Action detection method, system, apparatus and computer readable storage medium

Abstract

The application provides a motion detection method, system, device and computer readable storage medium. The method comprises the following steps: continuously acquiring channel impulse response data collected by multiple ultra-wideband radars; for each ultra-wideband radar, determining a candidate motion corresponding to the channel impulse response data based on the relative position between the ultra-wideband radar and the tail door; if the channel impulse response data collected by all the ultra-wideband radars meet the recognition condition of the same candidate motion, the candidate motion is determined as a target motion. The generated kicking motion near the tail door is detected by setting multiple ultra-wideband radars, and when at least two ultra-wideband radars simultaneously detect the same candidate motion, it is determined that the target motion is detected, so that the motion misrecognition caused by environmental interference can be reduced, and the detection accuracy of the kicking motion is improved.

Description

Technical Field

[0001] This invention relates to the field of vehicles, and more particularly to a motion detection method, system, device, and computer-readable storage medium. Background Technology

[0002] With the trend of automotive intelligence, intelligent tailgate opening has become a common and important requirement. When users are carrying heavy items and cannot free their hands, the foot-operated tailgate function allows users to control the tailgate opening with leg movements, which is extremely convenient.

[0003] In the prior art, the detection of kicking motions used to control the tailgate function is usually achieved by distance detection, such as capacitance detection. However, the environment in which vehicles are used is complex, and this method is easily affected by environmental factors, resulting in low accuracy in detecting kicking motions. Summary of the Invention

[0004] The main objective of this invention is to provide a motion detection method, system, device, and computer-readable storage medium, aiming to solve the problem of low detection accuracy of kicking motions in the prior art.

[0005] This invention provides an action detection method, the method comprising the following steps: The motion detection method includes: Continuously acquire channel impulse response data collected by multiple ultra-wideband radars; For each of the ultra-wideband radars, a candidate action corresponding to the channel impulse response data is determined based on the relative position between the ultra-wideband radar and the target detection area. If the channel impulse response data acquired by at least two of the ultra-wideband radars meet the identification conditions of the same candidate action, then the candidate action is determined as the target action to be detected.

[0006] Optionally, the target detection area is the tailgate of the vehicle, and the number of ultra-wideband radars is 2. The ultra-wideband radars include a first ultra-wideband radar and a second ultra-wideband radar, which are respectively disposed on both sides of the tailgate of the vehicle.

[0007] Optionally, the step of ensuring that the channel impulse response data acquired by at least two of the ultra-wideband radars satisfy the identification condition of the same candidate action includes: If a candidate action is detected based on the channel impulse response data collected by any of the ultra-wideband radars, a second detection region is determined, wherein the second detection region is the region where the first detection region of the ultra-wideband radar that detected the candidate action intersects with the first detection regions of other ultra-wideband radars. Based on the detection data of each of the ultra-wideband radars for the corresponding second detection area, it is determined whether there are at least two ultra-wideband radars that collect channel impulse response data that satisfy the identification conditions of the same candidate action.

[0008] The step of determining whether there are at least two channels impulse response data collected by the ultra-wideband radars that satisfy the identification condition of the same candidate action based on the detection data of the second detection area by each of the ultra-wideband radars includes: The arrangement of motion features obtained by each of the ultra-wideband radars detecting the corresponding second detection area is acquired; For each of the ultra-wideband radars, determine whether the arrangement of motion features satisfies the preset feature arrangement corresponding to the candidate action; If the motion feature arrangements of at least two of the ultra-wideband radars both satisfy the preset feature arrangement corresponding to the candidate action, then it is determined that the channel impulse response data collected by at least two of the ultra-wideband radars satisfy the identification condition of the same candidate action.

[0009] Optionally, determining the candidate action as the detected target action if at least two ultra-wideband radars acquire channel impulse response data that satisfy the identification condition of the same candidate action includes: If the channel impulse response data acquired by at least two of the ultra-wideband radars satisfy the identification condition of the same candidate action, then for the channel impulse response data of the ultra-wideband radars that satisfy the identification condition of the same candidate action, a motion feature arrangement of the candidate action is determined, wherein the motion feature arrangement includes multiple motion features arranged in the acquisition order. For each motion feature, determine whether the time periods of the motion feature corresponding to all the channel impulse response data are consistent; If the time period of the motion feature corresponding to all the channel impulse response data is consistent, then the candidate action is determined as the target action to be detected.

[0010] Optionally, determining the candidate action corresponding to the channel impulse response data includes: Determine the directional velocity characteristics corresponding to the channel impulse response data; The motion characteristics of the target object are determined based on the directional velocity characteristics; Match candidate actions corresponding to the motion features.

[0011] Optionally, the directional velocity feature includes the object direction and the object velocity; determining the motion feature of the target object based on the directional velocity feature includes: Identify multiple directional velocity features that are continuous, have the same direction, and have a velocity greater than a preset velocity threshold. Determine whether the duration of the multiple directional velocity features is greater than a preset time threshold. If the duration of multiple directional velocity features is greater than the preset time threshold, then the detected motion feature is determined to be a unidirectional movement feature, wherein the direction of the unidirectional movement feature is the object direction corresponding to the multiple directional velocity features.

[0012] Optionally, the matching of candidate actions corresponding to the motion features includes: Obtain a plurality of continuously defined motion features; The motion features are arranged according to the detection order of the motion features; In the preset kicking motion, a matching kicking motion is matched with a preset feature arrangement that is consistent with the motion feature arrangement; The matching kicking motion is selected as the candidate motion.

[0013] Optionally, the step of obtaining the motion feature arrangement based on the detection order of the motion features includes: Obtain the motion cycle features of the motion feature arrangement; Determine whether the action cycle characteristics meet the action recognition conditions; If the motion cycle feature satisfies the motion recognition condition, then the matching kicking motion is arranged and matched based on the motion feature.

[0014] Optionally, determining whether the action cycle feature satisfies the action recognition condition includes: Obtain the maximum speed, maximum acceleration, and cycle duration from the action cycle characteristics; Determine whether the maximum speed is within a preset speed range, whether the maximum acceleration is greater than a preset acceleration threshold, and whether the period duration is less than a preset duration threshold; If the maximum speed is within a preset speed range, the maximum acceleration is greater than a preset acceleration threshold, and the period duration is less than a preset duration threshold, then the action recognition condition is met.

[0015] This invention also provides an action detection system, including a processor and a memory connected to the processor. The memory stores program code for a data acquisition module, program code for a candidate action acquisition module, and program code for a target action recognition module. The processor is configured to execute the data acquisition module program code to continuously acquire channel impulse response data collected by multiple ultra-wideband radars; The processor is also configured to execute candidate action acquisition module code to determine, for each of the ultra-wideband radars, a candidate action corresponding to the channel impulse response data based on the relative position between the ultra-wideband radar and the target detection area. The processor is also configured to execute target action recognition module code to determine the candidate action as the detected target action if the channel impulse response data acquired by at least two of the ultra-wideband radars satisfy the recognition conditions of the same candidate action.

