WO2021180060A1

WO2021180060A1 - Channel gate control method, apparatus and system

Info

Publication number: WO2021180060A1
Application number: PCT/CN2021/079701
Authority: WO
Inventors: 任志浩; 徐晖
Original assignee: 杭州海康威视数字技术股份有限公司
Priority date: 2020-03-11
Filing date: 2021-03-09
Publication date: 2021-09-16
Also published as: CN113393603B; CN113393603A

Abstract

A channel gate control method, apparatus and system. The method comprises: collecting current images in the gate channel entrance direction, the current images at least comprising a two-dimensional image and a depth image; performing pedestrian detection on a first target on the basis of the two-dimensional image; acquiring the position of the first target on the basis of the depth image; when the position of the first target is at a set distance threshold, triggering non-contact verification on the first target, wherein the distance threshold is determined according to a movement rate; and controlling opening of the gate channel on the basis of the verification result. Therefore, the waiting time of stopping in front of a channel gate, waiting for the identification result and then passing the gate is saved, the passing efficiency of the channel gate is improved, senseless passing effect is achieved, user experience is improved, and the control and determination accuracy of the gate channel is improved.

Description

Control method, device and system of channel gate

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on March 11, 2020, with an application number of 202010165868.2 and the invention title is "a method and system for controlling access gates", the entire content of which is incorporated herein by reference Applying.

Technical field

This application relates to the field of security, and in particular to a method, device and system for controlling access gates.

Background technique

In the current channel gate control method, the personnel can only be released after they have passed the identity verification. Regardless of the verification method, including but not limited to swiping card, fingerprint, face, iris, vein, etc., the person needs to stop in front of the gate and take the necessary actions during the travel process of the person. The face authorization verification is For example, during the passage of people, it is necessary for the passers to walk to the gates of the gates to stand, and they can only pass after the facial recognition authentication is qualified, which affects the passage efficiency of the gates.

Summary of the invention

The present application provides a method, device and system for controlling passage gates, so as to improve the passage efficiency of passage gates. The specific technical solutions are as follows:

In the first aspect, an embodiment of the present application provides a method for controlling an access gate, the method including:

Collect the current image in the entrance direction of the gate channel, where the current image includes at least a two-dimensional image and a depth image;

Perform pedestrian detection on the first target based on the two-dimensional image;

The position of the first target is acquired based on the depth image, and when the position of the first target is at a set distance threshold, non-contact verification of the first target is triggered, wherein the distance threshold is determined according to the movement rate;

Based on the verification result, the opening of the gate channel is controlled.

In the second aspect, an embodiment of the present application provides a control device for a passage gate, the device including:

The acquisition module is used to acquire the current image in the entrance direction of the gate channel, where the current image includes at least a two-dimensional image and a depth image;

The detection module is used to perform pedestrian detection on the first target based on the two-dimensional image;

The verification module is configured to obtain the position of the first target based on the depth image, and when the position of the first target is at a set distance threshold, trigger non-contact verification of the first target, wherein the distance threshold is determined according to the moving speed;

The control module is used to control the opening of the gate channel based on the verification result.

In a third aspect, an embodiment of the present application provides a gate system, including a verification module and an image data processing module, the system also includes a memory and a processor,

The memory stores instructions that can be executed by the processor, and the instructions are executed by the processor so that the processor controls the verification module and the image data processing module to execute the method provided in the first aspect of the embodiment of the present application.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium in which a computer program is stored. When the computer program is executed by a processor, the method provided in the first aspect of the embodiment of the present application is implemented.

In the fifth aspect, the embodiment of the present application provides a computer program product for executing at runtime: the method provided in the first aspect of the embodiment of the present application.

Applying the embodiment of the application, the non-contact verification of the first target is triggered by detecting whether the position of the first target is within the distance threshold determined according to the moving rate, so that the verification is performed on the first target relative to the passage from far to near The process is executed at the same time period, which avoids the waiting time in the prior art of stopping in front of the channel gate and waiting for the recognition result before passing, improves the passage efficiency of the channel gate, and achieves the effect of non-inductive passage. The user experience is improved, and the control and judgment accuracy of the gate channel is improved.

Description of the drawings

In order to explain the embodiments of the present application and the technical solutions of the prior art more clearly, the following briefly introduces the drawings that need to be used in the embodiments and the prior art. Obviously, the drawings in the following description are merely the present invention. For some of the embodiments of the application, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

FIG. 1 is a schematic diagram of the installation position of a depth camera device according to an embodiment of the application.

FIG. 2 is a schematic diagram of the installation position of a depth camera device according to another embodiment of the application.

FIG. 3 is a schematic flowchart of a single-channel gate control method equipped with a depth camera module according to an embodiment of the application.

