Human body detection method for elevator non-contact calling

Abstract

The embodiment of the invention discloses a human body detection method for elevator non-contact calling, and the method comprises the steps: obtaining each frame of data of an intermediate frequency signal of a human body target, and carrying out the fast Fourier transform and CFAR detection to obtain first point cloud data; effective points are screened out according to whether target points in the first point cloud data meet detection distance and energy threshold requirements or not to obtain second point cloud data, target effective points are screened out based on the second point cloud data, the number of clustering targets is obtained according to the number of the target effective points, and motion parameters are calculated; and according to the number of the clustered targets and the motion parameters, determining a call landing triggering count, if the call landing triggering count is greater than or equal to a call landing triggering threshold value, controlling an elevator to go to a target floor, and associating screening of effective points with a triggering distance, thereby reducing interference of a long-distance target in a radar beam coverage range on call landing triggering. And the call landing triggering counting is confirmed according to the motion parameters, so that the possibility of false triggering caused by normal passing of the personnel is reduced, and the accuracy and the stability of call landing triggering are improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of passenger elevators, and in particular to a human body detection method for contactless elevator calls. Background Art

[0002] With the improvement of people's living standards and the continuous advancement of the intelligent process, smart elevators have emerged, especially those that can realize contactless elevator calls, which can not only prevent the spread of viruses, but also make people's lives more comfortable and convenient. In the related technologies, there are solutions that use high-definition cameras and infrared to achieve contactless elevator calls, solutions that use intelligent voice control systems to achieve contactless elevator calls, and solutions that use radars to achieve contactless elevator calls, but these solutions are more susceptible to false triggering due to the influence of the surrounding environment. Summary of the invention

[0003] The following is a summary of the subject matter of the detailed description of the present disclosure. This summary is not intended to limit the scope of the claims.

[0004] The disclosed embodiment provides a human body detection method for contactless elevator calling, which can improve the accuracy of elevator calling triggering and improve the stability and reliability of elevator calling triggering.

[0005] An embodiment of the first aspect of the present disclosure is a method for detecting a human body in a contactless elevator call, comprising:

[0006] Acquire each frame of intermediate frequency signal of the human target, perform fast Fourier transform and CFAR detection on each frame of data, and obtain first point cloud data;

[0007] Filter out valid points according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, and obtain second point cloud data of the valid points;

[0008] Filtering out valid target points based on the second point cloud data, obtaining the number of clustered targets according to the number of valid target points, and calculating motion parameters;

[0009] Determining an elevator call trigger count according to the number of cluster targets and the motion parameter;

[0010] If the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator is controlled to go to the target floor.

[0011] According to certain embodiments of the first aspect of the present disclosure, the intermediate frequency signal has multiple reference channels, each frame of data includes multiple sub-data corresponding to the reference channels, and performing fast Fourier transform and CFAR detection on each frame of data to obtain the first point cloud data includes:

[0012] Performing one-dimensional fast Fourier transform and two-dimensional fast Fourier transform on the sub-data corresponding to each reference channel to obtain a two-dimensional matrix of distance and Doppler corresponding to each reference channel;

[0013] Performing the CFAR detection after taking the modulus of the distance and Doppler two-dimensional matrix corresponding to the detection channel to obtain third point cloud data, wherein the detection channel is determined from the plurality of reference channels based on a preset condition;

[0014] Based on the third point cloud data, sequentially performing a three-dimensional fast Fourier transform and a four-dimensional fast Fourier transform on the distance and Doppler two-dimensional matrix corresponding to each reference channel to obtain fourth point cloud data;

[0015] Coordinate transformation is performed on the third point cloud data and the fourth point cloud data to obtain the first point cloud data.

[0016] According to some embodiments of the first aspect of the present disclosure, the first point cloud data includes a set of target points, a first distance lateral component of the target points, a first distance longitudinal component of the target points, a first distance height component of the target points, a first speed of the target points, and a first energy of the target points;

[0017] The detection distance and energy threshold include the maximum horizontal detection distance, the minimum vertical detection distance, the maximum vertical detection distance, the minimum detection height, the maximum detection height, and the target point energy threshold;

[0018] The step of selecting valid points according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements to obtain second point cloud data of the valid points includes:

[0019] If the first distance lateral component of the target point in the target point set is between the negative value of the maximum lateral detection distance in a closed interval and the positive value of the maximum lateral detection distance, the first distance longitudinal component is between the minimum longitudinal detection distance in a closed interval and the maximum longitudinal detection distance, the first distance height component is between the minimum detection height and the maximum detection height in a closed interval, and the first energy is greater than or equal to the target point energy threshold, then the corresponding target point is set as a valid point, and the valid points are combined into a valid point set. Based on the first distance lateral component corresponding to the target point, the first distance longitudinal component corresponding to the target point, the first speed corresponding to the target point, and the first energy corresponding to the target point, the second distance lateral component of the valid point, the second distance longitudinal component of the valid point, the second speed of the valid point, and the second energy of the valid point are obtained, and the valid point set, the second distance lateral component, the second distance longitudinal component, the second speed, and the second energy are combined into the second point cloud data.

[0020] According to certain embodiments of the first aspect of the present disclosure, the motion parameters include a lateral weighted average distance of the target effective point, a longitudinal weighted average distance of the target effective point, and a weighted average speed of the target effective point;

[0021] The step of screening out target valid points based on the second point cloud data, obtaining the number of clustered targets according to the number of target valid points, and calculating motion parameters includes:

[0022] Based on the second point cloud data, in a first iteration, the valid point with the maximum second energy and a clustering flag of 0 is selected from the valid point set as a centroid point, the valid point within a preset distance range from the centroid point is selected from the valid point set as an associated distance point, the centroid point and the associated distance point are used as the target valid points, and the clustering flags of the valid points corresponding to the centroid point and the associated distance point are set to 1;

[0023] Count the number of target valid points of this iteration. If the number of points is greater than or equal to 2, add 1 to the number of clustered targets, and calculate the horizontal weighted average distance of the target valid points, the vertical weighted average distance of the target valid points, and the weighted average speed of the target valid points of this iteration;

[0024] The loop is iterated until the clustering flags of the valid points in the valid point set are all 1, or the number of clustering targets reaches the maximum number of clustering targets in a single frame.