[0016] This invention also provides an action detection device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the action detection method described above.

[0017] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the action detection method described above.

[0018] This invention proposes an action detection method, system, device, and computer-readable storage medium that continuously acquires channel impulse response data collected by multiple ultra-wideband radars. For each ultra-wideband radar, based on the relative position between the ultra-wideband radar and the target detection area, a candidate action corresponding to the channel impulse response data is determined. If the channel impulse response data collected by at least two ultra-wideband radars satisfy the identification conditions of the same candidate action, then the candidate action is determined as the detected target action. By setting up multiple ultra-wideband radars to detect kicking actions generated near the tailgate of a vehicle, and determining that a target action has been detected when at least two ultra-wideband radars simultaneously detect the same candidate action, the method can reduce false recognition of actions caused by environmental interference and improve the detection accuracy of kicking actions. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the first embodiment of the motion detection method of the present invention; Figure 2 This is a schematic diagram of the ultra-wideband radar setup in the motion detection method of this invention. Figure 3 This is a schematic diagram of a vertical kick scene that does not enter the vehicle projection area in the action detection method of this embodiment of the invention; Figure 4 This is a schematic diagram of a vertical kick scene entering the vehicle projection area in the motion detection method of this embodiment of the invention; Figure 5 This is a schematic diagram of a clockwise side kick scenario in the motion detection method of this invention. Figure 6 This is a schematic diagram of a counter-clockwise side kick scenario in the motion detection method of this invention. Figure 7 This is a schematic diagram of an environmental pedestrian interference scenario in the motion detection method of this invention. Figure 8 This is a detailed flowchart of the object velocity determination in the motion detection method of this invention. Figure 9 This is a schematic diagram of the action recognition conditions in the action detection method of this invention. Figure 10 This is a schematic diagram of the module structure of the motion detection device according to an embodiment of the present invention. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0023] This invention provides an action detection method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the motion detection method of the present invention. The method includes the following steps: Step S10: Continuously acquire channel impulse response data collected by multiple ultra-wideband radars; UWB (Ultra Wide Band) radar is a radar device based on UWB technology.

[0024] CIR (Channel Impulse Response) indicates the signal strength reflected back after a signal emitted by an ultra-wideband radar travels through different distances. Channel impulse response data reflects the environmental characteristics of the ultra-wideband radar and is usually represented by an array. The values ​​at different positions in the array represent the reflection intensity of the radar's emitted signal at different distances.

[0025] The ultra-wideband radar is installed at the rear of the vehicle. The specific location can be determined based on actual needs, such as installing it under the chassis at the rear of the vehicle. The detection area of ​​the ultra-wideband radar can include a certain area outside the rear of the vehicle and a certain area under the rear chassis, as shown in [reference needed]. Figure 2 The detection area of ​​the ultra-wideband radar is defined by taking the ultra-wideband radar as its apex and pointing towards the rear of the vehicle. This area will be referred to as the first detection area in the following text.

[0026] In this embodiment, there are multiple ultra-wideband radars. It should be noted that the ultra-wideband radars referred to in this embodiment are those used for motion detection; the specific number of ultra-wideband radars can be set based on actual needs. Figure 2 The system employs two ultra-wideband (UWB) radars: a first UWB1 and a second UWB2. Both UWB1 and UWB2 are positioned on either side of the tailgate. Their first detection areas are arranged opposite each other. The following explanation uses a configuration of two UWB radars as an example; configurations for other numbers of UWB radars can be similarly implemented and will not be elaborated upon further.

[0027] Step S20: For each ultra-wideband radar, based on the relative position between the ultra-wideband radar and the target detection area, determine the candidate action corresponding to the channel impulse response data. The target detection area is the area targeted by the action detection method, that is, the action detection method detects the actions that occur within the target detection area. The number and location of the target detection areas can be set according to actual needs. For example, the target detection areas can be set for the front of the vehicle, the tailgate, etc. It is understood that when there are multiple target detection areas, the action detection method of this application is executed for each target detection area. In this embodiment and subsequent embodiments, the target detection area set for the tailgate is used as an example for illustration.

[0028] Different ultra-wideband radars are installed in different locations on the vehicle. Therefore, the relative positions between different ultra-wideband radars and the tailgate are also different. The signal characteristics reflected by each ultra-wideband radar are also different when a kicking action occurs at the tailgate. Therefore, in this embodiment, the channel impulse response data is analyzed based on the relative position between the ultra-wideband radar and the tailgate to determine the kicking action that occurred at the tailgate.

[0029] The kicking action is a kicking action set in the vehicle to trigger tailgate control; the specific type of kicking action can be preset by the factory or set by the user; the action can be a kicking action, a gesture action or other limb action, etc.; the kicking action can be a vertical kick, a horizontal kick, etc.; this embodiment and subsequent embodiments use tailgate control as the vehicle control and kicking action as the target action as an example for explanation. Other types of vehicle control and actions can be implemented by analogy and will not be described in detail.

[0030] The candidate action is a kicking motion determined by channel impulse response data from a single ultra-wideband radar.

[0031] Step S30: If the channel impulse response data collected by at least two ultra-wideband radars meet the identification conditions of the same candidate action, then the candidate action is determined as the target action to be detected.

[0032] If the channel impulse response data acquired by any ultra-wideband radar does not meet the identification conditions for the same candidate action, then no target action is detected.

[0033] Because different ultra-wideband radars have different relative positions to the tailgate, they can detect actions occurring at the tailgate from different directions. When multiple ultra-wideband radars collect channel impulse response data pointing to the same candidate action, meaning the same candidate action can be identified in all directions of the ultra-wideband radars, it can be determined that a candidate action has actually occurred at the tailgate. In this case, the candidate action is taken as the target action. In this embodiment, the identification conditions for satisfying the same candidate action are set to at least two. In practical applications, the number of channel impulse response data that meet the identification conditions for the same candidate action can be set based on actual needs, such as setting a proportion or a specific number. The specific number can be set based on actual needs. For example, if more than half of the channel impulse response data meet the identification conditions for the same candidate action, then the candidate action is determined to be the target action. Alternatively, if all channel impulse response data meet the identification conditions for the same candidate action, then the candidate action is determined to be the target action.

[0034] The target action is a kicking motion performed by the user as determined by this embodiment.

[0035] Compared to a single detection device, this embodiment combines multi-directional signal acquisition to comprehensively detect kicking actions, avoiding the problem of significant detection deviations caused by severe environmental interference, which is common with single devices.

[0036] This embodiment uses multiple ultra-wideband radars to detect kicking motions near the tailgate of the vehicle. When at least two ultra-wideband radars detect the same candidate motion simultaneously, the target motion is determined to be detected. This reduces the false recognition of motions caused by environmental interference and improves the detection accuracy of kicking motions.

