagram of a smoothing process of a touch motion track according to an embodiment of the present invention; FIG.

4 is a schematic flow chart of the second embodiment of step S10 of FIG. 1;

FIG. 5 is a schematic diagram of determining whether a touch distance reaches a threshold distance according to an embodiment of the present invention; FIG.

6 is a schematic diagram showing the refinement process of an embodiment of step S20 in FIG. 1;

7 is a schematic flow chart of a second embodiment of a gesture recognition method based on a touch remote controller according to the present invention;

8 is a schematic diagram of functional modules of a first embodiment of a gesture recognition device based on a touch remote controller according to the present invention;

9 is a schematic diagram of a refinement function module of the first embodiment of the computing module of FIG. 8;

10 is a schematic diagram of a refinement function module of the second embodiment of the computing module of FIG. 8;

11 is a schematic diagram of a refinement function module of an embodiment of the identification module of FIG. 8;

FIG. 12 is a schematic diagram of functional modules of a second embodiment of a gesture recognition device based on a touch remote controller of the present invention.

The implementation, functional features, and advantages of the present invention will be further described in conjunction with the embodiments.

detailed description

It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The main solution of the embodiment of the present invention is: during the touch operation, the touch motion track is smoothed according to the point information on the touch motion track collected in real time, and the touch distance is calculated according to the smoothed touch motion track to determine Whether the touch distance reaches a threshold distance; when the touch distance reaches a threshold distance, the gesture recognition is performed according to the smoothed touch motion track, and the gesture recognition result is output. In the process of the touch motion, when the smoothed touch motion track reaches the threshold distance, the gesture recognition is performed, so that the smart TV synchronizes the touch action of the touch remote controller in the process of sliding the touch panel, and the response is rapid.

Since the existing touch remote controller needs to wait for the user's finger to lift and end the sliding motion, the gesture motion recognition is performed, so that the touch action of the smart television in response to the touch remote controller is slow.

Based on the above problems, the present invention provides a gesture recognition method based on a touch remote controller.

1 is a schematic flow chart of a first embodiment of a gesture recognition method based on a touch remote controller according to the present invention.

In an embodiment, the touch remote controller-based gesture recognition method includes:

Step S10: During the touch operation, the touch motion track is smoothed according to the point information on the touch motion track collected in real time, and the touch distance is calculated according to the smoothed touch motion track to determine whether the touch distance reaches the threshold. distance;

The touch remote controller includes a touch panel, and the touch panel may also be a touch screen, and may be a touch panel capable of detecting touch points, such as a resistive touch panel or a capacitive inductive touch panel. When the touch remote controller is in a normal standby state, whenever the user touches the touchpad with a finger, the touch remote controller detects the number of touched points, and acquires the sliding of the user's finger according to the sequence of the detected touched points. Touch the motion track. Since the detected touched point may have interference or hopping, the obtained touch motion trajectory may be very unsmooth, and the speed and accuracy of the gesture recognition are greatly reduced. Therefore, an embodiment of the present invention smoothes the touch motion trajectory. To improve the speed and accuracy of gesture recognition. There are many ways to smooth the processing, for example, linear smoothing, function fitting smoothing, exponential smoothing, etc., and calculating the touch distance according to the smoothed touch motion trajectory to determine whether the distance of the touch motion reaches the distance at which gesture recognition can be performed.

Referring to FIG. 2, FIG. 2 is a schematic diagram of a refinement process of the first embodiment of step S10 of FIG.

Step S11, when collecting the point information of the second point, determining a point from the line connecting the first point and the second point as a smoothed point according to a preset rule, and smoothing the point The point as the starting point for smoothing the trajectory;

Step S12, determining, according to a preset rule, a point from the line of the subsequently collected point and the adjacent previous point as the smoothed point, and using the smoothed point as the intermediate point of the smoothing processing track;

Step S13, calculating a distance from the intermediate point to the starting point, and using the distance as the touch distance;

Step S14: determining whether the touch distance reaches the threshold distance. When the touch distance does not reach the threshold distance, the process returns to step S12. When the touch distance reaches the threshold distance, the process proceeds to step S20.