FIG. 4 is a schematic diagram of the positional relationship between the gate and the distance threshold according to an embodiment of the application.

FIG. 5 is a schematic diagram of the space of a multi-channel gate equipped with a depth camera module according to an embodiment of the application.

Fig. 6 is a schematic flow chart of a method for controlling a multi-channel gate equipped with a depth camera module according to an embodiment of the application.

FIG. 7 is a schematic diagram of the space of a multi-channel gate equipped with a depth camera module according to another embodiment of the application.

FIG. 8 is a schematic flowchart of a method for controlling a multi-channel gate equipped with a depth camera module according to another embodiment of the application.

Fig. 9 is a schematic structural diagram of a control device for a passage gate according to an embodiment of the application.

Fig. 10 is a schematic structural diagram of a multi-channel gate system according to an embodiment of the application.

Detailed ways

In order to make the purpose, technical solutions, and advantages of the present application clearer, the following further describes the present application in detail with reference to the accompanying drawings and embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

After the embodiment of the application detects that the first target is a pedestrian through the acquired two-dimensional image information, it determines whether the first target is within a set threshold from the gate entrance by acquiring the three-dimensional spatial information of the first target, and if so, triggers The non-contact verification of the first target, further, through the acquired two-dimensional image information, to identify the second target accompanying the first target, when the second target is identified as large luggage and/or children, Based on the acquired three-dimensional space information, the three-dimensional space size of the first target and the second target are determined, and the passage time of the gate passage is controlled according to the three-dimensional space size.

For multiple channels, by acquiring the three-dimensional space information of the first target within a threshold set from the gate entrance, the velocity and moving direction of the first target are calculated based on the three-dimensional space information of the first target, and according to the velocity and movement of the first target Direction, predict the gate channel to be entered by the first target, and trigger the gate channel to verify the first target.

In order to collect two-dimensional images and three-dimensional spatial information of the target, a depth camera module can be used. One of the implementation manners uses binocular stereo vision equipment, that is, at least two VIS (Visual, visible light) cameras; the second implementation manner, uses structured light machine vision equipment, that is, at least one projector, one NIR (Near Infrared) , Near-infrared) camera and a VIS camera; the third implementation mode uses a ToF (Time of Flight) machine vision device solution, that is, there is at least one ToF camera and one VIS camera.

In addition, considering factors such as cost and system robustness can also increase the number of cameras. For example, in the first embodiment, a NIR camera can also be added to meet the needs of biometric identification under extremely low illumination conditions; in the second embodiment, a NIR camera can be added to increase the measurement accuracy and accuracy. Correspondingly, you can also consider adding a projector. quantity. It can be designed flexibly according to actual scenarios and user needs.

The two-dimensional images collected by the VIS camera in the depth camera module can be directly used for pedestrian detection, biometric recognition and other functions, or it can be completed by separately configuring other imaging modules.

Referring to FIG. 1, FIG. 1 is a schematic diagram of an installation position of a depth camera device according to an embodiment of the application. In this example, a depth camera is installed at the end of the gate body extending along the entrance of the passage, and the monitored field of view covers the passage.

Refer to FIG. 2, which is another schematic diagram of the installation of the depth camera device according to the embodiment of the application. In this example, the depth camera is separated from the gate body, for example, is installed in the upper area of the gate. This installation scheme is more suitable for indoors or occasions where the depth camera and other equipment are conveniently fixed above the gate, and can have a better field of view. The equipment on the left side of the gate near the entrance in Figure 2 is used to interact with pedestrians and display information such as facial images.

Example one

Referring to FIG. 3, FIG. 3 is a schematic diagram of a flow chart of a single-channel gate control installed with a depth camera module according to an embodiment of the application.

Step 301: Control the depth camera device to collect the current image within its field of view to obtain two-dimensional image and three-dimensional spatial information, that is, a depth map.

Step 302: Perform first target detection based on the two-dimensional image to determine whether there is a pedestrian, that is, perform pedestrian detection on the first target, if yes, perform step 303, otherwise, return to step 302.

In this step, it is possible to first detect whether there is a pedestrian based on the gait, and then further perform the first target detection on the detected pedestrian. In this way, the search range of the target can be narrowed by the gait recognition first.

Step 303: Acquire the current position of the first target based on the depth map, and determine whether the current position is within a preset distance threshold, and if so, trigger a non-contact verification process based on the two-dimensional image, for example, triggering a non-contact verification process based on the two-dimensional image. The verification process of face recognition, otherwise, keep the current state.

In this step, the position of the depth camera can be used as the origin of the world coordinate system, and the three-dimensional space information of the first target in the world coordinate system can be calculated based on the depth map.