[0037] Furthermore, in the second embodiment of the motion detection method of the present invention based on the first embodiment of the present invention, step S30 includes the following steps: Step S40: If a candidate action is detected based on the channel impulse response data collected by any ultra-wideband radar, then a second detection area is determined, wherein the second detection area is the area where the first detection area of ​​the ultra-wideband radar that detected the candidate action intersects with the first detection areas of other ultra-wideband radars. Step S50: Based on the detection data of each ultra-wideband radar for the corresponding second detection area, determine whether there are at least two ultra-wideband radars that collect channel impulse response data that satisfy the identification conditions of the same candidate action.

[0038] When at least one ultra-wideband radar acquires channel impulse response data indicating the detection of a candidate action, the determination of the target action is triggered. It is understood that, due to the different locations of different ultra-wideband radars, the detected data will differ. Therefore, when a kicking action occurs, not all channel impulse response data from all ultra-wideband radars can directly determine the kicking action. However, as long as one ultra-wideband radar detects a candidate action, it is considered that a kicking action may have occurred, and therefore, the subsequent target action determination process is executed.

[0039] The second detection area is the area where the first detection area of ​​the ultra-wideband radar that detected the candidate action intersects with the first detection areas of the other ultra-wideband radars. In other words, it identifies the ultra-wideband radars capable of detecting the area corresponding to the triggered candidate action. In practical applications, for a specific detection area, such as a vehicle tailgate, the first detection areas of all ultra-wideband radars can be positioned facing the tailgate, and the intersection of all the first detection areas of the ultra-wideband radars constitutes the second detection area. The specific range of the second detection area can be set based on actual needs, as shown in [reference needed]. Figure 2 The first ultra-wideband radar UWB1 is located on the left side of the tailgate, and the second ultra-wideband radar UWB2 is located on the right side of the tailgate. The line connecting the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2 is parallel to the outer side of the tailgate. The first detection area of ​​the first ultra-wideband radar UWB1 faces the second ultra-wideband radar UWB2, and the first detection area of ​​the second ultra-wideband radar UWB2 faces the first ultra-wideband radar UWB1. The intersection area between the two radars is determined according to the setting parameters of the ultra-wideband radars. Specifically:

[0040] Among them, S l S is the arc radius of the first detection region of the first ultra-wideband radar UWB1; r d is the arc radius of the first detection area of ​​the second ultra-wideband radar UWB2; d is the distance between the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2, which is determined during vehicle manufacturing; S k S is the length of the second detection area in the direction perpendicular to the outer side of the tailgate; k The value can be set according to actual needs to adjust the second detection area.

[0041] It is understandable that the second detection area is the intersection area of ​​each ultra-wideband radar. Therefore, it can ensure the comprehensiveness of the detection of the second detection area by each ultra-wideband radar. At the same time, the edge areas in the first detection area are removed to reduce the interference of edge factors on the detection of kicking action and reduce the possibility of false detection.

[0042] After determining the second detection area, the relevant data for the second detection area based on the channel impulse response data are used to determine whether each ultra-wideband radar meets the recognition criteria for the same candidate action. It is understood that the channel impulse response data contains echo signals with specific distances. Therefore, based on the distance in the signals, it is possible to determine which signals are reflected from the second detection area to the ultra-wideband radar. Thus, it is possible to determine the signals used to detect objects in the second detection area, and consequently, whether the recognition criteria for the same candidate action are met.

[0043] Further, step S50 includes the following steps: Step S51: Obtain the arrangement of motion features obtained by each ultra-wideband radar detecting the second detection area; Step S52: For each ultra-wideband radar, determine whether the arrangement of motion features satisfies the preset feature arrangement corresponding to the candidate action. Step S53: If the motion feature arrangement of all ultra-wideband radars satisfies the preset feature arrangement corresponding to the candidate action, then it is determined that the channel impulse response data collected by all ultra-wideband radars satisfies the identification condition of the same candidate action.

[0044] Motion characteristics indicate the type of motion of the target object, such as kicking, retracting the leg, or side kicking.

[0045] The motion feature arrangement is a set obtained by arranging motion features in the order of detection.

[0046] For a specific kicking action, it can be broken down into the sequential execution of multiple action features. Therefore, a preset feature arrangement is set for each kicking action in advance, such as: See Figure 3If the kicking action is a vertical kick that does not enter the vehicle projection area, then its corresponding preset feature arrangement is kicking (close to the ultra-wideband sensor) and retracting the leg (away from the ultra-wideband sensor).

[0047] See Figure 4 If the kicking action is a vertical kick that enters the vehicle projection area, then its corresponding preset feature arrangement is kicking (first approaching the UWB sensor and then moving away from the UWB sensor) and retracting the leg (first approaching the UWB sensor and then moving away from the UWB sensor).

[0048] See Figure 5 If the kicking action is a side kick, then its corresponding preset feature arrangement is diagonal kick, side kick, and diagonal leg retraction; the movement of the action relative to the ultra-wideband sensor is determined based on the direction of the side kick and the position of the ultra-wideband sensor.

[0049] When the arrangement of motion features detected by each ultra-wideband radar corresponds to the preset arrangement of features corresponding to the candidate action, it is considered to meet the recognition conditions of the same candidate action.

[0050] See also Figure 3 For vertical kicks that do not enter the vehicle's projection area: During the kicking motion, the foot moves closer to the first UWB1 and the second UWB2 radars. It should be noted that although the foot does not move directly toward the first UWB1 or the second UWB2 radars, the distance between the foot and the first UWB1 and the second UWB2 radars is shortened during the kicking motion. Therefore, it is considered to be moving closer to the first UWB1 and the second UWB2 radars. During the leg retraction movement, the foot moves away from the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2.

[0051] Therefore, if the first ultra-wideband radar UWB1 detects that the object first approaches and then moves away, and the second ultra-wideband radar UWB2 detects that the object first approaches and then moves away, then it is considered that the two ultra-wideband radars simultaneously meet the recognition conditions of a vertical kick that has not entered the vehicle's projection area, and the target action is determined to be a vertical kick.

[0052] See Figure 4 For vertical kicks that enter the vehicle's projection area: During the kicking motion, the foot moves close to the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2, passes through the line connecting the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2, and then moves away from the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2. During the leg retraction movement, the foot moves away from the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2, passes through the line connecting the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2, and then moves closer to the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2. Therefore, if the first ultra-wideband radar UWB1 detects that the object first approaches and then moves away + first moves away and then approaches, and the second ultra-wideband radar UWB2 detects that the object first approaches and then moves away + first moves away and then approaches, then it is considered that the two ultra-wideband radars simultaneously meet the recognition conditions of a vertical kick entering the vehicle's projection area, and the target action is determined to be a vertical kick.

[0053] See Figure 5 For clockwise side kicks (taking the example of not entering the vehicle projection area; in the determination of entering the vehicle projection area, analogous to vertical kicks, the opposite direction feature of the same action is added): During the left front kick, the foot moves closer to the first ultra-wideband radar UWB1 and further away from the second ultra-wideband radar UWB2. During the side kick, the foot moves from the first ultra-wideband radar UWB1 to the second ultra-wideband radar UWB2; During the leg retraction movement, the foot moves closer to the first ultra-wideband radar UWB1 and further away from the second ultra-wideband radar UWB2.