As shown in FIG. 3, in order to improve the smoothing processing speed, an embodiment of the present invention adopts linear smoothing as a preset rule of smoothing processing, and when the point information of two points is collected, the first point and the first point to be collected are collected. A point determined on the two points is used as a smoothed point, and the smoothed point is used as a starting point of the smoothing processing track; and the subsequently collected point is connected to the adjacent previous point. The point is used as a smoothed point, and the smoothed point is used as an intermediate point of the smoothed processed track; preferably, the midpoint of two adjacent points is used as the smoothed point. In the process of collecting the point information, the smoothing process is performed in real time, and when the intermediate point is obtained, the distance between the intermediate point and the starting point is calculated, and the distance is used as the touch distance to determine whether the distance of the touch motion is reachable for gesture recognition. When the touch distance does not reach the threshold distance, the subsequent collected points are further smoothed to obtain an intermediate point of the smoothed processed track, and the touch distance is calculated until the touch distance reaches the threshold distance, indicating that the condition for performing gesture recognition is reached. Gesture recognition is available.

Referring to FIG. 4, FIG. 4 is a schematic diagram of a refinement process of the second embodiment of step S10 of FIG. Based on the first embodiment of step S10, the step S10 further includes:

Step S15, when the distance between the intermediate point and the starting point is less than the threshold distance, predicting the next point on the trajectory according to the intermediate point and the adjacent previous smoothed point prediction;

Step S16, calculating a distance from the predicted point to the starting point, and using the distance as the touch distance;

Step S17: determining whether the touch distance reaches the threshold distance. When the touch distance does not reach the threshold distance, the process returns to step S12. When the touch distance reaches the threshold distance, the process proceeds to step S20.

As described above, the smoothed points are obtained by smoothing the adjacent two points. For example, when the points P0 and P1 are collected, the starting point Q0 of the smoothed processing track is obtained, and when the point P2 is collected, The intermediate point Q1 is obtained, and so on, when the point P8 is collected, the intermediate point Q7 can be obtained, that is, the time of obtaining the smoothed point always lags behind the time when the same number of point information is collected, because one or two are collected. At the time of the point, the prediction accuracy of the touch motion track is low. Therefore, it is necessary to determine the smoothed touch motion track according to the actually collected points for gesture recognition, and when at least two smoothed points are obtained, The next point on the trajectory can be smoothed according to the newly obtained intermediate point and the adjacent previous smoothed point, for example, the newly obtained intermediate point is connected with the adjacent previous smoothed point. In an embodiment of the present invention, as shown in FIG. 3, when the point P8 is collected, the intermediate point Q7 is obtained, and since the distance from the intermediate point Q7 to the starting point Q0 is less than the threshold distance, A point is taken as an predicted point Q8 on the extension line of the line connecting the intermediate point Q6 to the intermediate point Q7, and preferably, the distance from the predicted point Q8 to the intermediate point Q7 is the same as the distance from the intermediate point Q7 to the intermediate point Q6, and The distance from the predicted point Q8 to the starting point Q0 is calculated, and the distance is used as the touch distance to determine whether the distance of the touch motion reaches a distance at which the gesture recognition can be performed. When the predicted point-to-start distance is less than the threshold distance, the predicted point will be discarded. For example, when the predicted point Q8 to the starting point Q0 is less than the threshold distance, the predicted point Q8 is discarded, and the point P9 is collected. At this time, the intermediate point Q8 is obtained, and the touch point is continuously calculated and/or the next point on the smoothing processing track is predicted.

Step S20: When the touch distance reaches a threshold distance, perform gesture recognition according to the smoothed touch motion track, and output a gesture recognition result.