[0025] According to certain embodiments of the first aspect of the present disclosure, the calculating of the lateral weighted average distance of the target effective point, the longitudinal weighted average distance of the target effective point, and the weighted average speed of the target effective point of this iteration includes:

[0026] In this iteration, a third distance lateral component of the target effective point, a third distance longitudinal component of the target effective point, a third velocity of the target effective point, and a third energy of the target effective point are obtained based on the second distance lateral component, the second distance longitudinal component, the second velocity, and the second energy;

[0027] summing the products of each of the third distance transverse components multiplied by each of the third energies to obtain a first intermediate value;

[0028] summing the products of each of the third distance longitudinal components multiplied by each of the third energies to obtain a second intermediate value;

[0029] summing the products of each of the third speeds and each of the third energies to obtain a third intermediate value;

[0030] summing up the third energies to obtain a fourth intermediate value;

[0031] The first intermediate value is divided by the fourth intermediate value to obtain the lateral weighted average distance, the second intermediate value is divided by the fourth intermediate value to obtain the longitudinal weighted average distance, and the third intermediate value is divided by the fourth intermediate value to obtain the weighted average speed.

[0032] According to certain embodiments of the first aspect of the present disclosure, the step of confirming the elevator call trigger count according to the number of cluster targets and the motion parameter includes:

[0033] If the number of cluster targets is greater than 0, and the horizontal weighted average distance is between the negative value of the horizontal trigger distance in the closed interval and the positive value of the horizontal trigger distance, the vertical weighted average distance is less than or equal to the vertical trigger distance, and the weighted average speed is less than or equal to the trigger speed upper limit, then the elevator call trigger count is increased by 1;

[0034] If all the target valid points satisfy that the absolute value of the horizontal weighted average distance is greater than the horizontal protection distance, or if all the target valid points satisfy that the vertical weighted average distance is greater than the vertical protection distance, then the elevator call trigger count is cleared;

[0035] If the number of clustered targets is equal to 0, the target loss count is increased by 1. If the target loss count is greater than the maximum allowed number of lost frames, the elevator call trigger count is cleared.

[0036] According to certain embodiments of the first aspect of the present disclosure, if the elevator call trigger count is greater than or equal to the elevator call trigger threshold, controlling the elevator to go to the target floor includes:

[0037] If the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator call command and the status query command are sent in a periodic cycle, and the indicator light is turned on. The elevator is controlled to go to the target floor according to the elevator call command, and the elevator status is obtained according to the status query command. If the elevator status is that the elevator has arrived at the target floor and is in the process of opening the door or the door opening is completed, the indicator light is turned off, the elevator call command is stopped from being reported, and the status query command is continued to be sent in a periodic cycle. The elevator status is obtained according to the status query command. If the elevator status is that the elevator has completed closing the door at the target floor, the status query command is stopped from being sent, and the next detection is re-entered after waiting for the preset protection time.

[0038] An embodiment of the second aspect of the present disclosure is an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the human body detection method for contactless elevator calls as described above.

[0039] An embodiment of the third aspect of the present disclosure is a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the human body detection method for contactless elevator calls as described above is implemented.

[0040] The disclosed embodiment includes at least the following beneficial effects: first, each frame of the intermediate frequency signal of the human target is obtained, and each frame of the data is subjected to fast Fourier transform and CFAR detection to obtain the first point cloud data; then, valid points are screened out according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, and the second point cloud data of valid points are obtained; based on the second point cloud data, valid points of the target are screened out, and the number of clustered targets is obtained according to the number of valid points of the target, and the motion parameters are calculated; according to the number of clustered targets and the motion parameters, the elevator call trigger count is confirmed; if the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator is controlled to go to the target floor. Valid points are screened according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, and the screening of valid points is associated with the trigger distance, which minimizes the interference of the elevator call trigger by the long-distance target within the coverage range of the radar beam, and improves the accuracy of the elevator call trigger. The elevator call trigger count is confirmed according to the motion parameters, which reduces the possibility of false triggering caused by normal passing of personnel, and improves the stability and reliability of the elevator call trigger.

[0041] Other features and advantages of the present disclosure will be set forth in the following description, and in part will be apparent from the description, or may be learned by practicing the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings are used to provide further understanding of the technical solution of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the technical solution of the present disclosure and do not constitute a limitation on the technical solution of the present disclosure.

[0043] Figure 1 An optional flow chart of a human body detection method for contactless elevator call provided in an embodiment of the present disclosure;

[0044] Figure 2 A schematic top view of a detection scene provided in an embodiment of the present disclosure;

[0045] Figure 3 A schematic diagram of an elevator call detection algorithm provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present disclosure more clear, the present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure and are not used to limit the present disclosure.

[0047] In the embodiments of the present disclosure, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present disclosure belongs. The terms used herein are only for the purpose of describing the embodiments of the present disclosure and are not intended to limit the present disclosure.

[0049] To facilitate understanding of the technical solution provided by the embodiments of the present disclosure, some key terms used in the embodiments of the present disclosure are explained here:

[0050] Constant False Alarm Rate Detector (CFAR) is an important technology in radar signal processing. CFAR detection aims to maintain a fixed false alarm rate, that is, in the absence of target signals, the probability of a false alarm generated by the detection algorithm is kept at a predetermined level. This is achieved by dynamically adjusting the detection threshold in a complex and changeable interference environment to ensure the detection performance of the radar system under different signal conditions.

[0051] The embodiments of the present disclosure are further described below in conjunction with the accompanying drawings.

[0052] Reference Figure 1 As shown, Figure 1 An optional flow chart of a method for detecting human bodies in a contactless elevator call is provided in an embodiment of the present disclosure. The method for detecting human bodies in a contactless elevator call includes but is not limited to steps 101 to 105. Each step is introduced in turn below.

[0053] Step 101: acquiring each frame of intermediate frequency signal of a human target, performing fast Fourier transform and CFAR detection on each frame of data, and obtaining first point cloud data.