[0054] Therefore, if the first ultra-wideband radar UWB1 detects that the object first approaches, then moves away, and then approaches again, and the second ultra-wideband radar UWB2 detects that the object first moves away, then approaches, and then moves away again, then it is considered that the two ultra-wideband radars simultaneously meet the recognition conditions of a clockwise side kick, and the target action is determined to be a side kick.

[0055] See Figure 6 For a counter-clockwise side kick (assuming it doesn't enter the vehicle's projection area): During the right front kick, the foot moves closer to the second ultra-wideband radar UWB2 and further away from the first ultra-wideband radar UWB1. During the side kick, the foot moves from the second ultra-wideband radar UWB2 to the first ultra-wideband radar UWB1; During the leg retraction movement, the foot moves closer to the second ultra-wideband radar UWB2 and further away from the first ultra-wideband radar UWB1.

[0056] Therefore, if the second ultra-wideband radar UWB2 detects that the object first approaches, then moves away, and then approaches again, and the first ultra-wideband radar UWB1 detects that the object first moves away, then approaches, and then moves away again, then it is considered that both ultra-wideband radars simultaneously meet the recognition conditions of a counterclockwise side kick, and the target action is determined to be a side kick.

[0057] It also allows for the arrangement of motion characteristics for some disruptive actions in the environment, as shown in [reference]. Figure 7 For distracting objects in the environment, such as pedestrians moving from left to right: As the object moves from a distance to the first ultra-wideband radar UWB1, both the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2 detect the object approaching. During the process of moving from the first ultra-wideband radar UWB1 to the second ultra-wideband radar UWB2, the first ultra-wideband radar UWB1 detected the object moving away, while the second ultra-wideband radar UWB2 detected the object moving closer. As the object moves away from the second ultra-wideband radar UWB2, both the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2 detect that the object is moving away.

[0058] Therefore, by arranging these motion characteristics, we can identify them as environmental interference and not treat them as target actions, thus avoiding false detections.

[0059] In this embodiment, the accurate identification of kicking motions is achieved by determining the arrangement of motion characteristics of each ultra-wideband radar.

[0060] Furthermore, in the third embodiment of the motion detection method of the present invention based on the first embodiment, step S30 includes: Step S31: For each channel impulse response data, determine the motion feature arrangement of the candidate actions, wherein the motion feature arrangement includes multiple motion features arranged in the acquisition order. Step S32: For each motion feature, determine whether the time periods of the motion features corresponding to all the channel impulse response data are consistent; Step S33: If the time period of the motion feature corresponding to all the channel impulse response data is consistent, then the candidate action is determined as the target action to be detected.

[0061] The time period is the time covered by the data corresponding to the action feature; for example, in the channel impulse response data, if the data in the time period T1~T2 determines that the action feature is a front kick, then the time period corresponding to the kick is T1~T2.

[0062] It is understandable that, in the case of a kicking action being triggered, the action characteristics determined by each ultra-wideband radar at the same time should be the same; For example, a vertical kick that does not enter the vehicle's projection area: During the time period T1~T2, the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2 determine that the feet are close together, thereby determining the characteristics of the kicking motion. During the T3-T4 time period following T1-T2, the first ultra-wideband radar UWB1 and the second ultra-wideband radar UWB2 determine that the feet are moving away, thereby determining the characteristics of the leg retraction movement.

[0063] T2 and T3 can be a single time node or two time nodes with a small time difference. The maximum value of the specific time difference can be set based on actual needs. The same applies to the time nodes of subsequent continuous time periods, which will not be elaborated further.

[0064] For example, regarding a clockwise side kick: During the T1~T2 cycle, the first ultra-wideband radar UWB1 determines that the foot is approaching, and the second ultra-wideband radar UWB2 determines that the foot is moving away, thereby determining the action characteristics of the left front kick. During the T3-T4 time period following T1-T2, the first ultra-wideband radar UWB1 determines that the foot is moving away, while the second ultra-wideband radar UWB2 determines that the foot is moving closer, thus determining the action characteristics of the side kick. During the T5-T6 time period following T3-T4, the first ultra-wideband radar UWB1 determines that the foot is approaching, and the second ultra-wideband radar UWB2 determines that the foot is moving away, thereby determining the movement characteristics of the leg retraction.

[0065] When there are inconsistent motion characteristics in the time period, it is assumed that the motion characteristics determined by multiple ultra-wideband radars in the same period are different, and the determination of the kicking motion is not uniform. Therefore, it is assumed that no target motion has occurred.

[0066] It should be noted that in practical applications, different ultra-wideband radars have different detection timing sequences for foot movement at the same moment. Therefore, there may be time deviations between the data from ultra-wideband radars. When judging the consistency of time periods, a time difference threshold can be set. When the difference between time periods is less than or equal to the time difference threshold, the time periods are considered to be consistent. When the difference between time periods is greater than the time difference threshold, the time periods are considered to be inconsistent.

[0067] In this embodiment, by determining the consistency of the time period of motion features, the uniformity of kicking action detection by multiple ultra-wideband radars can be guaranteed, thereby improving the accuracy of kicking action detection.

[0068] Furthermore, in the fourth embodiment of the motion detection method of the present invention based on the first embodiment, step S20 includes the following steps: Step S21: Determine the directional velocity characteristics corresponding to the channel impulse response data; Step S22: Determine the motion characteristics of the target object based on the directional velocity characteristics; Step S23: Match candidate actions corresponding to motion features.

[0069] Directional velocity characteristics are the features of directional movement relative to the ultra-wideband radar, as indicated by channel impulse response data; specifically, they can include the direction and speed of movement of the target object relative to the ultra-wideband radar.

[0070] As the Doppler effect describes, when there is relative motion between a wave source and an observer, the frequency of the wave received by the observer deviates from the actual frequency of the wave source. In this embodiment, the ultra-wideband radar is the observer, and the target object within the detection area reflects the signal emitted by the ultra-wideband radar, making the target object the wave source. The movement between the target object and the ultra-wideband radar causes a change in the frequency of the signal received by the ultra-wideband radar, i.e., the channel impulse response data also changes. Therefore, the channel impulse response data can reflect the relative motion relationship between the target object and the ultra-wideband radar. When motion detection is required, the vehicle is stationary, i.e., the ultra-wideband radar is stationary. Therefore, the relative motion directly reflects the motion of the target object. Thus, the channel impulse response data can be used to determine the directional movement characteristics of the target object relative to the ultra-wideband radar.

[0071] Directional velocity characteristics reflect the relative motion of the target object with respect to the ultra-wideband radar, while the ultra-wideband radar is stationary. Therefore, directional velocity characteristics can determine the motion of the target object itself, i.e., motion characteristics.

[0072] Motion characteristics indicate the type of motion of the target object, such as kicking, retracting the leg, or side kicking.