In order to make the passers-by pass through the gate without waiting for verification, this embodiment wakes up the current channel gate when the first target reaches the distance threshold to trigger the verification process, that is, completes the opening of the gate when the first target is close to the gate passage entrance action. Specifically, the background main control chip can wake up the current channel gate, and the main control chip triggers the verification module to execute the verification process. The verification module can be a module installed inside the gate or a background verification server.

Refer to FIG. 4, which is a schematic diagram of the positional relationship between the gate and the distance threshold according to the embodiment of the application. Among them, the distance threshold is set according to the algorithm time, pedestrian moving average speed, and motor control time. The algorithm time includes one of pedestrian detection time, verification time, or any combination thereof. Specifically, the distance threshold is the pedestrian moving average speed The result of multiplying with the time-consuming algorithm and the sum of the time-consuming motor control can be expressed mathematically as:

Distance threshold = average pedestrian moving speed* (algorithm time-consuming + motor system control time-consuming)

Among them, the pedestrian moving average speed is obtained based on statistical data.

The applicable distance of pedestrian detection is generally longer than the applicable distance of biometric identification. Therefore, first perform pedestrian detection. If a pedestrian is detected, then perform biometric identification when the pedestrian travels to the gate of the gate and enters a certain area. Authentication, no need for passers-by to stop at the gate entrance and then identify.

Step 304: Extract a second target accompanying the first target based on the current image, and detect and recognize the second target according to the two-dimensional image. If the second target is detected as an object and/or child with a certain spatial size, the The first target and the second target are taken as a whole, and the three-dimensional space size information of the first target and the second target as a whole is obtained based on the depth map, and then step 305 is executed. Otherwise, it is determined that there is a trailing person and an alarm is triggered.

When detecting and recognizing the second target, you can use the pre-trained detection and recognition model based on the deep learning method to identify the category and size of the second target. If the space size of the second target is recognized to meet the set space size , And the second target is an object and/or a pedestrian, it can be confirmed that the detected second target is an object and/or child with a certain size of space.

In step 305, when the verification result of the verification process is passable, the channel is triggered to open, and according to the three-dimensional space size information determined in step 304, the length of time that the channel remains open, that is, the length of pass time is determined.

In this step, the length of the travel time can be determined according to the length information (depth information) in the three-dimensional space size and the pedestrian moving average speed. For example, the travel time length is the quotient of the length information and the pedestrian moving average speed, expressed in a mathematical formula for:

Travel time length = length information in the three-dimensional space size / pedestrian moving average speed

In order to monitor the current passage, the depth camera can monitor in real time during the passage of passengers. If someone is passing through the gate channel, the gate channel is controlled to remain open at all times to further prevent the passage of people from holding the gate.

Step 306: When the passage time is reached, the gate channel is closed.

In the process of passing personnel from far to near, the embodiment of the application narrows the search range through pedestrian detection, and then uses biometric identification to confirm the authority, which makes full use of the travel time of the passer, avoids the waiting time of the passer, and improves The passage efficiency is realized, and the non-inductive passage is realized; the passage time length determined according to the three-dimensional space size of the passer effectively reduces the occurrence of the passage of the passer The user body.

Example two

Referring to Figure 5, Figure 5 is a schematic diagram of a multi-channel gate equipped with a depth camera module according to an embodiment of the application. A depth camera device is installed in each channel, and the field of view of each depth camera device is same. The installation position of the depth camera device and the angle at which the image is collected can be the same.

Refer to FIG. 6, which is a schematic diagram of a process flow of controlling a multi-channel gate with a depth camera module installed in the second embodiment of the application.

Step 601: Control each depth camera device to collect the current image within its field of view, so as to obtain a two-dimensional image and three-dimensional spatial information, that is, a depth map.

For each current image, perform the following steps:

Step 602: Perform first target detection based on the two-dimensional image, and determine whether there is a passer-by, if yes, perform step 603, otherwise, return to step 602.

Step 603: Obtain the current position of the first target based on the depth map, and determine whether the current position is within a preset distance threshold, and if so, calculate the current movement rate and direction of the first target according to at least two consecutive depth images; otherwise, , The current gate status is maintained, and step 603 is returned.

In this step, since the channel will be matched to the first target later, the distance threshold also needs to consider the time-consuming channel matching. Specifically, the distance threshold is the sum of the algorithm time-consuming, motor control time-consuming, and channel-matching time-consuming The result of the product of the pedestrian moving average speed is mathematically expressed as:

Distance threshold = current moving average speed* (algorithm time-consuming + motor system control time-consuming + channel matching time-consuming)

In addition, the current moving rate and moving direction are calculated by the change of the depth information of the first target in two consecutive depth images, which is expressed mathematically as:

v=(d2-d1)/t

Among them, d2 and d1 are the depth values of the first target in two consecutive depth images, t is the time difference between the two consecutive depth images, and v is the current moving rate; the moving direction is determined according to the changes in the depth values of the two consecutive frames .