[0054] Specifically, the radar echo signal of the human target is obtained, and the radar echo signal is mixed to obtain the intermediate frequency signal. Among them, the radar echo signal is the signal received by the radar receiver after the radar transmission signal is reflected and scattered after encountering the target object. Fast Fourier transform is an algorithm for efficiently calculating discrete Fourier transform and its inverse transform. Discrete Fourier transform is a mathematical tool used to convert signals from time domain to frequency domain. However, the complexity of directly calculating discrete Fourier transform is high, especially when the number of sampling points of the signal is large, the amount of calculation will increase sharply. Therefore, people have proposed the fast Fourier transform algorithm to greatly reduce the amount of calculation.

[0055] In some embodiments, the intermediate frequency signal has multiple reference channels, each frame of data includes sub-data corresponding to the multiple reference channels, and each frame of data is subjected to fast Fourier transform and CFAR detection to obtain first point cloud data, including:

[0056] Perform one-dimensional fast Fourier transform and two-dimensional fast Fourier transform on the sub-data corresponding to each reference channel to obtain the distance and Doppler two-dimensional matrix corresponding to each reference channel;

[0057] Perform CFAR detection on the distance corresponding to the detection channel and the Doppler two-dimensional matrix after taking the modulus, and obtain the third point cloud data, wherein the detection channel is determined from multiple reference channels based on preset conditions;

[0058] Based on the third point cloud data, three-dimensional fast Fourier transform and four-dimensional fast Fourier transform are sequentially performed on the distance and Doppler two-dimensional matrix corresponding to each reference channel to obtain fourth point cloud data;

[0059] Coordinate transformation is performed on the third point cloud data and the fourth point cloud data to obtain first point cloud data.

[0060] Specifically, the one-dimensional fast Fourier transform is the application of the fast Fourier transform on a one-dimensional signal, which is used to convert a one-dimensional time domain signal into a frequency domain signal, or to convert a frequency domain signal back to a time domain signal. The two-dimensional fast Fourier transform is the application of the fast Fourier transform on a two-dimensional signal or image, which is used to convert a two-dimensional space domain signal into a frequency domain signal, or to convert a frequency domain signal back to a space domain signal. The three-dimensional fast Fourier transform is the application of the fast Fourier transform on a three-dimensional signal or data, which is used to convert a three-dimensional space domain signal into a frequency domain signal, or to convert a frequency domain signal back to a space domain signal. The four-dimensional fast Fourier transform is the application of the fast Fourier transform on a four-dimensional signal or data, which is used to convert a four-dimensional space domain signal into a frequency domain signal, or to convert a frequency domain signal back to a space domain signal.

[0061] It is understandable that for each frame of data, a one-dimensional fast Fourier transform is performed, and then the mean or background noise after the one-dimensional fast Fourier transform of the previous frame is subtracted, and then a two-dimensional fast Fourier transform is performed to obtain the distance and Doppler two-dimensional matrix corresponding to each reference channel. Among them, the background noise refers to the mean after the one-dimensional fast Fourier transform when there is no target, and its value is continuously updated. Perform a one-dimensional fast Fourier transform and a two-dimensional fast Fourier transform, and subtract the mean or background noise after the one-dimensional fast Fourier transform of the previous frame after the one-dimensional fast Fourier transform. In this way, human targets standing still after entering the trigger area of ​​the contactless elevator call can be detected normally, and human targets within the detection range can be continuously tracked without being lost. At the same time, static interference targets in the environment can be filtered out, reducing the possibility of false triggering, and improving the stability and reliability of the elevator call trigger.

[0062] Specifically, the distance and Doppler two-dimensional matrix corresponding to the detection channel is modulo-measured and then CFAR detection is performed to obtain the third point cloud data. Based on the third point cloud data, the distance and Doppler two-dimensional matrix corresponding to each reference channel is sequentially subjected to three-dimensional fast Fourier transform and four-dimensional fast Fourier transform to obtain the fourth point cloud data. The third point cloud data and the fourth point cloud data are subjected to coordinate transformation to obtain the first point cloud data. Among them, CFAR detection refers to taking the signals of the two units in the sliding window with a length of n before and after one protection unit in the detected unit, and then removing some of the maximum values ​​of the signals of the two units and averaging them, and then selecting the maximum of the averaged signal and multiplying it by the threshold factor as the detection threshold, and finally comparing the detected unit with the detection threshold and outputting it as a constant false alarm.

[0063] Step 102: Filter out valid points according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, and obtain second point cloud data of the valid points.

[0064] Specifically, valid points are screened out according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, and second point cloud data of valid points are obtained, which can realize multi-target detection. Valid points are screened according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements. The screening of valid points is related to the triggering distance, which minimizes the interference of elevator call triggering by long-distance targets within the coverage range of the radar beam, thereby improving the accuracy of elevator call triggering.

[0065] In some embodiments, the first point cloud data includes a set of target points, a first distance lateral component of the target points, a first distance longitudinal component of the target points, a first distance height component of the target points, a first speed of the target points, and a first energy of the target points;

[0066] The detection distance and energy threshold include the maximum horizontal detection distance, the minimum vertical detection distance, the maximum vertical detection distance, the minimum detection height, the maximum detection height, and the target point energy threshold;

[0067] Valid points are screened out according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, and second point cloud data of valid points is obtained, including:

[0068] If the first distance lateral component of the target point in the target point set is between the negative value of the maximum lateral detection distance in the closed interval and the positive value of the maximum lateral detection distance, the first distance longitudinal component is between the minimum longitudinal detection distance in the closed interval and the maximum longitudinal detection distance, the first distance height component is between the minimum detection height in the closed interval and the maximum detection height, and the first energy is greater than or equal to the target point energy threshold, then the corresponding target point is set as a valid point, and the valid points are combined into a valid point set. Based on the first distance lateral component of the corresponding target point, the first distance longitudinal component of the corresponding target point, the first speed of the corresponding target point, and the first energy of the corresponding target point, the second distance lateral component of the valid point, the second distance longitudinal component of the valid point, the second speed of the valid point, and the second energy of the valid point are obtained, and the valid point set, the second distance lateral component, the second distance longitudinal component, the second speed, and the second energy are combined into second point cloud data.