[0073] It is understandable that different kicking actions have different characteristics. For example, the characteristics of a vertical kick are that the foot kicks out towards the rear of the car first and then retracts; the characteristics of a horizontal kick are that the foot kicks out to the left and forward first, then moves laterally, and then retracts. Therefore, when setting kicking actions, the motion characteristics of the kicking action are set based on the specific type of kicking action. In practical applications, the user's motion characteristics are detected, and the detected motion characteristics are matched with the motion characteristics of the kicking action to determine the target action.

[0074] It is understood that in this embodiment, the directional velocity feature reflects the motion of objects within the detection area of ​​the ultra-wideband radar, while the motion feature reflects the type of motion. The target action is then determined based on the motion feature matching. In this embodiment, the action is determined based on the characteristics of the kicking action. Therefore, it can avoid false triggering of the kicking action detection caused by objects accidentally approaching the vehicle in the environment, and can improve the accuracy of kicking action detection.

[0075] This embodiment uses an ultra-wideband radar to determine the directional velocity characteristics of target objects within the detection area, thereby identifying the motion characteristics of the target objects within the detection area and reflecting their motion. Based on this motion, the corresponding target action is matched, enabling detection based on the actual motion characteristics of the kicking action, thus improving detection accuracy.

[0076] Further, see Figure 8 S21 includes the following steps: S211, determine the signal phase change rate of the channel impulse response data, and determine the object direction corresponding to the channel impulse response data; S212, the signal frequency shift corresponding to the phase change rate of the matched signal; S213, obtain the signal wavelength corresponding to the channel impulse response data, and calculate the object speed based on the signal wavelength and signal frequency shift; S214, the directional velocity feature is obtained by associating the velocity of the associated object with the object's direction.

[0077] The signal phase change rate is the rate of change of the phase of the signal received by the ultra-wideband radar, as reflected in the channel impulse response data. The signal phase change rate is obtained by dividing the phase difference of the received signals at two different times by the time interval between them. Specifically, the signal phase change rate can be determined based on the received signal corresponding to the previously transmitted pulse and the pulse transmitted in the current time.

[0078] The object direction is the direction of relative motion of the object within the detection area relative to the ultra-wideband radar.

[0079] Signal frequency shift is a frequency change phenomenon caused by the relative motion between the wave source and the receiver; there is a direct relationship between signal frequency shift and the rate of change of signal phase:

[0080] Among them, f d For signal frequency shift; The rate of change of the signal phase; t represents the pulse repetition period of the ultra-wideband radar.

[0081] The signal wavelength is the wavelength of the output signal of the ultra-wideband radar; the specific value of the signal wavelength can be set based on actual needs.

[0082] The object velocity is the moving speed of the target object relative to the ultra-wideband radar; based on the Doppler frequency shift formula, there is a corresponding relationship between the object velocity, signal wavelength, and signal frequency shift:

[0083] Where v is the velocity of the object; λ is the wavelength of the signal.

[0084] The object velocity indicates the speed at which the target object moves relative to the ultra-wideband radar, and the object direction indicates the direction of movement of the target object relative to the ultra-wideband radar. Therefore, by combining the object velocity and the object direction, we can reflect the motion of the target object relative to the ultra-wideband radar, thus obtaining the directional velocity characteristics that reflect the directional movement relative to the ultra-wideband radar.

[0085] Furthermore, step S211 includes: S2111, wavelet analysis is performed on the channel impulse response data to obtain the positive signal energy and negative signal energy; S2112, if the positive signal energy is greater than the first energy threshold, then the direction of the object is determined to be away from the ultra-wideband radar, where the first energy threshold is a positive number; If the positive signal energy is less than or equal to the first energy threshold, then the direction of the object is determined to be not far from the ultra-wideband radar, where the first energy threshold is a positive number; S2113, if the negative signal energy is less than the second energy threshold, the direction of the object is determined to be close to the ultra-wideband radar, where the second energy threshold is a negative number.

[0086] If the negative signal energy is greater than or equal to the second energy threshold, then the direction of the object is determined to be not close to the ultra-wideband radar.

[0087] If the positive signal energy is less than or equal to the first energy threshold and the negative signal energy is greater than or equal to the second energy threshold, then it is considered that no kicking action was detected.

[0088] Wavelet analysis is a multi-resolution signal processing technique that can simultaneously capture the time and frequency domain features of a signal. Specifically, it uses wavelet basis functions to perform convolution operations on channel impulse response data to decompose signal components of different scales. The signal energy can be obtained by squaring or taking the modulus of the signal components.

[0089] Positive signal energy corresponds to the signal energy generated by motion moving away from the ultra-wideband radar.

[0090] Negative signal energy corresponds to the signal energy generated by motion approaching the ultra-wideband radar.

[0091] In practice, positive wavelet analysis can be performed on the channel impulse response data to obtain positive signal energy, and negative wavelet analysis can be performed on the channel impulse response data to obtain negative signal energy.

[0092] The energy threshold is the threshold for triggering motion detection; the first energy threshold is the energy threshold set for motion that is far away from the ultra-wideband radar, and the second energy threshold is the energy threshold set for motion that is close to the ultra-wideband radar; the specific value of the energy threshold can be set based on actual needs.

[0093] It is understandable that in practical applications, due to the complexity of environmental factors, objects that can be detected by accidental displacement may interfere with the detection of kicking actions. Therefore, in order to avoid the influence of accidental factors, an energy threshold is set in this embodiment. When the signal energy meets the energy threshold requirement, it is considered that a kicking action may have occurred, and the object direction is determined to proceed with subsequent steps. When the signal energy does not meet the energy threshold requirement, it is considered that no kicking action has occurred.

[0094] In this embodiment and subsequent embodiments, a positive direction indicates moving away from the ultra-wideband radar, and a negative direction indicates moving closer to the ultra-wideband radar; this will not be elaborated further.

[0095] Furthermore, step S211 includes: S2114, perform wavelet analysis on the channel impulse response data within the current noise detection period to obtain the signal energy; S2115 converts signal energy into signal decibels and calculates the average signal decibels within the current noise detection period; S2116, obtain the historical noise decibels of the previous noise detection cycle, and combine the average signal decibels and the historical noise decibels to obtain the noise decibels of the current cycle; S2117, Subtract the current period noise decibel from the average signal decibel to obtain the target signal; S2118, determine the rate of change of the target signal's phase.

[0096] Signal decibels indicate the signal strength corresponding to the signal energy; specifically:

[0097] Where dB is the signal decibel; E signal This refers to signal energy.

[0098] The current noise detection period is a noise statistics window; the length of the current noise detection period can be set according to actual needs, such as setting the current noise detection period to 3 times the pulse repetition period of the ultra-wideband radar; it can be understood that the current noise detection period is a sliding window; taking 3 times the pulse repetition period as an example, the current noise detection period averages the signal decibels corresponding to the three pulse repetition periods n, n-1, and n-2 to obtain the average signal decibels, where n is the current pulse period.

[0099] The previous noise detection period is the noise detection period corresponding to the n-1 pulse repetition period, which is specifically determined based on the signal decibels corresponding to the three pulse repetition periods n-1, n-2, and n-3.