The threshold distance is a minimum distance required for gesture recognition. When the touch distance reaches a threshold distance, gesture recognition is performed according to the smoothed touch motion track. As shown in FIG. 5, the starting point on the smooth processing track is The center of the circle, with a threshold distance MOV_LIMIT (for example, 80 or 90 points, etc.) as a radius, when the intermediate point or the predicted point exceeds the range of the circle, that is, the touch distance reaches a threshold distance, of course, The first point of the acquisition is the center of the circle. According to the position of the line connecting the first point and the center of the circle beyond the circle in the circle, the direction of the touch motion is identified as left shift, right shift, up shift or down shift, and the gesture recognition result is output.

Referring to FIG. 6, FIG. 6 is a schematic diagram of a refinement process of an embodiment of step S20 in FIG.

Step S21, calculating an angle formed by a line connecting the starting point to the predicted point and a reference line, and using the angle as a linear motion angle;

Step S22, when the linear motion angle is within the first angular range, determining that the touch motion is an upward shift, and when the linear motion angle is within the second angular range, determining that the touch motion is a left shift And determining that the touch motion is a downward shift when the linear motion angle is within a third angular range, and determining that the touch motion is a right shift when the linear motion angle is within a fourth angular range.

When the touch distance reaches the threshold distance, the angle formed by the line connecting the starting point to the predicted point and the reference line is the angle of the linear motion during the touch operation, and the two-dimensional coordinates can be established on the touch panel and selected. A certain direction is a reference line, that is, a line representing 0° in two-dimensional coordinates, and an angle formed by a line connecting the starting point to the predicted point and the reference line is taken as a linear motion angle; of course, it may also be based on the starting point and the predicted point. The coordinate information can calculate the linear motion angle from the starting point to the predicted point; the direction of the touch motion is divided into left shift, right shift, up shift and down shift, and as can be seen from FIG. 5, the direction of the touch motion is arbitrary. However, it will always be in the range of 360°. Therefore, when performing azimuth division, the horizontal angle to the left and the opposite sides of the two sides will be shifted to the left, and the horizontal to the right and the two sides will be shifted to the right. And a certain angular range on both sides as the upward shift, the vertical downward and a certain angle range on both sides as the downward shift, an embodiment of the present invention indicates 0° or 360° horizontally to the right, and 45° to 135° as the upper shift, 135° to 225° is divided into left shift, 225° to 315° is divided into downward shift, and 0° to 45° and 315° to 360° are divided into right shifts, so that the touch motion in each direction can be recognized as Move left, right, up or down.

In the embodiment, when the point information on the touch motion track is collected, the touch motion track is smoothed in real time, and when the touch distance reaches the threshold distance, the gesture recognition is performed, so that the smart TV synchronizes during the sliding of the touch panel by the user. Responsive to the touch action of the touch remote control, the response is rapid.

Referring to FIG. 7, FIG. 7 is a schematic flowchart diagram of a second embodiment of a gesture recognition method based on a touch remote controller according to the present invention. Based on the first embodiment of the above-described touch remote control-based gesture recognition method, the method further includes:

Step S30, when the touch distance reaches the threshold distance, the number of points from the starting point to the predicted point is N, and the N+1th point collected on the touch motion track is taken as the first point of the next piece of the touch motion track.

When the touch distance reaches the threshold distance, the gesture recognition is performed according to the smoothed touch motion track, and the gesture recognition result is output, indicating that the gesture recognition is completed in one stage; in actual use, the touch motion formed by the user's one gesture motion The length of the track is different, and it is likely that it has not ended when the threshold distance is reached. In the process of the user's finger continuing to move, there may even be a turn or even a change of direction, in order to make the situation of the smart TV screen switch and the user's finger swipe. Consistently, it is necessary to continuously identify the touch motion; in an embodiment of the present invention, the number of points collected and the number of points obtained by the smoothing process are calculated. When the touch distance reaches the threshold distance, the number of points from the starting point to the predicted point is N. The N+1th point acquired on the touch motion track is used as the first point of the next touch motion track, and the touch motion track is smoothed, the gesture recognition, and the gesture recognition result are output. Of course, if the distance between the intermediate point obtained after the smoothing process and the starting point reaches the threshold distance, and there is no predicted point, the number of the starting point to the intermediate point is calculated as N, and the N+1th obtained on the touch motion track is obtained. The point is the first point of the next touch motion track.