[0069] Specifically, if the first distance lateral component of the target point in the target point set is between the negative value of the maximum lateral detection distance in the closed interval and the positive value of the maximum lateral detection distance, the first distance longitudinal component is between the minimum longitudinal detection distance in the closed interval and the maximum longitudinal detection distance, the first distance height component is between the minimum detection height in the closed interval and the maximum detection height, and the first energy is greater than or equal to the target point energy threshold, that is, the target point in the first point cloud data meets the detection distance and energy threshold requirements, then the corresponding target point is set as a valid point. It can be understood that the effective points are screened according to the maximum lateral detection distance, the minimum longitudinal detection distance, the maximum longitudinal detection distance, the minimum detection height, and the maximum detection height, and the screening of the effective points is associated with the triggering distance, which minimizes the interference of the elevator call trigger by the long-distance targets within the coverage range of the radar beam, and improves the accuracy of the elevator call trigger.

[0070] Step 103: Filter out valid target points based on the second point cloud data, obtain the number of clustered targets according to the number of valid target points, and calculate motion parameters.

[0071] Specifically, the second point cloud data includes a valid point set, a second distance transverse component, a second distance longitudinal component, a second speed, and a second energy.

[0072] In some embodiments, the motion parameters include a lateral weighted average distance of the target effective point, a longitudinal weighted average distance of the target effective point, and a weighted average speed of the target effective point;

[0073] Filter out valid target points based on the second point cloud data, obtain the number of clustered targets according to the number of valid target points, and calculate motion parameters, including:

[0074] Based on the second point cloud data, in the first iteration, a valid point with the largest second energy and a clustering flag of 0 is selected from the valid point set as the centroid point, and a valid point within a preset distance range from the centroid point is selected from the valid point set as an associated distance point, and the centroid point and the associated distance point are used as target valid points, and the clustering flags of the valid points corresponding to the centroid point and the associated distance point are set to 1;

[0075] Count the number of target effective points of this iteration. If the number of points is greater than or equal to 2, the number of clustered targets is increased by 1, and the horizontal weighted average distance of the target effective points of this iteration, the vertical weighted average distance of the target effective points, and the weighted average speed of the target effective points are calculated;

[0076] The loop is iterated until the clustering flags of the valid points in the valid point set are all 1, or the number of clustering targets reaches the maximum number of clustering targets in a single frame.

[0077] It is understandable that the first iteration screens out the target effective points of the first iteration, and when the number of target effective points is greater than or equal to 2, the horizontal weighted average distance of the target effective points of the first iteration, the vertical weighted average distance of the target effective points, and the weighted average speed of the target effective points are calculated. After multiple iterations, the horizontal weighted average distance, the vertical weighted average distance, and the weighted average speed of all target effective points calculated for multiple iterations are obtained. The weighted average speed of the target effective points is calculated, and the elevator call trigger count is subsequently confirmed based on the weighted average speed of the target effective points, which can reduce the possibility of false triggering caused by normal passage of personnel and improve the stability and reliability of the elevator call trigger.

[0078] Specifically, the number of target effective points of this iteration is counted. If the number of points is less than 2, there is no need to calculate the lateral weighted average distance of the target effective points of this iteration, the longitudinal weighted average distance of the target effective points, and the weighted average speed of the target effective points.

[0079] In some embodiments, calculating the lateral weighted average distance of the target effective points, the longitudinal weighted average distance of the target effective points, and the weighted average speed of the target effective points of this iteration includes:

[0080] In this iteration, a third distance lateral component of the target effective point, a third distance longitudinal component of the target effective point, a third velocity of the target effective point, and a third energy of the target effective point are obtained based on the second distance lateral component, the second distance longitudinal component, the second velocity, and the second energy;

[0081] summing the products of each third distance lateral component multiplied by each third energy to obtain a first intermediate value;

[0082] summing the products of the respective third distance longitudinal components multiplied by the respective third energies to obtain a second intermediate value;

[0083] summing the products of each third speed multiplied by each third energy to obtain a third intermediate value;

[0084] summing up the third energies to obtain a fourth intermediate value;

[0085] The first intermediate value is divided by the fourth intermediate value to obtain the lateral weighted average distance, the second intermediate value is divided by the fourth intermediate value to obtain the longitudinal weighted average distance, and the third intermediate value is divided by the fourth intermediate value to obtain the weighted average speed.

[0086] Specifically, first, the third energies are summed to obtain the fourth intermediate value, then the products of the lateral components of the third distances and the third energies are summed to obtain the first intermediate value, the products of the longitudinal components of the third distances and the third energies are summed to obtain the second intermediate value, and the products of the third speeds and the third energies are summed to obtain the third intermediate value. Finally, the first intermediate value is divided by the fourth intermediate value to obtain the lateral weighted average distance, the second intermediate value is divided by the fourth intermediate value to obtain the longitudinal weighted average distance, and the third intermediate value is divided by the fourth intermediate value to obtain the weighted average speed. It can be understood that the weighted average speed of the target effective point is calculated, and the elevator call trigger count is subsequently confirmed according to the weighted average speed of the target effective point, which can reduce the possibility of false triggering caused by normal passing of personnel and improve the stability and reliability of the elevator call trigger. The lateral weighted average distance of the target effective point and the longitudinal weighted average distance of the target effective point are calculated, and the confirmation of the elevator call trigger count can be associated with the trigger distance, thereby improving the elevator call trigger response speed and stability and reducing jitter.

[0087] Step 104: confirm the elevator call trigger count according to the number of clustered targets and the motion parameters.

[0088] Specifically, the motion parameters include a lateral weighted average distance of the target effective points, a longitudinal weighted average distance of the target effective points, and a weighted average speed of the target effective points.

[0089] In some embodiments, according to the number of clustered targets and the motion parameters, the elevator call trigger count is determined, including:

[0090] If the number of cluster targets is greater than 0, and the horizontal weighted average distance is between the negative value of the horizontal trigger distance and the positive value of the horizontal trigger distance in the closed interval, the vertical weighted average distance is less than or equal to the vertical trigger distance, and the weighted average speed is less than or equal to the trigger speed upper limit, then the elevator call trigger count is increased by 1;

[0091] If all target valid points satisfy that the absolute value of the horizontal weighted average distance is greater than the horizontal protection distance, or all target valid points satisfy that the vertical weighted average distance is greater than the vertical protection distance, the elevator call trigger count is cleared;

[0092] If the number of clustered targets is equal to 0, the target loss count is increased by 1. If the target loss count is greater than the maximum allowed number of lost frames, the elevator call trigger count is cleared.