In order to improve the accuracy of the current moving speed, multiple speed values can be calculated, and then the average value of the current moving speed can be obtained.

Step 604: Assign a matching channel according to the current moving direction of the first target, and trigger the verification process of the matching channel.

Step 605: Extract a second target accompanying the first target based on the current image, and detect and recognize the second target according to the two-dimensional image. If the second target is detected as an object and/or child with a certain space size, the The first target and the second target are taken as a whole, and the three-dimensional space size information of the first target and the second target as a whole is obtained based on the depth map, and then step 606 is executed. Otherwise, it is determined that there is a trailing person and an alarm is triggered.

Step 606: When the verification result of the verification process is passable, trigger the channel to open, and determine the length of time that the channel remains open according to the three-dimensional space size information determined in step 605 and the current moving average speed, that is, the length of travel time.

In this step, the travel time length can be determined according to the length information in the three-dimensional space size and the current moving average speed. For example, the travel time length is the quotient of the length information and the pedestrian moving average speed, which is expressed mathematically as:

Travel time length = length information / current moving average speed

Step 607: When the passage time is reached, the gate channel is closed.

In this embodiment, the moving direction is predicted by the depth image, and the matching channel is allocated according to the moving direction, which further improves the user experience of non-inductive traffic; the current moving average speed calculated in real time is used to determine the length of the traffic time, so that each traffic The passing time of the personnel can be different, which improves the humanization of the turnstile and more effectively reduces the probability of the passage holding the passing personnel.

Example three

Refer to FIG. 7, which is a schematic diagram of a multi-channel gate equipped with a depth camera module according to an embodiment of the present application, wherein at least one of the multi-channels has a width larger than other channels, so as to facilitate the carrying of large luggage Of passers-by.

Referring to FIG. 8, FIG. 8 is a schematic diagram of a process of controlling a multi-channel gate with a depth camera module installed in the third embodiment.

Step 801: Control each depth camera device to collect the current image within its field of view, so as to obtain a two-dimensional image and three-dimensional spatial information, that is, a depth map.

For each current image, perform the following steps:

Step 802: Perform a first target detection based on the two-dimensional image to determine whether there is a passer-by. If yes, perform step 803; otherwise, return to step 802.

Step 803: Obtain the current position of the first target based on the depth map, and determine whether the current position is within a preset distance threshold. If so, perform step 804; otherwise, maintain the current gate state and return to step 803.

In this step, the determination of the distance threshold is the same as in step 603.

Step 804: Extract the second target accompanying the first target based on the current image, and detect and recognize the second target according to the two-dimensional image. If the second target is detected as an object and/or child with a certain space size, the The first target and the second target as a whole, obtain the three-dimensional space size information of the first target and the second target based on the depth map, allocate the matching channel according to the three-dimensional space size, and trigger the matching channel verification process; otherwise, it is determined that there is a trailing person To trigger an alarm.

In this step, matching channels can be allocated according to the maximum width of the three-dimensional space size.

Step 805: Calculate the current movement rate of the first target according to at least two consecutive depth image frames.

In this step, the current movement rate is calculated by changing the depth information of the first target in two consecutive depth image frames, which is expressed mathematically as:

v=(d2-d1)/t

Among them, d2 and d1 are the depth values of the first target in two consecutive frames, t is the time difference between the two consecutive frames, and v is the current moving rate.

In step 806, when the verification result of the verification process is passable, the channel is triggered to open, and according to the determined three-dimensional space size information and the current moving average rate, the length of time that the channel remains open is determined, that is, the length of travel time.

In this step, the travel time length can be determined according to the length information (depth information) in the three-dimensional space size and the current moving average speed. For example, the travel time length is the quotient of the length information and the pedestrian moving average speed, which is expressed mathematically as :

Travel time length = length information in the three-dimensional space size/current moving average speed

In step 807, when the passage time is reached, the gate channel is closed.

In this embodiment, according to the overall space size of the first target and the second target, matching passages are allocated to the passers-by and their carry-on luggage, which avoids human error selection and realizes the automation and intelligence of passage; according to real-time calculations The current moving average speed and the length of the space size determine the length of the passage time, so that the passage time of each passer can be different, which improves the humanization of the gate and the intelligence of the control, and more effectively reduces the passage of the passage. Probability of personnel.

In the above embodiment, the images collected by the depth camera can also be used for video surveillance in the gate area and integrated into the overall intelligent system, for example, into the building monitoring system, the toll system of the subway gate, and so on.