[0100] The historical noise decibel is the noise decibel determined in the previous noise detection cycle; it can be understood that the historical noise decibel of the previous noise detection cycle is obtained by the average signal decibel determined in the previous noise detection cycle and the historical noise decibel of the noise detection cycle before that.

[0101] In practical applications, sudden environmental events may cause abrupt changes in signal noise. Therefore, in order to avoid significant changes in signal processing due to sudden events, this embodiment combines the historical noise decibels of the previous noise detection cycle with the average signal decibels of the current noise detection cycle to comprehensively determine the current period noise decibels of the current noise detection cycle; thereby enabling the noise decibels to be updated smoothly and eliminating the impact of sudden events.

[0102] Specifically, the current periodic noise level can be calculated by weighted summation using the average signal level in decibels and historical noise levels in decibels:

[0103] Where noise is the current periodic noise in decibels; old_noise is the historical noise in decibels; ave_dB is the average signal in decibels; and A is the weight of the historical noise in decibels. The specific value can be set according to actual needs.

[0104] After obtaining the current period noise level in decibels, subtracting the current period noise level from the average signal level in decibels yields the denoised signal, i.e., the target signal.

[0105] Where snr is the target signal.

[0106] It should be noted that signal energy includes positive signal energy and negative signal energy. The positive signal energy can be denoised to obtain the positive target signal, and the negative signal energy can be denoised to obtain the negative target signal. The corresponding signal phase change rate can then be obtained from the positive and negative target signals.

[0107] The signal phase change rate is determined based on the denoised target signal, which enables the elimination of environmental background noise interference and improves detection accuracy.

[0108] Furthermore, S2118 includes the following steps: S21181, retain the maximum frequency signal corresponding to each moment in the target signal, and combine the maximum frequency signals corresponding to each moment to obtain the target frequency signal; S21182, determine the rate of change of the signal phase corresponding to the target frequency signal.

[0109] The maximum frequency signal is the signal with the largest absolute frequency at a given moment.

[0110] It is understandable that at the same moment, the signal received by the ultra-wideband radar contains environmental interference, such as the slight shaking of ground debris, which has a very low frequency. However, when a kicking action occurs, the signal received by the ultra-wideband radar due to the kicking action has a higher frequency. Therefore, in order to further avoid the influence of environmental factors, in this embodiment, only the highest frequency signal is selected to determine the signal phase change rate, so as to have a strong sensitivity to the real kicking action and reduce the impact of environmental interference.

[0111] Furthermore, the directional velocity feature includes the object's direction and its velocity; step S22 includes the following steps: Step S221: Determine multiple directional velocity features that are continuous, have the same object direction, and have an object velocity greater than a preset velocity threshold. Step S222: Determine whether the duration of the velocity features in multiple directions is greater than a preset time threshold. Step S223: If the duration of multiple directional velocity features is greater than a preset time threshold, then the detected motion feature is determined to be a unidirectional movement feature, wherein the direction of the unidirectional movement feature is the object direction corresponding to the multiple directional velocity features.

[0112] The kicking action is a continuous action, and the motion characteristics reflect the specific type of kicking action. In this embodiment, the motion characteristics are determined by continuously defined directional velocity characteristics.

[0113] In this embodiment, continuous directional velocity characteristics refer to directional velocity characteristics determined by continuous pulse repetition periods in an ultra-wideband radar, such as directional velocity characteristics determined by n-1 and n being continuous directional velocity characteristics.

[0114] The preset speed threshold indicates the minimum speed required for a kicking motion; the specific value can be set based on actual needs; the preset speed threshold can be set separately for positive and negative object speeds, with positive speed thresholds being positive numbers and negative speed thresholds being negative numbers.

[0115] It is understandable that when a kicking action occurs, its corresponding motion characteristics usually exhibit similar motions for a certain duration. For example, a kick will continuously approach the ultra-wideband radar for a certain period of time. In terms of directional velocity characteristics, if the kicking action begins in cycle T1, then in the previous cycle, T0, the directional velocity characteristics show that the object direction is negative or the object velocity is greater than the negative velocity threshold. In cycle T1, the directional velocity characteristics show that the object direction is negative and the object velocity is less than the negative velocity threshold. During the duration of the kicking action, the directional velocity characteristics consistently show that the object direction is negative and the object velocity is less than the negative velocity threshold. When the kicking action ends, such as in cycle Tx, the directional velocity characteristics show that the object direction is positive or the object velocity is greater than the negative velocity threshold. At this point, it can be determined that the kicking action occurred in cycles T1 to Tx, reflecting multiple directional velocity characteristics with continuous, negative object directions and object velocities less than the negative velocity threshold, thus identifying the kicking action.

[0116] For example, regarding leg retraction, the object will continuously move away from the ultra-wideband radar for a certain period of time. In terms of directional velocity characteristics, if leg retraction begins in cycle T1, then in the previous cycle (T0), the directional velocity characteristics show the object's direction as negative or its velocity as less than the positive velocity threshold. In cycle T1, the directional velocity characteristics show the object's direction as positive and its velocity as greater than the positive velocity threshold. Throughout the leg retraction cycle, the directional velocity characteristics consistently show the object's direction as positive and its velocity as greater than the positive velocity threshold. When leg retraction ends, such as in cycle Tx, the directional velocity characteristics show the object's direction as positive or its velocity as greater than the positive velocity threshold. At this point, it can be determined that the leg retraction action occurred between cycles T1 and Tx, reflecting multiple directional velocity characteristics with continuous, negative object directions and object velocities greater than preset velocity thresholds, thus identifying the leg retraction action.

[0117] Therefore, by analyzing multiple directional velocity features that are continuous, have the same object orientation, and have an object speed greater than a preset speed threshold, it is possible to identify features of single-direction movement, such as positive or negative movement.

[0118] In practical applications, since the kicking and retracting movements require a certain duration, this embodiment sets a preset time threshold to determine the minimum duration determined by the movement characteristics; the specific value of the preset time threshold can be set based on actual needs.

[0119] Therefore, if a movement meets the unidirectional motion characteristic and the duration requirement for motion execution is met, it can be determined as a unidirectional movement characteristic.

[0120] It should be noted that some action features involve multiple directions. For example, for a side kick from left to right, the kicker will first approach the ultra-wideband radar at a slightly slower speed than the kicking leg. After the foot moves directly towards the ultra-wideband radar, it will then move away from the ultra-wideband radar at a slightly slower speed than the retraction leg. In this case, a speed range and a negative-to-positive directional arrangement can be set for the side kick as conditions for determining the action features, thereby determining the action features of the composite direction.

[0121] Further, step S23 includes the following steps: Step S231: Obtain multiple continuously defined motion features; Step S232: Obtain the motion feature arrangement according to the detection order of the motion features; Step S233: In the preset kicking action, match the kicking action that is consistent with the preset feature arrangement and the motion feature arrangement; Step S234: Match the kicking motion as a candidate motion.