[0093] It can be understood that when the number of cluster targets is greater than 0, the elevator call trigger count is confirmed based on whether the horizontal weighted average distance is between the negative value of the horizontal trigger distance in the closed interval and the positive value of the horizontal trigger distance, the vertical weighted average distance is less than or equal to the vertical trigger distance, and the weighted average speed is less than or equal to the trigger speed upper limit. The confirmation of the elevator call trigger count is associated with the trigger distance to improve the elevator call trigger response speed and stability and reduce jitter. In addition, the speed requirement is introduced in the confirmation of the elevator call trigger count, which can ensure timely response while reducing the possibility of false triggering caused by normal personnel passing by, and improve the stability and reliability of the elevator call trigger.

[0094] In some embodiments, reference Figure 2 As shown, Figure 2 A schematic top view of a detection scene provided in an embodiment of the present disclosure, Figure 2 Rx max is the maximum lateral detection distance, Ry min is the minimum longitudinal detection distance, Rx tri is the lateral trigger distance, Ry tri is the longitudinal trigger distance, Rx pro is the lateral protection distance, Ry pro is the longitudinal protection distance, Ry max is the maximum longitudinal detection distance, Rx avr is the lateral weighted average distance, Ry avr It is the longitudinal weighted average distance. The detection zone, protection zone and trigger zone are confirmed based on these data.

[0095] Specifically, the lateral protection distance is equal to 1.2 times the lateral trigger distance, the longitudinal protection distance is equal to 1.2 times the longitudinal trigger distance, the lateral maximum detection distance is equal to 1.3 times the lateral trigger distance, and the longitudinal maximum detection distance is equal to 1.3 times the longitudinal trigger distance.

[0096] Step 105: If the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator is controlled to go to the target floor.

[0097] In some embodiments, if the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator is controlled to go to the target floor, including: if the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator call command and the status query command are sent periodically and the indicator light is turned on, the elevator is controlled to go to the target floor according to the elevator call command, and the elevator status is obtained according to the status query command. If the elevator status is that the elevator has reached the target floor and is in the process of opening the door or the door opening is completed, the indicator light is turned off, the elevator call command is stopped from being reported, and the status query command is continued to be sent periodically and cycled, and the elevator status is obtained according to the status query command. If the elevator status is that the elevator has completed closing the door on the target floor, the status query command is stopped, and the next detection is re-entered after waiting for the preset protection time.

[0098] It is understandable that the indicator light is lit to remind the elevator user that the elevator call is triggered successfully and the elevator is heading to the target floor. The indicator light is turned off to indicate that the elevator has reached the target floor and is in the state of opening the door or the door opening is completed. When the elevator status is that the elevator has reached the target floor and is in the state of opening the door or the door opening is completed, the elevator call command is stopped from being reported, and the status query command is continued to be sent periodically. When the elevator status is that the elevator has closed the door at the target floor, the status query command is stopped from being sent, and the next detection is re-entered after waiting for the preset protection time. The indicator light can timely remind the elevator user whether the elevator call is triggered successfully. By sending the status query command, the elevator status can be tracked in real time, and the next step can be determined according to the elevator status, which improves the rate of the process from the successful elevator call trigger to the elevator reaching the target floor and the door closing is completed to the next detection, and the elevator scheduling strategy can also be optimized according to the elevator status.

[0099] It can be seen that each frame of the intermediate frequency signal of the human target is obtained, and each frame of data is fast Fourier transformed and CFAR detected to obtain the first point cloud data. According to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, valid points are screened out to obtain the second point cloud data of valid points; based on the second point cloud data, valid target points are screened out, and the number of clustered targets is obtained according to the number of valid target points, and the motion parameters are calculated; according to the number of clustered targets and motion parameters, the elevator call trigger count is confirmed; if the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator is controlled to go to the target floor. According to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, valid points are screened. The screening of valid points is associated with the trigger distance, which minimizes the interference of long-distance targets within the coverage range of the radar beam on the elevator call trigger, and improves the accuracy of the elevator call trigger. The elevator call trigger count is confirmed according to the motion parameters, which reduces the possibility of false triggering caused by normal passing of personnel, and improves the stability and reliability of the elevator call trigger.

[0100] In some embodiments, the elevator contactless call device includes a 60GHz FMCW millimeter wave radar with 4 transmit and 4 receive, an ADC acquisition module, a 485 serial port, and a processor. Among them, 4 transmit and 4 receive refer to 4 transmitting antennas and 4 receiving antennas, with a total of 16 channels. The FMCW millimeter wave radar has a distance resolution of 5.6cm, a speed resolution of 0.2m / s, and a frame period of 100ms. Among them, the channel with the best signal quality is selected from the data of the 16 channels obtained as the detection channel.

[0101] Initialize the basic configuration of each peripheral module. The elevator control unit can send parameter configuration data about data processing to the elevator contactless call device through the 485 serial port. The radar chip is configured for RF through the MCU SPI, so that the radar antenna transmits the designed FMCW signal. The FMCW millimeter wave radar sends a transmission signal (i.e., the aforementioned FMCW signal), uses the ADC acquisition module of the MCU to collect the reflected radar echo signal, and mixes the radar echo signal to obtain an intermediate frequency signal. The next step is the processor to process the intermediate frequency signal. The intermediate frequency signal has multiple reference channels, and each frame of data includes sub-data corresponding to multiple reference channels. For the intermediate frequency signal of the human target, the processor performs a one-dimensional fast Fourier transform on the sub-data corresponding to each reference channel, and then subtracts the mean value or background noise after the one-dimensional fast Fourier transform of the previous frame, where the mean value or background noise after the one-dimensional fast Fourier transform of the previous frame is subtracted from the mean value after the one-dimensional fast Fourier transform of the previous frame for each elevator call trigger count, and then subtracts the background noise. Then, a two-dimensional fast Fourier transform is performed to obtain a two-dimensional matrix of distance and Doppler. The two-dimensional matrix of distance and Doppler corresponding to the detection channel is modulo and then CFAR detection is performed to obtain the third point cloud data, which refers to the distance index, velocity index and energy value. Based on the third point cloud data, the two-dimensional matrix of distance and Doppler corresponding to each reference channel is sequentially subjected to a three-dimensional fast Fourier transform and a four-dimensional fast Fourier transform to obtain the fourth point cloud data, which refers to the azimuth index and the pitch index. The third point cloud data and the fourth point cloud data are subjected to coordinate conversion to obtain the first point cloud data with spatial coordinate information. According to whether the target points in the first point cloud data meet the requirements of the detection distance and energy threshold, valid points are screened out to obtain the second point cloud data of valid points; based on the second point cloud data, valid target points are screened out, the number of clustered targets is obtained according to the number of valid target points, and the motion parameters are calculated; according to the number of clustered targets and the motion parameters, the elevator call trigger count is confirmed.