As shown in FIG. 9, FIG. 9 is a schematic structural diagram of a control device for a passage gate to which an embodiment of the present application is applied, and the control device may include:

The acquisition module 910 is configured to acquire a current image in the entrance direction of the gate channel, where the current image includes at least a two-dimensional image and a depth image;

The detection module 920 is configured to perform pedestrian detection on the first target based on the two-dimensional image;

The verification module 930 is configured to obtain the position of the first target based on the depth image, and when the position of the first target is within a set distance threshold, trigger non-contact verification of the first target, wherein the distance threshold is determined according to the moving speed;

The control module 940 is configured to control the opening of the gate channel based on the verification result.

Optionally, the detection module 920 may also be used to extract the second target accompanying the first target based on the two-dimensional image at the same time when the first target is non-contacted verification, and to detect and recognize the second target; If it is detected that the second target is an object and/or a child with a set space size, acquiring three-dimensional space size information of the first target and the second target based on the depth image;

The control module 940 may be specifically used to trigger the opening of the gate channel when the verification is passed, and control the length of time that the gate channel remains open according to the three-dimensional space size information.

Optionally, the gate channel is a single channel;

The detection module 920 can also be used to determine that the second target is a trailing person and trigger an alarm if it is detected that the second target is not an object and/or child with a set space size;

When the detection module 920 is used to obtain the three-dimensional space size information of the first target and the second target based on the depth image, it can be specifically used to take the first target and the second target as a whole, and determine the three-dimensional size of the whole according to the image data of the depth image. Space size information;

The control module 940 can be specifically used to divide the length information in the three-dimensional space size information by the moving average speed as the length of time that the gate channel remains open, where the pedestrian moving average speed is the statistical value of the pedestrian movement sample; where , The distance threshold is set according to the time consumption of all algorithms, the average speed of pedestrians, and the time consumption of motor control. The algorithm time consumption includes one of the time consumption of pedestrian detection and verification or any combination thereof.

Optionally, the gate channel includes more than two single channels;

The device may also include:

The detection module 920 can also be used to extract the second target accompanying the first target based on the two-dimensional image when the position of the first target is at the set distance threshold, and detect and recognize the second target; if the second target is detected If the target is an item and/or child with a set space size, the current movement rate and direction of the first target are calculated according to at least two consecutive depth images, the matching channel is allocated according to the moving direction, and the matching channel’s non-compliance with the first target is triggered. Contact verification; at the same time as the non-contact verification of the first target, the second target accompanying the first target is extracted based on the two-dimensional image, and the second target is detected and identified. If the second target is detected as having For items and/or children whose spatial size is set, the three-dimensional spatial size information of the first target and the second target is acquired based on the depth image;

The control module 940 may be specifically used to trigger the opening of the gate channel when the verification is passed, and control the length of time that the gate channel remains open according to the three-dimensional space size information and the current movement rate of the first target.

Optionally, the gate channel includes two or more single channels, and the width of each channel is different;

The detection module 920 can also be used to extract the second target accompanying the first target based on the two-dimensional image when the position of the first target is at the set distance threshold, and detect and recognize the second target; if the second target is detected If the target is an item and/or child with a set spatial size, the three-dimensional spatial size information of the first target and the second target is acquired based on the depth image, the matching channel is allocated according to the three-dimensional spatial size information, and the matching channel is triggered for the first target. Contact verification: Calculate the current movement rate of the first target based on at least two consecutive depth images at the same time during the non-contact verification of the first target;

Optionally, the device may also include:

The detection module 920 can also be used to determine that the second target is a trailing person and trigger an alarm if it is detected that the second target is not an object and/or a child with a set space size;

The detection module 920, when used to obtain the three-dimensional space size information of the first target and the second target based on the depth image, can be specifically used to take the first target and the second target as a whole, and determine the overall size according to the image data of the depth image. Three-dimensional space size information;

The control module 940 can be specifically used to divide the length information in the three-dimensional space size information with the current movement rate of the first target as the length of time the gate channel remains open; wherein the distance threshold is time-consuming and pedestrian based on all algorithms The moving average speed, motor control time, and allocation matching channel time-consuming settings are set, where the algorithm time-consuming includes one of pedestrian detection time, verification time-consuming one or any combination thereof, and the pedestrian moving average speed is the statistical value of the pedestrian movement sample.

Optionally, the detection module 920 can be specifically used to calculate the depth information change of the first target in two consecutive frames of depth images, and use the ratio of the depth information change to the time difference between the two consecutive frames of depth images as the current movement rate; The average value of two or more current moving speeds to obtain the current moving average speed of the first target;

The control module 940 can be specifically used to divide the length information in the three-dimensional space size information by the current moving average speed of the first target as the length of time the gate channel remains open; where the distance threshold is: algorithm time-consuming, The product of the time-consuming motor system control, the time-consuming allocation and matching channel and the average pedestrian moving speed.