In the whole process of the touch motion, the embodiment continuously smoothes the touch motion track, recognizes the gesture, and outputs the gesture recognition result, determines the gesture motion of the entire touch motion, and causes the smart TV to perform screen switching following the user's finger swipe.

In order to better illustrate the gesture recognition scheme of the above touch remote controller, the following will be specifically explained by an example.

The touch remote controller transmits the point information on the collected touch motion track to the set top box of the smart TV in real time, and the touch remote controller and the set top box transmit information through Bluetooth. Since the Bluetooth protocol specifies that information is sent every 7 ms, the touch is set. The control remote controller collects the point information on the touch motion track every 7ms. Since the point information is transmitted from the touch remote controller to the set top box, it takes about 1 to 2ms, and the touch remote controller needs to wait for the previous point information transmission to be completed. The point information can be collected and transmitted again, so the point information is collected once every 8 to 12 ms. Each time the point information is collected, there are touch points on the touch panel, indicating that the touch motion continues, and vice versa indicates that the touch motion ends. The touch motion track is smoothed in the set top box, and the midpoint of the collected point is used as the smoothed point, and each touch point on the touch panel may be coordinate-labeled in advance, and the smoothed point may also be touched by the touch panel. The coordinates above indicate that Table 1 below gives information on the acquired points and the smoothed points (including predicted points).

Table 1 Information on collected points and smoothed points (including predicted points)

采集的点Collected point	P0 P0	P1 P1	P2 P2	P3 P3	P4 P4	P5 P5	P6P6	P7 P7	P8 P8	P9 P9	P10 P10
X坐标 X coordinate	24twenty four	26 26	28 28	32 32	4444	80 80	120 120	140140	176 176	218 218	280 280
Y坐标 Y coordinate	222222	242 242	244 244	246 246	248248	242 242	238 238	226226	210 210	202 202	184 184
时间 Time	9466 9466	9476 9476	9486 9486	9498 9498	9523 9523	9535 9535	9546 9546	9550 9550	9558 9558	9564 9564	9575 9575
平滑处理后的点 Smoothed point	Q0Q0	Q1 Q1	Q2 Q2	Q3 Q3	Q4 Q4	Q5 Q5	Q6 Q6	Q7 Q7	Q8 Q8	Q9Q9	Q10 Q10
X坐标 X coordinate	2525	27 27	30 30	38 38	6262	100 100	138 138	158158	197 197	249 249
Y坐标 Y coordinate	232 232	243243	245 245	247 247	245 245	240 240	235235	218 218	206 206	193 193