[0102] Determine whether the elevator call trigger count is greater than or equal to the elevator call trigger threshold. If the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator contactless elevator call device periodically sends the elevator call command and status query command to the elevator control unit through the 485 serial port, and lights up the indicator light; the elevator control unit controls the elevator to the target floor according to the elevator call command; the FMCW millimeter-wave radar receives and analyzes the elevator status replied by the elevator control unit. If the elevator status is that the elevator has reached the target floor and is in the process of opening the door or the door opening is completed, the FMCW millimeter-wave radar turns off the indicator light, stops reporting the elevator call command, and continues to send the status query command periodically. The FMCW millimeter-wave radar receives and analyzes the elevator status replied by the elevator control unit. If the elevator status is that the elevator has closed the door on the target floor, the FMCW millimeter-wave radar stops sending the status query command. The elevator contactless elevator call device waits for the preset protection time and then re-enters the next detection.

[0103] The elevator call detection algorithm is used in the process from confirming the elevator call trigger count based on the number of clustered targets and motion parameters to the elevator reaching the target floor and re-entering the next detection. Figure 3 As shown, Figure 3 A schematic diagram of an elevator call detection algorithm provided in an embodiment of the present disclosure, Figure 3 The number of targets N in is the number of clustering targets, V avr is the weighted average velocity, V tri is the trigger speed upper limit, TriCnt is the elevator call trigger count, TriThres is the elevator call trigger threshold, the radar floor is the target floor mentioned above, LostCnt is the target loss count, and LostCntMax is the maximum allowed number of lost frames.

[0104] Specifically, the horizontal trigger distance refers to the horizontal outer boundary distance of the trigger area, which is sent to the FMCW millimeter wave radar by the elevator control unit. The longitudinal trigger distance refers to the longitudinal outer boundary distance of the trigger area, which is sent to the FMCW millimeter wave radar by the elevator control unit. The horizontal protection distance refers to the horizontal outer boundary distance of the protection area, the longitudinal protection distance refers to the longitudinal outer boundary distance of the protection area, the horizontal maximum detection distance refers to the horizontal outer boundary distance of the detection area, and the longitudinal maximum detection distance refers to the longitudinal outer boundary distance of the detection area. The longitudinal minimum detection distance refers to the longitudinal inner boundary distance of the detection area, which is determined according to the actual installation position of the FMCW millimeter wave radar. The maximum detection height refers to the upper boundary distance of the detection area, which is determined according to the actual installation position of the FMCW millimeter wave radar. The minimum detection height refers to the lower boundary distance of the detection area, which is determined according to the actual installation position of the FMCW millimeter wave radar. The upper limit of the trigger speed is determined according to the demand and actual measurement, the target point energy threshold is determined according to the actual measurement and experience, the elevator call trigger threshold is determined according to the demand and actual measurement, and the maximum allowable number of lost frames is determined according to the actual measurement.

[0105] It is understandable that the 60GHz FMCW millimeter-wave radar with 4 transmitters and 4 receivers can better accommodate the two important indicators of data volume and resolution, and effectively reduce costs while ensuring performance, which is conducive to market promotion and implementation. In addition, based on a human body detection method for elevator contactless calling, elevator contactless calling is realized, which provides a good experience for elevator users. At the same time, the hardware cost used is low, the implementation is simple, the cost-effectiveness is high, and it is also easy to promote in a large area of ​​the market and promote the transfer and transformation of technological achievements.

[0106] The following is an actual example to illustrate the principle of the embodiment of the present application:

[0107] A 60GHz FMCW millimeter-wave radar with 4 transmit and 4 receive (4 transmit antennas, 4 receive antennas, 16 channels in total) is used, with a distance resolution of 5.6cm, a speed resolution of 0.2m / s, and a frame period of 100ms. The radar first configures the RF chip through the MCU SPI, uses the ADC acquisition module to receive the echo signal, and then performs FFT digital signal processing and CFAR detection to generate point cloud data with distance, speed, azimuth, pitch angle and energy information; then after coordinate conversion, valid points are selected according to a certain strategy, and the weighted average coordinates, speed and energy information are calculated; finally, the human body detection algorithm is used to determine the elevator call trigger, and the intelligent elevator is triggered through the 485 serial port to execute the elevator call action. The following describes the processing process of each frame of data.

[0108] First, initialization: power on each peripheral module, initialize the configuration, initialize the variables, and read and verify the configuration from FLASH (including the parameter of "radar location"). Second, parameter configuration: check whether the 485 serial port buffer has received the configuration data sent by the elevator control unit. If yes, update the configuration. Third, RF configuration: configure the radar chip through the SPI of the single-chip microcomputer to make the radar antenna emit the designed FMCW signal. Fourth, AD acquisition: use the ADC of the single-chip microcomputer to collect the intermediate frequency signal obtained after the radar echo signal is mixed and processed. Fifth, time-frequency conversion and CFAR detection: first, perform the distance dimension (one-dimensional) FFT fast Fourier transform on the sampling data of the 16 channels of the current frame in turn, and then subtract the mean value or background noise after the one-dimensional FFT of the previous frame. The background noise refers to the mean value after the one-dimensional FFT when there is no target, and continuously update it, and then perform the speed dimension (two-dimensional) FFT fast Fourier transform to generate the distance-Doppler two-dimensional matrix. Before doing the two-dimensional FFT, the mean value or background noise after the one-dimensional FFT of the previous frame is subtracted. Specifically, the mean value after the one-dimensional FFT of the previous frame is subtracted from the five frames before each elevator call trigger count, and the background noise is subsequently subtracted. The purpose of this is to be able to normally detect human targets standing still after entering the trigger area, while also filtering out interference targets in the environment and not triggering by mistake. Then, the distance-Doppler two-dimensional matrix corresponding to the pre-selected detection channel is modulo-measured and CFAR detection is performed to obtain the point cloud data of the human target, including the distance index, velocity index and energy value. The detection channel is a channel with the best signal quality selected in advance based on the time domain sampling data waveform of the 16 channels. Finally, according to the CFAR detection results, the distance-Doppler two-dimensional matrix corresponding to other reference channels is screened for targets, and the data corresponding to the screened target points are subjected to antenna dimension (three-dimensional and four-dimensional) FFT fast Fourier transform to obtain the azimuth index and pitch index of the human target point cloud data, and then through coordinate conversion, point cloud data with spatial coordinate information is generated. Each target point contains five characteristic information: "distance transverse component Rx, distance longitudinal component Ry, distance height component Rz, speed V, energy P". Among them, CFAR detection refers to taking the signals of the two units in the sliding window with a length of n before and after one protection unit in the detected unit, and then removing some of the maximum values ​​of the signals of these two units and averaging them, and then selecting the maximum of the averaged signal and multiplying it by the threshold factor as the detection threshold, and finally comparing the detected unit with the detection threshold and outputting it as a constant false alarm.