Refer to FIG. 10, which is a schematic diagram of a multi-channel gate system to which the control method of the present application is applied. Each channel in the gate system includes a depth camera module for acquiring two-dimensional images and depth information, a verification module for verifying each channel, and image processing for processing image data Module, the system also includes memory and processor, for example, the main control chip MCU, the modules of each channel communicate and control through the main control chip MCU:

The main control chip controls the image acquisition of the depth camera module,

Control the image processing module to perform pedestrian detection on the first target based on the two-dimensional image, and obtain the position of the first target based on the depth image. When the position of the first target is within the set distance threshold, control the verification module to trigger the detection of the first target The non-contact verification, based on the verification result, controls the opening of the gate channel.

Through the above control, the gate channel control method in the embodiment of the present application is realized.

The modules of each channel can be integrated according to specific design requirements, for example, the depth camera module and the verification module are integrated, or the image processing module and the depth camera module are integrated; the integration of the modules The separation does not affect the realization of its function.

The aforementioned processor may be a general-purpose processor, including CPU (Central Processing Unit), NP (Network Processor), etc.; it may also be DSP (Digital Signal Processor), ASIC ( Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array, Field Programmable Gate Array) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

The memory may include RAM (Random Access Memory, random access memory), and may also include NVM (Non-Volatile Memory, volatile memory), such as at least one disk storage. Optionally, the memory may also be at least one storage device located far away from the foregoing processor.

The gate system of the present application controls gate channels through images, and does not require a complicated infrared beaming system, which is beneficial to cost reduction, facilitates the integration and systemization of modules, and facilitates the dispatch and control of the gate system.

The embodiment of the present application also provides a computer-readable storage medium, and a computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the following steps are implemented:

The embodiment of the present application also provides a computer program product, which is used to execute any of the above-mentioned channel gate control methods at runtime.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, it can be implemented in the form of a computer program product in whole or in part. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present invention are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website, computer, server, or data center. Transmission to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, DSL (Digital Subscriber Line)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center integrated with one or more available media. The usable medium may be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD (Digital Versatile Disc)), or a semiconductor medium (such as an SSD (Solid State Disk)), etc. .

For the device/multi-channel gate system/computer-readable storage medium/computer program product embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment description. .

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply one of these entities or operations. There is any such actual relationship or order between. Moreover, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes those that are not explicitly listed Other elements of, or also include elements inherent to this process, method, article or equipment. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article, or equipment that includes the element.

The various embodiments in this specification are described in a related manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, multi-channel gate system, computer readable storage medium, and computer program product embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For related parts, please refer to the description of the method embodiments. That's it.

The above descriptions are only preferred embodiments of this application, and are not intended to limit this application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in the protection of this application. Within the range.

Claims

A method for controlling channel gates, characterized in that the method includes:

Acquiring a current image in the entrance direction of the gate passage, where the current image includes at least a two-dimensional image and a depth image;

Performing pedestrian detection on the first target based on the two-dimensional image;

The position of the first target is acquired based on the depth image, and when the position of the first target is within a set distance threshold, a non-contact verification of the first target is triggered, and the distance threshold is based on the moving speed Sure;

Based on the verification result, the opening of the gate channel is controlled.
The control method of claim 1, wherein after the step of triggering non-contact verification of the first target when the position of the first target is at a set distance threshold, the The method also includes:

At the same time when non-contact verification is performed on the first target, extracting a second target accompanying the first target based on the two-dimensional image, and detecting and identifying the second target;

If it is detected that the second target is an object and/or a child with a set space size, acquiring three-dimensional space size information of the first target and the second target based on the depth image;

The step of controlling the opening of the gate channel based on the verification result includes:

When the verification is passed, the gate channel is triggered to open, and the length of time that the gate channel remains open is controlled according to the three-dimensional space size information.
The control method according to claim 2, wherein the gate channel is a single channel;

After the step of detecting and identifying the second target, the method further includes:

If it is detected that the second target is not an object and/or child with a set space size, it is determined that the second target is a trailing person, and an alarm is triggered;

The step of acquiring three-dimensional space size information of the first target and the second target based on the depth image includes:

Taking the first target and the second target as a whole, and determining the three-dimensional space size information of the whole according to the image data of the depth image;

The step of controlling the length of time that the gate channel remains open according to the three-dimensional space size information includes:

The result of dividing the length information in the three-dimensional space size information by the moving average speed is used as the length of time the gate channel remains open, where the pedestrian moving average speed is a statistical value of pedestrian movement samples; the distance threshold is based on All algorithm time-consuming, pedestrian moving average speed, motor control time-consuming setting, algorithm time-consuming includes one of pedestrian detection time, verification time-consuming one or any combination thereof.
The control method according to claim 1, wherein the gate channel includes two or more single channels;