[0122] In this embodiment, the continuity of motion features refers to the continuity of the sequence of motion features. Periods of undetermined motion features are allowed between consecutive motion features. At the same time, a preset number of periods can be set. When the target action is not clearly determined for a consecutive preset number of periods, the previously determined motion features are discarded.

[0123] The detection order of motion features is the chronological order in which the motion features are determined. For example, motion features determined by n-1 pulse repetition cycles precede motion features determined by n pulse repetition cycles.

[0124] The motion feature arrangement is a set obtained by arranging motion features in the order of detection.

[0125] For a specific kicking action, it can be broken down into the sequential execution of multiple action features. Therefore, a preset feature arrangement is set for each kicking action in advance, such as: If the kicking action is a vertical kick, then its corresponding preset feature arrangement is kicking (close to the ultra-wideband sensor) and retracting the leg (away from the ultra-wideband sensor).

[0126] If the kicking action is a side kick, its corresponding preset feature arrangement is: kicking (approaching the ultrawideband sensor and the speed decreases), side kicking (approaching and then moving away from the ultrawideband sensor), and retracting the leg (moving away from the ultrawideband sensor).

[0127] When the actual detected motion feature arrangement matches the preset feature arrangement of a certain kicking action, it is determined that the kicking action has been detected, that is, the target action is determined.

[0128] It should be noted that a match can be determined as long as the motion feature arrangement contains a preset kicking action. For example, if the motion feature arrangement is leg retraction, kicking, and leg retraction, then it contains the preset feature arrangement corresponding to the vertical kick of kicking and retraction. Therefore, a vertical kick is detected.

[0129] In this embodiment, the target action can be accurately determined by matching the sequential arrangement of motion features.

[0130] Furthermore, step S232 is followed by the following steps: Step S235: Obtain the motion cycle features arranged by motion features; Step S236: Determine whether the action cycle features meet the action recognition conditions; Step S237: If the motion cycle features meet the motion recognition conditions, then the matching kicking motion is arranged based on the motion features.

[0131] The motion cycle feature is the related signal feature of the cycle designed by arranging motion features; specifically, it can include velocity features, time features, acceleration features, etc.

[0132] The action recognition condition is that when a kicking action occurs, the motion feature arrangement must have the motion periodic features.

[0133] It is understandable that in practical applications, due to environmental interference, there may be scenarios that conform to the characteristics of a kicking action, which may lead to accidental touches. Therefore, in this embodiment, in addition to the features, action recognition conditions are set for the action cycle characteristics of the signal to further set the conditions for the signal characteristics of the kicking scene, thereby further reducing the probability of accidental touches.

[0134] Further, step S236 includes the following steps: Step S2361: Obtain the maximum speed, maximum acceleration, and cycle duration from the motion cycle features; Step S2362: Determine whether the maximum speed is within the preset speed range, whether the maximum acceleration is greater than the preset acceleration threshold, and whether the period duration is less than the preset duration threshold. Step S2363: If the maximum speed is within the preset speed range, the maximum acceleration is greater than the preset acceleration threshold, and the period duration is less than the preset duration threshold, then the action recognition condition is met.

[0135] The maximum velocity is determined among all moments in the sequence of motion characteristics. Figure 9 Chinese f c f is the maximum negative velocity in the maximum velocity range. s This is the maximum positive velocity within the maximum speed range.

[0136] The preset speed range is the required speed range for a kicking motion. The specific value can be set based on actual needs. It is understood that in practical applications, kicking motions are usually within a certain speed range. Therefore, in order to distinguish them from environmental interference, the possible speeds of kicking motions are set to obtain the preset speed range, and the maximum speed is evaluated based on the preset speed range to clarify the possibility of kicking motions.

[0137] The maximum acceleration is the maximum acceleration determined across all moments in the sequence of motion characteristics; Figure 9 In the equation, a1 represents the maximum negative acceleration within the maximum acceleration, and a2 represents the maximum positive acceleration within the maximum acceleration.

[0138] The preset acceleration threshold is the acceleration requirement set for kicking movements; the specific value can be set based on actual needs. Kicking movements typically exhibit certain acceleration characteristics, such as a significant acceleration at the initial stage of the kick. Therefore, to match the characteristics of kicking movements, a preset acceleration threshold is set to determine the likelihood of a kicking action based on acceleration. It should be noted that different motion characteristics have different acceleration requirements. Therefore, a corresponding preset acceleration threshold can be set for each motion characteristic based on its type, and the motion characteristic can be compared with the corresponding preset acceleration threshold. The preset acceleration threshold requirement is met when the maximum acceleration of all motion characteristics is greater than the corresponding preset acceleration threshold.

[0139] The period duration is the time difference between the earliest and latest signals from the signal sources of the motion feature arrangement; Figure 9 C s S is the earliest signal acquisition time. e This is the latest time the signal was acquired.

[0140] The preset duration threshold is the required completion time for the kicking action. To avoid environmental interference, this embodiment sets a more difficult condition to meet, that is, the kicking action must be completed within the preset duration threshold, thereby reducing the execution time of the kicking action. The shorter the time, the more difficult it is for environmental interference to mistakenly trigger the kicking action, thus reducing the probability of mistaken triggering.

[0141] In this embodiment, by setting corresponding conditions for the maximum speed, maximum acceleration, and cycle duration in the action cycle characteristics, it is possible to further combine the characteristics of the kicking action to set signal feature requirements, thereby further reducing the misidentification of environmental interference noise.

[0142] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0143] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0144] This invention also provides an action detection system, including a processor and a memory connected to the processor. The memory stores program code for a data acquisition module, program code for a candidate action acquisition module, and program code for a target action recognition module. The processor is configured to execute the data acquisition module program code to continuously acquire channel impulse response data collected by multiple ultra-wideband radars; The processor is also configured to execute candidate action acquisition module code to determine, for each of the ultra-wideband radars, a candidate action corresponding to the channel impulse response data based on the relative position between the ultra-wideband radar and the target detection area. The processor is also configured to execute target action recognition module code to determine the candidate action as the detected target action if the channel impulse response data acquired by at least two of the ultra-wideband radars satisfy the recognition conditions of the same candidate action.

[0145] Reference Figure 10 In terms of hardware structure, the motion detection device may include components such as a communication module 10, a memory 20, and a processor 30. In the motion detection device, the processor 30 is connected to both the memory 20 and the communication module 10. The memory 20 stores a computer program, which is executed by the processor 30. When the computer program is executed, it implements the steps of the above-described method embodiments.

[0146] The communication module 10 can connect to external communication devices via a network. The communication module 10 can receive requests from external communication devices, and can also send requests, instructions and information to external communication devices. External communication devices can be other motion detection devices, servers or Internet of Things devices, such as televisions, etc.

[0147] The memory 20 can be used to store software programs and various data. The memory 20 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as continuously acquiring channel impulse response data from multiple ultra-wideband radars), etc.; the data storage area may include a database, and may store data or information created based on system usage. Furthermore, the memory 20 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0148] The processor 30 is the control center of the motion detection device. It connects various parts of the device via interfaces and lines, and performs various functions and processes data by running or executing software programs and / or modules stored in the memory 20, and by calling data stored in the memory 20, thereby providing overall monitoring of the motion detection device. The processor 30 may include one or more processing units; optionally, it may integrate an application processor and a modem processor. The application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may not be integrated into the processor 30.