[0109] Sixth, filter out valid points based on point cloud data and perform target clustering. The steps are as follows:

[0110] Step 1: Check whether each point in the point cloud set meets the detection distance and energy threshold requirements, that is,

[0111] -Rxmax≤Rx≤Rxmax, Rymin≤Ry≤Rymax, Rzmin≤Rz≤Rzmax and P≥Pmin, if they do not meet the requirements, they will be discarded and the point cloud set will be updated;

[0112] Step 2: Find the point with the largest energy and a clustering flag of 0 in the new point cloud set as the centroid of one of the targets after clustering. Then traverse the new point cloud set to filter out the points whose distance difference with the centroid of this target meets the requirements. Finally, the centroid and the filtered points are used as the valid points of the target, and the clustering flags of these points are set to 1.

[0113] Step 3: Determine the number of valid points n of the target. If n<2, discard them directly. If n≥2, add 1 to the number of clustered targets and calculate the weighted average coordinates Rxavr, Ryavr and weighted average speed Vavr of all valid point coordinates Rx, Ry and speed V as the clustering result of the target. For example, if the number of valid points is 3, then Rxavr=(Rx1*P1+Rx2*P2+Rx3*P3) / (P1+P2+P3), Ryavr=(Ry1*P1+Ry2*P2+Ry3*P3) / (P1+P2+P3), Vavr=(V1*P1+V2*P2+V3*P3) / (P1+P2+P3);

[0114] Repeat steps 2 and 3 until all point clustering flags in the point cloud set are 1, or the number of clustered targets N reaches the maximum number of targets Nmax in a single frame.

[0115] Seventh, use the elevator call detection algorithm to determine the elevator call trigger and output the result. The steps are as follows:

[0116] Step 1: Update the elevator call trigger count TriCnt according to the number of clustered targets N, the target positions Rxavr, Ryavr and speeds Vavr, and the set trigger conditions:

[0117] If N>0, traverse each target point to determine whether there is a target that satisfies -Rxtri≤Rxavr≤Rxtri&&Ryavr≤Rytri&&Vavr≤Vtri, if yes, then TriCnt++;

[0118] Otherwise, if all target points satisfy |Rxavr|>Rxpro||Ryavr>Rypro, then TriCnt=0;

[0119] Otherwise, if N=0, the target loss count is LostCnt++. If the target loss count is greater than the maximum allowed number of lost frames, that is, LostCnt>LostCntMax, then TriCnt=0;

[0120] Step 2: Determine whether the elevator call trigger count meets the elevator call trigger threshold, that is, if TriCnt ≥ TriThres, periodically send the elevator call command and status query command to the intelligent elevator control unit through the serial port, and light up the indicator light at the same time;

[0121] Step 3: Immediately analyze the elevator status replied by the elevator control unit. If the elevator reaches the floor where the radar is located and is in the process of opening the door / opening the door, the radar turns off the indicator light, stops reporting the elevator call command, continues to periodically send the status query command, and analyzes the replied data;

[0122] If the elevator completes closing the door at the floor where the radar is located, the radar stops sending status query commands, waits for the set protection time, and then re-enters the next detection.

[0123] The embodiment of the present disclosure also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned human body detection method for contactless elevator call when executing the computer program. The electronic device can be any intelligent terminal including a computer.

[0124] The embodiment of the present disclosure further provides a storage medium storing a computer program, which implements the above-mentioned human body detection method for contactless elevator calling when the computer program is executed by a processor.

[0125] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate to describe the embodiments of the present disclosure, such as being able to be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0126] It should be understood that in the present disclosure, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0127] It should be understood that in the description of the embodiments of the present disclosure, the meaning of multiple (or multiple items) is more than two, greater than, less than, exceed, etc. are understood to not include the number, and above, below, within, etc. are understood to include the number.

[0128] It should also be understood that the various implementations provided in the embodiments of the present disclosure can be combined arbitrarily to achieve different technical effects.

[0129] The above is a specific description of the preferred implementation of the present disclosure, but the present disclosure is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions under the shared conditions without violating the spirit of the present disclosure. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present disclosure.

Claims

1. A human body detection method for elevator contactless elevator call, characterized in that: include: Acquire each frame of intermediate frequency signal of the human target, perform fast Fourier transform and CFAR detection on each frame of data, and obtain first point cloud data; Filter out valid points according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements, and obtain second point cloud data of the valid points; Filtering out valid target points based on the second point cloud data, obtaining the number of clustered targets according to the number of valid target points, and calculating motion parameters; Determining an elevator call trigger count according to the number of cluster targets and the motion parameter; If the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator is controlled to go to the target floor.