The time in Table 1 is the system time when the point is collected on the touch panel. The value of the threshold distance MOV_LIMIT is set to 90. When P0 and P1 are acquired, the starting point Q0 of the smoothed processing track is obtained. When P2 is acquired, it is obtained. At the intermediate point Q1, the distance from the intermediate point Q1 to the starting point Q0 is calculated at the same time. From the above table, the distance from the intermediate point Q1 to the starting point Q0 is 11 < 90, taking the predicted point Q2 from the starting point Q0 to the extension point of the intermediate point Q1, the distance from the predicted point Q2 to the intermediate point Q1 is the same as the distance from the intermediate point Q1 to the starting point Q0, and calculated from the predicted point Q2 to the starting point The distance of Q0 is 22 < 90. As the touch motion continues, points P2, P3, P4, and P5 are continuously collected, and intermediate points Q2, Q3, and Q4 are obtained, and predicted points Q3, Q4, and Q5, due to intermediate points Q2, Q3, The distance between Q4 and predicted points Q3, Q4 and Q5 to the starting point Q0 is less than 90. When the point P6 is collected, the intermediate point Q5 is obtained, and the distance from the intermediate point Q5 to the starting point Q0 is still less than 90, so that gesture recognition is not possible yet. The coordinates of the point Q6 obtained from the intermediate points Q4 and Q5 are (138, 235), and the distance from the predicted point Q6 to the starting point Q0 is 113. > 90, the condition for performing gesture recognition is reached, and the linear motion angle from the starting point Q0 to the predicted point Q6 is calculated. According to the calculation formula of Sinα, the distance 113 from the predicted point Q6 to the starting point Q0 is the oblique side, and the predicted point Q6 is The distance 3 of the starting point Q0 along the Y axis is the opposite side of the right triangle, and the angle of α is calculated to be 1.5°, that is, the direction of the touch motion from P0 to P6 is right shift. The touch motion continues, and P7 is the first point of the next touch motion trajectory. The above process is repeated to obtain the starting point Q7, the intermediate point Q8 and the predicted point Q9 of the smoothing trajectory, due to the intermediate point Q8 and the predicted point Q9. The distance of Q7 is less than 90, so gesture recognition is not yet possible. When P10 is acquired, the intermediate point Q9 is obtained, and the distance from the intermediate point Q9 to the starting point Q7 is calculated. From the above table, the distance from the intermediate point Q9 to the starting point Q7 is 94. > 90, the condition for performing gesture recognition is reached, and the linear motion angle of the calculation starting point Q7 to the predicted point Q9 is 14.9°, that is, the direction of the touch motion from P7 to P10 is right shift, and P10 is the first touch motion track of the next segment. Repeat the above process.

The invention further provides a gesture recognition device based on a touch remote controller.

Referring to FIG. 8, FIG. 8 is a schematic diagram of functional modules of a first embodiment of a gesture recognition apparatus based on a touch remote controller according to the present invention.

In an embodiment, the touch remote controller-based gesture recognition apparatus includes: a calculation module 10 and an identification module 20.

The calculating module 10 is configured to perform smoothing on the touch motion track according to the point information on the touch motion track collected in real time during the touch operation, and calculate the touch distance according to the smoothed touch motion track to determine the Whether the touch distance reaches a threshold distance;

Referring to FIG. 9, FIG. 9 is a schematic diagram of a refinement function module of the first embodiment of the computing module 10 of FIG. The calculation module 10 includes a smoothing unit 11, a calculating unit 12, and a comparing unit 13.

The smoothing unit 11 is configured to determine a point from the line connecting the first point and the second point as a smoothed point according to a preset rule when the point information of the second point is collected, and Using the smoothed point as a starting point of the smoothing track; and

The calculating unit 12 is configured to calculate a distance from the intermediate point to the starting point, and use the distance as the touch distance;

The comparing unit 13 is configured to determine whether the touch distance reaches a threshold distance.

Referring to FIG. 10, FIG. 10 is a schematic diagram of a refinement function module of the second embodiment of the computing module 10 of FIG. Based on the first embodiment of the computing module 10 described above, the computing module 10 further includes: a prediction unit 14.

The prediction unit 14 is configured to: when the distance between the intermediate point and the starting point is less than the threshold distance, predict the smoothing of the next point on the trajectory according to the intermediate point and the adjacent previous smoothed point;

The calculating unit 12 is further configured to calculate a distance from the predicted point to the starting point, and use the distance as the touch distance;

The comparing unit 13 is further configured to determine whether the touch distance reaches a threshold distance.

The identification module 20 is configured to perform gesture recognition according to the smoothed touch motion track when the touch distance reaches a threshold distance, and output a gesture recognition result.

Referring to FIG. 11, FIG. 11 is a schematic diagram of a refinement function module of an embodiment of the identification module 20 of FIG. The identification module 20 includes an angle unit 21 and a determination unit 22.