[0149] although Figure 10 Not shown, but the above-described motion detection device may further include a circuit control module, which is connected to a power supply to ensure the normal operation of other components. Those skilled in the art will understand that... Figure 10 The motion detection device structure shown does not constitute a limitation on the motion detection device. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0150] The present invention also proposes a computer-readable storage medium having a computer program stored thereon. The computer-readable storage medium may be... Figure 10The memory 20 in the motion detection device may also be at least one of ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk, etc. The computer-readable storage medium includes several instructions to cause a terminal device with a processor (which may be a television, automobile, mobile phone, computer, server, terminal, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0151] In this invention, the terms "first," "second," "third," "fourth," and "fifth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0152] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0153] Although embodiments of the present invention have been shown and described above, the scope of protection of the present invention is not limited thereto. It is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, and substitutions to the above embodiments within the scope of the present invention, and such changes, modifications, and substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A motion detection method, characterized in that, The motion detection method includes: Continuously acquire channel impulse response data collected by multiple ultra-wideband radars; For each of the ultra-wideband radars, a candidate action corresponding to the channel impulse response data is determined based on the relative position between the ultra-wideband radar and the target detection area. If the channel impulse response data acquired by at least two of the ultra-wideband radars meet the identification conditions of the same candidate action, then the candidate action is determined as the target action to be detected.

2. The motion detection method as described in claim 1, characterized in that, The target detection area is the tailgate of the vehicle. The number of ultra-wideband radars is 2. The ultra-wideband radars include a first ultra-wideband radar and a second ultra-wideband radar. The first ultra-wideband radar and the second ultra-wideband radar are respectively arranged on both sides of the tailgate of the vehicle.

3. The motion detection method as described in claim 1, characterized in that, The step of ensuring that the channel impulse response data acquired by at least two of the ultra-wideband radars meet the identification conditions for the same candidate action includes: If a candidate action is detected based on the channel impulse response data collected by any of the ultra-wideband radars, a second detection region is determined, wherein the second detection region is the region where the first detection region of the ultra-wideband radar that detected the candidate action intersects with the first detection regions of other ultra-wideband radars. Based on the detection data of each of the ultra-wideband radars for the corresponding second detection area, it is determined whether there are at least two ultra-wideband radars that collect channel impulse response data that satisfy the identification conditions of the same candidate action.

4. The motion detection method as described in claim 3, characterized in that, The step of determining whether there are at least two channels impulse response data collected by the ultra-wideband radars that satisfy the identification condition of the same candidate action based on the detection data of the second detection area by each of the ultra-wideband radars includes: The arrangement of motion features obtained by each of the ultra-wideband radars detecting the corresponding second detection area is acquired; For each of the ultra-wideband radars, determine whether the arrangement of motion features satisfies the preset feature arrangement corresponding to the candidate action; If the motion feature arrangements of at least two of the ultra-wideband radars both satisfy the preset feature arrangement corresponding to the candidate action, then it is determined that the channel impulse response data collected by at least two of the ultra-wideband radars satisfy the identification condition of the same candidate action.

5. The motion detection method as described in claim 1, characterized in that, If the channel impulse response data acquired by at least two of the ultra-wideband radars satisfy the identification condition of the same candidate action, then determining the candidate action as the detected target action includes: If the channel impulse response data acquired by at least two of the ultra-wideband radars satisfy the identification condition of the same candidate action, then for the channel impulse response data of the ultra-wideband radars that satisfy the identification condition of the same candidate action, a motion feature arrangement of the candidate action is determined, wherein the motion feature arrangement includes multiple motion features arranged in the acquisition order. For each motion feature, determine whether the time periods of the motion feature corresponding to all the channel impulse response data are consistent; If the time period of the motion feature corresponding to all the channel impulse response data is consistent, then the candidate action is determined as the target action to be detected.

6. The motion detection method as described in claim 1, characterized in that, The process of determining the candidate action corresponding to the channel impulse response data includes: Determine the directional velocity characteristics corresponding to the channel impulse response data; The motion characteristics of the target object are determined based on the directional velocity characteristics; Match candidate actions corresponding to the motion features.

7. The motion detection method as described in claim 6, characterized in that, The directional velocity feature includes the object's direction and object's velocity; determining the motion features of the target object based on the directional velocity feature includes: Identify multiple directional velocity features that are continuous, have the same direction, and have a velocity greater than a preset velocity threshold. Determine whether the duration of the multiple directional velocity features is greater than a preset time threshold. If the duration of multiple directional velocity features is greater than the preset time threshold, then the detected motion feature is determined to be a unidirectional movement feature, wherein the direction of the unidirectional movement feature is the object direction corresponding to the multiple directional velocity features.

8. The motion detection method as described in claim 6, characterized in that, The candidate actions matched with the motion features include: Obtain a plurality of continuously defined motion features; The motion features are arranged according to the detection order of the motion features; In the preset kicking motion, a matching kicking motion is matched with a preset feature arrangement that is consistent with the motion feature arrangement; The matching kicking motion is selected as the candidate motion.

9. The motion detection method as described in claim 8, characterized in that, The step of arranging the motion features according to the detection order of the motion features includes: Obtain the motion cycle features of the motion feature arrangement; Determine whether the action cycle characteristics meet the action recognition conditions; If the motion cycle feature satisfies the motion recognition condition, then the matching kicking motion is arranged and matched based on the motion feature.

10. The motion detection method as described in claim 9, characterized in that, Determining whether the action cycle feature satisfies the action recognition condition includes: Obtain the maximum speed, maximum acceleration, and cycle duration from the action cycle characteristics; Determine whether the maximum speed is within a preset speed range, whether the maximum acceleration is greater than a preset acceleration threshold, and whether the period duration is less than a preset duration threshold; If the maximum speed is within a preset speed range, the maximum acceleration is greater than a preset acceleration threshold, and the period duration is less than a preset duration threshold, then the action recognition condition is met.

11. A motion detection system, characterized in that, It includes a processor and a memory connected to the processor, the memory storing the program code of the data acquisition module, the program code of the candidate action acquisition module, and the program code of the target action recognition module; The processor is configured to execute the data acquisition module program code to continuously acquire channel impulse response data collected by multiple ultra-wideband radars; The processor is also configured to execute candidate action acquisition module code to determine, for each of the ultra-wideband radars, a candidate action corresponding to the channel impulse response data based on the relative position between the ultra-wideband radar and the target detection area. The processor is also configured to execute target action recognition module code to determine the candidate action as the detected target action if the channel impulse response data acquired by at least two of the ultra-wideband radars satisfy the recognition conditions of the same candidate action.

12. A motion detection device, characterized in that, The motion detection device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the motion detection method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the motion detection method as described in any one of claims 1 to 10.

Images