2. The method for detecting human body in a non-contact elevator call according to claim 1, characterized in that: The intermediate frequency signal has a plurality of reference channels, each frame of data includes a plurality of sub-data corresponding to the reference channels, and performing fast Fourier transform and CFAR detection on each frame of data to obtain first point cloud data includes: Performing one-dimensional fast Fourier transform and two-dimensional fast Fourier transform on the sub-data corresponding to each reference channel to obtain a two-dimensional matrix of distance and Doppler corresponding to each reference channel; Performing the CFAR detection after taking the modulus of the distance and Doppler two-dimensional matrix corresponding to the detection channel to obtain third point cloud data, wherein the detection channel is determined from the plurality of reference channels based on a preset condition; Based on the third point cloud data, sequentially performing a three-dimensional fast Fourier transform and a four-dimensional fast Fourier transform on the distance and Doppler two-dimensional matrix corresponding to each reference channel to obtain fourth point cloud data; Coordinate transformation is performed on the third point cloud data and the fourth point cloud data to obtain the first point cloud data.

3. The method for detecting human body in a non-contact elevator call according to claim 1, characterized in that: The first point cloud data includes a set of target points, a first distance lateral component of the target points, a first distance longitudinal component of the target points, a first distance height component of the target points, a first speed of the target points, and a first energy of the target points; The detection distance and energy threshold include the maximum horizontal detection distance, the minimum vertical detection distance, the maximum vertical detection distance, the minimum detection height, the maximum detection height, and the target point energy threshold; The step of selecting valid points according to whether the target points in the first point cloud data meet the detection distance and energy threshold requirements to obtain second point cloud data of the valid points includes: If the first distance lateral component of the target point in the target point set is between the negative value of the maximum lateral detection distance in a closed interval and the positive value of the maximum lateral detection distance, the first distance longitudinal component is between the minimum longitudinal detection distance in a closed interval and the maximum longitudinal detection distance, the first distance height component is between the minimum detection height and the maximum detection height in a closed interval, and the first energy is greater than or equal to the target point energy threshold, then the corresponding target point is set as a valid point, and the valid points are combined into a valid point set. Based on the first distance lateral component corresponding to the target point, the first distance longitudinal component corresponding to the target point, the first speed corresponding to the target point, and the first energy corresponding to the target point, the second distance lateral component of the valid point, the second distance longitudinal component of the valid point, the second speed of the valid point, and the second energy of the valid point are obtained, and the valid point set, the second distance lateral component, the second distance longitudinal component, the second speed, and the second energy are combined into the second point cloud data.

4. The method for detecting human body in a non-contact elevator call according to claim 3, characterized in that: The motion parameters include a lateral weighted average distance of the target effective point, a longitudinal weighted average distance of the target effective point, and a weighted average speed of the target effective point; The step of screening out target valid points based on the second point cloud data, obtaining the number of clustered targets according to the number of target valid points, and calculating motion parameters includes: Based on the second point cloud data, in a first iteration, the valid point with the maximum second energy and a clustering flag of 0 is selected from the valid point set as a centroid point, the valid point within a preset distance range from the centroid point is selected from the valid point set as an associated distance point, the centroid point and the associated distance point are used as the target valid points, and the clustering flags of the valid points corresponding to the centroid point and the associated distance point are set to 1; Count the number of target valid points of this iteration. If the number of points is greater than or equal to 2, add 1 to the number of clustered targets, and calculate the horizontal weighted average distance of the target valid points, the vertical weighted average distance of the target valid points, and the weighted average speed of the target valid points of this iteration; The loop is iterated until the clustering flags of the valid points in the valid point set are all 1, or the number of clustering targets reaches the maximum number of clustering targets in a single frame.

5. The method for detecting human body in a non-contact elevator call according to claim 4, characterized in that: The calculating of the lateral weighted average distance of the target effective point, the longitudinal weighted average distance of the target effective point, and the weighted average speed of the target effective point of this iteration includes: In this iteration, a third distance lateral component of the target effective point, a third distance longitudinal component of the target effective point, a third velocity of the target effective point, and a third energy of the target effective point are obtained based on the second distance lateral component, the second distance longitudinal component, the second velocity, and the second energy; summing the products of each of the third distance transverse components multiplied by each of the third energies to obtain a first intermediate value; summing the products of each of the third distance longitudinal components multiplied by each of the third energies to obtain a second intermediate value; summing the products of each of the third speeds and each of the third energies to obtain a third intermediate value; summing up the third energies to obtain a fourth intermediate value; The first intermediate value is divided by the fourth intermediate value to obtain the lateral weighted average distance, the second intermediate value is divided by the fourth intermediate value to obtain the longitudinal weighted average distance, and the third intermediate value is divided by the fourth intermediate value to obtain the weighted average speed.

6. The method for detecting human body in a non-contact elevator call according to claim 4, characterized in that: The step of confirming the elevator call trigger count according to the number of cluster targets and the motion parameter includes: If the number of cluster targets is greater than 0, and the horizontal weighted average distance is between the negative value of the horizontal trigger distance in the closed interval and the positive value of the horizontal trigger distance, the vertical weighted average distance is less than or equal to the vertical trigger distance, and the weighted average speed is less than or equal to the trigger speed upper limit, then the elevator call trigger count is increased by 1; If all the target valid points satisfy that the absolute value of the horizontal weighted average distance is greater than the horizontal protection distance, or if all the target valid points satisfy that the vertical weighted average distance is greater than the vertical protection distance, then the elevator call trigger count is cleared; If the number of clustered targets is equal to 0, the target loss count is increased by 1. If the target loss count is greater than the maximum allowed number of lost frames, the elevator call trigger count is cleared.

7. The method for detecting human body in a non-contact elevator call according to claim 1, characterized in that: If the elevator call trigger count is greater than or equal to the elevator call trigger threshold, controlling the elevator to go to the target floor includes: If the elevator call trigger count is greater than or equal to the elevator call trigger threshold, the elevator call command and the status query command are sent in a periodic cycle, and the indicator light is turned on. The elevator is controlled to go to the target floor according to the elevator call command, and the elevator status is obtained according to the status query command. If the elevator status is that the elevator has arrived at the target floor and is in the process of opening the door or the door opening is completed, the indicator light is turned off, the elevator call command is stopped from being reported, and the status query command is continued to be sent in a periodic cycle. The elevator status is obtained according to the status query command. If the elevator status is that the elevator has completed closing the door at the target floor, the status query command is stopped from being sent, and the next detection is re-entered after waiting for the preset protection time.

8. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the human body detection method for contactless elevator calling according to any one of claims 1 to 7 when executing the computer program.

9. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a human body detection method for contactless elevator calls according to any one of claims 1 to 7 is implemented.

Images