After the step of acquiring the position of the first target based on the depth image, the method further includes:

When the position of the first target is within the set distance threshold, extracting a second target accompanying the first target based on the two-dimensional image, and detecting and identifying the second target;

If it is detected that the second target is an article and/or child with a set space size, the current movement rate and direction of the first target are calculated according to at least two consecutive depth images, and the matching is allocated according to the movement direction Channel to trigger the non-contact verification of the first target by the matching channel;

At the same time as the non-contact verification of the first target, the second target accompanying the first target is extracted based on the two-dimensional image, and the second target is detected and recognized. If the second target is an item and/or a child with a set space size, then acquiring three-dimensional space size information of the first target and the second target based on the depth image;

The step of controlling the opening of the gate channel based on the verification result includes:

When the verification is passed, the gate channel is triggered to open, and the length of time the gate channel remains open is controlled according to the three-dimensional space size information and the current movement rate of the first target.
The control method according to claim 1, wherein the gate channel includes two or more single channels, and the width of each channel is different;

After the step of acquiring the position of the first target based on the depth image, the method further includes:

When the position of the first target is within the set distance threshold, extracting a second target accompanying the first target based on the two-dimensional image, and detecting and identifying the second target;

If it is detected that the second target is an article and/or a child with a set spatial size, the three-dimensional size information of the first target and the second target is acquired based on the depth image, and the three-dimensional space size information is obtained according to the three-dimensional space. The size information allocates a matching channel, and triggers the non-contact verification of the first target by the matching channel;

Calculating the current movement rate of the first target according to at least two consecutive depth images at the same time when the first target is subjected to non-contact verification;

The step of controlling the opening of the gate channel based on the verification result includes:

When the verification is passed, the gate channel is triggered to open, and the length of time the gate channel remains open is controlled according to the three-dimensional space size information and the current movement rate of the first target.
The control method according to claim 4 or 5, wherein after the step of detecting and recognizing the second target, the method further comprises:

If it is detected that the second target is not an object and/or child with a set space size, it is determined that the second target is a trailing person, and an alarm is triggered;

The step of acquiring three-dimensional space size information of the first target and the second target based on the depth image includes:

Taking the first target and the second target as a whole, and determining the three-dimensional space size information of the whole according to the image data of the depth image;

The step of controlling the length of time that the gate channel remains open according to the three-dimensional space size information and the current movement rate of the first target includes:

The result of dividing the length information in the three-dimensional space size information by the current movement rate of the first target is used as the length of time that the gate channel remains open; the distance threshold is based on the time consumption of all algorithms, the average speed of pedestrian movement, Motor control time-consuming and allocation matching channel time-consuming settings, where the algorithm time-consuming includes one of pedestrian detection time, verification time-consuming one or any combination thereof, and the pedestrian moving average velocity is a statistical value of pedestrian movement samples.
7. The control method according to claim 6, wherein the step of calculating the current moving rate and moving direction of the first target according to at least two consecutive depth images comprises:

Calculate the depth information change of the first target in two consecutive frames of depth images, and use the ratio of the depth information change to the time difference between the two consecutive frames of depth images as the current movement rate;

Obtaining an average value of two or more current moving rates to obtain the current moving average rate of the first target;

The step of controlling the length of time that the gate channel remains open according to the three-dimensional space size information and the current movement rate of the first target includes:

The result of dividing the length information in the three-dimensional space size information by the current moving average speed of the first target is used as the length of time the gate channel remains open; the distance threshold is: algorithm time consumption, motor system control consumption The product of the sum of the time and the time-consuming distribution matching channel and the pedestrian moving average velocity.
A control device for a passage gate, characterized in that the device comprises:

An acquisition module for acquiring a current image in the entrance direction of the gate passage, the current image including at least a two-dimensional image and a depth image;

The detection module is configured to perform pedestrian detection on the first target based on the two-dimensional image;

The verification module is configured to acquire the position of the first target based on the depth image, and trigger non-contact verification of the first target when the position of the first target is within a set distance threshold. The distance threshold is determined according to the moving speed;

The control module is used to control the opening of the gate channel based on the verification result.
The control device according to claim 8, wherein the detection module is further configured to extract based on the two-dimensional image at the same time when the verification module performs non-contact verification on the first target. The second target accompanying the first target is detected and identified; if the second target is detected as an object and/or child with a set space size, it is acquired based on the depth image Three-dimensional space size information of the first target and the second target;

The control module is specifically configured to trigger the opening of the gate channel when the verification is passed, and control the length of time that the gate channel remains open according to the three-dimensional space size information.
9. The control device of claim 9, wherein the gate channel is a single channel;