The angle unit 21 is configured to calculate an angle formed by a line connecting the starting point to the predicted point and a reference line, and use the angle as a linear motion angle;

The determining unit 22 is configured to: when the linear motion angle is within a first angular range, determine that the touch motion is upward, and when the linear motion angle is within a second angular range, determine the The touch motion is a left shift, and when the linear motion angle is within a third angle range, determining that the touch motion is a downward shift, and when the linear motion angle is within a fourth angular range, determining the touch motion Move to the right.

In the embodiment, when the point information on the touch motion track is collected, the touch motion track is smoothed in real time, and when the touch distance reaches the threshold distance, the gesture recognition is performed, and the gesture recognition result is output, so that the user slides the touch panel. In the middle, the smart TV immediately responds to the touch action of the touch remote controller and responds quickly.

Referring to FIG. 12, FIG. 12 is a schematic diagram of functional modules of a second embodiment of a gesture recognition device based on a touch remote controller according to the present invention. The gesture remote control based gesture recognition device further includes a starting point module 30.

The starting point module 30 is configured to calculate, when the touch distance reaches a threshold distance, the number of points from the starting point to the predicted point is N, and the N+1 points collected on the touch motion track are used as the one point.

The above are only the preferred embodiments of the present invention, and are not intended to limit the scope of the invention, and the equivalent structure or equivalent process transformations made by the description of the present invention and the drawings are directly or indirectly applied to other related technical fields. The same is included in the scope of patent protection of the present invention.

Claims

A gesture recognition method based on a touch remote controller, comprising the following steps:

Step A: During the touch operation, the touch motion track is smoothed according to the point information on the touch motion track collected in real time, and the touch distance is calculated according to the smoothed touch motion track to determine whether the touch distance reaches the threshold. distance;

Step B: When the touch distance reaches a threshold distance, perform gesture recognition according to the smoothed touch motion track, and output a gesture recognition result.
The touch remote controller-based gesture recognition method according to claim 1, wherein the smoothing processing comprises linear smoothing, function fitting smoothing, and exponential smoothing.
The touch remote controller-based gesture recognition method according to claim 1, wherein the step A comprises:

Step A1: When collecting the point information of the second point, determine a point from the line connecting the first point and the second point as a smoothed point according to a preset rule, and smooth the processed The point as the starting point for smoothing the trajectory;

Step A2: Determine a point from the line between the subsequently collected point and the adjacent previous point as a smoothed point according to a preset rule, and use the smoothed point as an intermediate point of the smoothed processing track;

Step A3: Calculate the distance from the intermediate point to the starting point, and use the distance as the touch distance;

In step A4, it is determined whether the touch distance reaches the threshold distance. When the touch distance does not reach the threshold distance, the process returns to step A2. When the touch distance reaches the threshold distance, the process proceeds to step B.
The gesture remote control-based gesture recognition method of claim 3, further comprising: after the step A3,

Step A5: When the distance between the intermediate point and the starting point is less than the threshold distance, predict the smoothing of the next point on the trajectory according to the intermediate point and the adjacent previous smoothed point;

Step A6: calculating a distance from the predicted point to the starting point, and using the distance as the touch distance;

In step A7, it is determined whether the touch distance reaches the threshold distance. When the touch distance does not reach the threshold distance, the process returns to step A2. When the touch distance reaches the threshold distance, the process proceeds to step B.
The touch remote controller-based gesture recognition method according to claim 4, wherein the step B comprises:

Step B1, calculating an angle formed by the line connecting the starting point to the predicted point and the reference line, and using the angle as a linear motion angle;

Step B2, when the linear motion angle is within the first angular range, determining that the touch motion is an upward shift, and when the linear motion angle is within a second angular range, determining that the touch motion is a left shift And determining that the touch motion is a downward shift when the linear motion angle is within a third angular range, and determining that the touch motion is a right shift when the linear motion angle is within a fourth angular range.
The touch remote controller-based gesture recognition method according to claim 4, wherein the method further comprises:

The starting point on the smoothing processing track is the center of the circle, and the threshold distance is used as the radius. When the intermediate point or the predicted point exceeds the range of the circle, it indicates that the touch distance reaches the threshold distance.
The touch remote controller-based gesture recognition method of claim 6, wherein the method further comprises:

According to the position of the line connecting the first point and the center of the circle beyond the circle in the circle, the direction of the touch motion is identified as left shift, right shift, up shift or down shift, and the gesture recognition result is output.
The touch remote controller-based gesture recognition method according to claim 4, wherein the method further comprises:

Step C: When the touch distance reaches the threshold distance, the number of points from the starting point to the predicted point is N, and the N+1th point collected on the touch motion track is taken as the first point of the next piece of the touch motion track.
A gesture recognition device based on a touch remote controller, comprising:

The calculation module is configured to smooth the touch motion track according to the point information on the touch motion track collected in real time during the touch operation, and calculate the touch distance according to the smoothed touch motion track to determine whether the touch distance is Reaching the threshold distance;

The identification module is configured to perform gesture recognition according to the smoothed touch motion track when the touch distance reaches a threshold distance, and output a gesture recognition result.
The gesture remote control-based gesture recognition apparatus according to claim 9, wherein the smoothing processing comprises linear smoothing, function fitting smoothing, and exponential smoothing.
The touch control device-based gesture recognition device of claim 9, wherein the calculation module comprises:

a smoothing unit, configured to determine a point from the line connecting the first point and the second point as a smoothed point according to a preset rule when collecting the point information of the second point, and smooth the point The processed point serves as the starting point for smoothing the trajectory; and

Determining a point from the subsequently acquired point and the adjacent previous point as a smoothed point according to a preset rule, and using the smoothed point as an intermediate point of the smoothed processing track;

a calculating unit, configured to calculate a distance from the intermediate point to the starting point, and use the distance as the touch distance;

The comparing unit is configured to determine whether the touch distance reaches a threshold distance.
The touch remote controller-based gesture recognition apparatus according to claim 11, wherein the calculation module further comprises a prediction unit;

The prediction unit is configured to: when the distance between the intermediate point and the starting point is less than the threshold distance, predict the smoothing of the next point on the trajectory according to the intermediate point and the adjacent previous smoothed point;

The calculating unit is further configured to calculate a distance from the predicted point to the starting point, and use the distance as the touch distance;

The comparing unit is further configured to determine whether the touch distance reaches a threshold distance.
The touch recognition device-based gesture recognition device of claim 12, wherein the identification module comprises:

An angle unit, configured to calculate an angle formed by a line connecting the starting point to the predicted point and a reference line, and using the angle as a linear motion angle;

a determining unit, configured to: when the linear motion angle is within a first angular range, determine that the touch motion is an upward shift, and when the linear motion angle is within a second angular range, determine that the touch motion is Moving to the left, when the linear motion angle is within the third angular range, determining that the touch motion is a downward shift, and when the linear motion angle is within the fourth angular range, determining that the touch motion is a right shift .
The touch recognition device-based gesture recognition device according to claim 12, wherein the identification module is further configured to use a starting point on the smoothing processing track as a center and a threshold distance as a radius to be a circle. Or when the predicted point exceeds the range of the circle, it indicates that the touch distance reaches a threshold distance.
The gesture recognition device based on the touch remote controller according to claim 14, wherein the identification module is further configured to position the line in the circle according to the line connecting the first point and the center of the circle beyond the circle The direction in which the touch motion is recognized is left shifted, right shifted, moved up or down, and the gesture recognition result is output.
The gesture remote control-based gesture recognition apparatus of claim 12, wherein the gesture remote control-based gesture recognition apparatus further comprises:

a starting point module, when the touch distance reaches a threshold distance, the number of points from the starting point to the predicted point is N, and the N+1th point collected on the touch motion track is taken as the first point of the next piece of the touch motion track .