The detection module is further configured to determine that the second target is a trailing person and trigger an alarm if it is detected that the second target is not an object and/or a child with a set space size;

When the detection module is used for acquiring the three-dimensional space size information of the first target and the second target based on the depth image, it is specifically used for taking the first target and the second target as a whole , Determining the overall three-dimensional space size information according to the image data of the depth image;

The control module is specifically configured to divide the length information in the three-dimensional space size information by the moving average speed as the length of time for which the gate channel remains open, wherein the pedestrian moving average speed is the result of the pedestrian movement sample Statistical value; the distance threshold is set according to all algorithm time-consuming, pedestrian moving average speed, motor control time-consuming, and the algorithm time-consuming includes one of the time-consuming pedestrian detection and the time-consuming verification or any combination thereof.
The control device according to claim 8, wherein the gate channel includes more than two single channels;

The detection module is further configured to extract a second target accompanying the first target based on the two-dimensional image when the position of the first target is within a set distance threshold, and perform Detection and recognition; if it is detected that the second target is an item and/or a child with a set space size, the current movement rate and direction of the first target are calculated according to at least two consecutive depth images, and according to the movement The matching channel is assigned to the direction to trigger the non-contact verification of the first target by the matching channel; at the same time as the non-contact verification of the first target, based on the two-dimensional image extraction and the first target A second target accompanied by a target is detected and identified. If it is detected that the second target is an object and/or a child with a set space size, the first target is acquired based on the depth image. Three-dimensional space size information of the target and the second target;

The control module is specifically configured to trigger the opening of the gate channel when the verification is passed, and control the length of time that the gate channel remains open according to the three-dimensional space size information and the current movement rate of the first target.
The control device according to claim 8, wherein the gate channel includes more than two single channels, and the width of each channel is different;

The detection module is further configured to extract a second target accompanying the first target based on the two-dimensional image when the position of the first target is within a set distance threshold, and perform Detection and recognition; if it is detected that the second target is an article and/or child with a set spatial size, the three-dimensional spatial size information of the first target and the second target is acquired based on the depth image, and the three-dimensional space size information of the first target and the second target is acquired according to the The three-dimensional space size information allocates a matching channel to trigger the non-contact verification of the first target by the matching channel; at the same time as the non-contact verification of the first target, according to at least two consecutive depth images Calculating the current moving speed of the first target;

The control module is specifically configured to trigger the opening of the gate channel when the verification is passed, and control the length of time that the gate channel remains open according to the three-dimensional space size information and the current movement rate of the first target.
The control device according to claim 11 or 12, wherein the device further comprises:

The detection module is further configured to determine that the second target is a trailing person and trigger an alarm if it is detected that the second target is not an object and/or a child with a set space size;

When the detection model is used in the acquisition of the three-dimensional space size information of the first target and the second target based on the depth image, it is specifically used to integrate the first target and the second target as a whole , Determining the overall three-dimensional space size information according to the image data of the depth image;

The control module is specifically configured to divide the length information in the three-dimensional space size information by the current movement rate of the first target as the length of time that the gate channel remains open; the distance threshold is based on all algorithms Time-consuming, pedestrian moving average speed, motor control time-consuming, and allocation matching channel time-consuming setting, where algorithm time-consuming includes one of pedestrian detection time, verification time-consuming one or any combination thereof, said pedestrian moving average speed is pedestrian The statistical value of the moving sample.
The control device according to claim 13, wherein the detection module is specifically configured to calculate the depth information change of the first target in two consecutive frames of depth images, and compare the depth information change with the two consecutive frames of depth information. The ratio of the time difference between the images is used as the current moving speed; calculating the average of two or more current moving speeds to obtain the current moving average speed of the first target;

The control module is specifically configured to divide the length information in the three-dimensional space size information by the current moving average speed of the first target as the length of time the gate channel remains open; the distance threshold is: The product of the time-consuming algorithm, the time-consuming motor system control, the time-consuming allocation of matching channels and the average velocity of pedestrians.
A gate system includes a verification module and an image data processing module, characterized in that the system also includes a memory and a processor,

The memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor controls the verification module and the image data processing module to execute the control according to any one of claims 1 to 7 method.
The gate system of claim 15, wherein the gate body in the gate system is equipped with a depth camera device for collecting two-dimensional images and depth images, and the depth camera device is located along the gate. The end of the gate body extending from the entrance of the passage; and/or a depth camera device for collecting two-dimensional images and depth images is installed in the upper area of the gate passage in the gate system.
A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the control method according to any one of claims 1 to 7 is realized.
A computer program product, which is characterized in that it is used to execute the control method according to any one of claims 1 to 7.