Channel detection method, apparatus and system

Abstract

The present disclosure relates to the technical field of channel detection. Provided are a channel detection method, apparatus and system. The channel detection method comprises: receiving point cloud data from a scanning LiDAR arranged on at least one side of a channel; on the basis of the point cloud data, determining the contour of an object to be detected that enters the channel and the distance from said object to the at least one side of the channel; on the basis of the contour of said object and the distance from said object to the at least one side of the channel, determining the category of said object; acquiring a restricted-object category corresponding to the type of the channel; and when the category of said object comprises the restricted-object category corresponding to the type of the channel, outputting alarm prompt information.

Description

Channel detection method, device and system

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is based on the application with CN application number 202311656859.3 and application date December 5, 2023, and claims its priority. The disclosed content of the CN application is hereby introduced as a whole into this application. Technical Field

[0003] The present disclosure relates to the technical field of channel detection, and in particular to a channel detection method, device, and system. Background Art

[0004] Currently, most access detection systems use light curtains, photoelectric sensors, or input / output (I / O) lidars to detect human entry. For example, several light curtains are installed at intervals within a passageway. When a person enters or moves within the passageway, the light curtains are triggered, triggering an alarm. For example, multiple I / O lidars can be installed at the entrances and exits of designated areas within the passageway. When an I / O lidar detects a person entering the designated area, the system issues an alarm. Summary of the Invention

[0005] The present disclosure provides a channel detection method, device, and system.

[0006] According to a first aspect of the present disclosure, a channel detection method is proposed, comprising: receiving point cloud data from a scanning laser radar arranged on at least one side of a channel; determining, based on the point cloud data, the outline of an object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel; determining the category of the object to be detected based on the outline of the object to be detected and the distance from the object to be detected to at least one side of the channel; obtaining a restricted object category corresponding to the type of the channel; and outputting an alarm prompt message when the category of the object to be detected includes the restricted object category corresponding to the type of the channel.

[0007] In some embodiments, determining the category of the object to be detected based on the outline of the object to be detected and the distance from the object to be detected to at least one side of the channel includes: determining the category of the object to be detected based on the outline of the object to be detected, the distance from the object to be detected to at least one side of the channel, and the length of time that the object to be detected stays in a designated area in the channel.

[0008] In some embodiments, the outline of the object to be detected includes the width of the object to be detected, and determining the category of the object to be detected based on the outline of the object to be detected, the distance of the object to be detected to at least one side of the channel, and the residence time of the object to be detected in a designated area in the channel includes: determining a first candidate category of the object to be detected based on a comparison result of the width of the object to be detected with a range of width values ​​of objects of known categories; determining a second candidate category of the object to be detected based on a comparison result of the distance of the object to be detected to at least one side of the channel with a range of distance values ​​of objects of known categories to at least one side of the channel; determining a third candidate category of the object to be detected based on a comparison result of the residence time of the object to be detected in the designated area of ​​the channel with a range of residence time of objects of known categories in the designated area of ​​the channel; and determining the category of the object to be detected based on the first candidate category, the second candidate category, and the third candidate category of the object to be detected.

[0009] In some embodiments, determining the category of the object to be detected based on the first candidate category, the second candidate category, and the third candidate category of the object to be detected includes: when at least two candidate categories among the first candidate category, the second candidate category, and the third candidate category are the same, taking the same candidate category as the category of the object to be detected.

[0010] In some embodiments, the method further includes: outputting a detection failure prompt message when the first candidate category, the second candidate category, and the third candidate category are all different.

[0011] In some embodiments, the type of the channel is a vehicle channel or a pedestrian channel, the restricted object categories corresponding to the vehicle channel include people, and the restricted object categories corresponding to the pedestrian channel include vehicles.

[0012] In some embodiments, the channel detection method is applied to a vehicle security inspection scenario based on an X-ray scanning device, and the channel detection method further includes: when the type of the channel is a vehicle channel and the category of the object to be detected includes people, controlling the X-ray scanning device to be turned off.

[0013] In some embodiments, the channel detection method further includes: before receiving point cloud data from a scanning laser radar arranged on at least one side of the channel, in response to satisfying the start-up condition of the scanning laser radar, controlling the scanning laser radar to turn on.

[0014] In some embodiments, the activation conditions of the scanning laser radar include at least one of the following: the current time period is a set channel detection time period; and the presence of the object to be detected is determined based on an image captured by a camera corresponding to the channel.

[0015] In some embodiments, the scanning laser radar arranged on at least one side of the channel includes a first laser radar arranged on one side of the channel and a second laser radar arranged on the other side of the channel, and the first laser radar and the second laser radar are arranged in an staggered manner.

[0016] In some embodiments, determining the outline of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel based on the point cloud data includes: removing background point cloud data from the point cloud data to obtain a point cloud point set of the object to be detected; determining the outline of the object to be detected and the distance from the object to be detected to the scanning laser radar based on the point cloud point set of the object to be detected; determining the distance from the object to be detected to at least one side of the channel based on the distance from the object to be detected to the scanning laser radar and the scanning angle of the scanning laser radar.

[0017] According to a second aspect of the present disclosure, a channel detection device is proposed, comprising: a receiving module configured to receive point cloud data from a scanning laser radar arranged on at least one side of the channel; a first determination module configured to determine, based on the point cloud data, the outline of an object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel; a second determination module configured to determine the category of the object to be detected based on the outline of the object to be detected and the distance from the object to be detected to at least one side of the channel; an acquisition module configured to acquire a restricted object category corresponding to the type of the channel; and an output module configured to output an alarm prompt message when the category of the object to be detected includes a restricted object category corresponding to the type of the channel.

[0018] According to a third aspect of the present disclosure, a channel detection system is proposed, comprising: the channel detection device as described above; and a scanning laser radar arranged on at least one side of the channel for collecting point cloud data.

[0019] In some embodiments, the channel detection system further includes: a camera for capturing images of the object to be detected.

[0020] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the instructions are executed by a processor, the channel detection method as described above is implemented.

[0021] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0023] The present disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings.

[0024] FIG1 is a flow chart of a channel detection method according to some embodiments of the present disclosure;

[0025] FIG2 is a schematic diagram of the layout of a scanning laser radar according to some embodiments of the present disclosure;

[0026] FIG3a is a schematic diagram of a scanning result according to some embodiments of the present disclosure;

[0027] FIG3 b is a schematic diagram of a scanning result according to some embodiments of the present disclosure;

[0028] FIG4 is a flow chart of a channel detection method according to some embodiments of the present disclosure;

[0029] FIG5 is a schematic structural diagram of a channel detection device according to some embodiments of the present disclosure;

[0030] FIG6 is a schematic structural diagram of a channel detection device according to other embodiments of the present disclosure;

[0031] FIG7 is a schematic diagram of the composition of a channel detection system according to some embodiments of the present disclosure;

[0032] FIG8 is a schematic diagram of the structure of a computer system according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0033] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0034] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0035] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0036] Technologies, methods and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, such technologies, methods and equipment should be considered part of the authorization specification.

[0037] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0038] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0039] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0040] Related technologies typically use light curtains, photoelectric sensors, or input / output (I / O) laser radars to detect whether a person has entered a passageway. These systems suffer from low detection accuracy, blind spots, and high missed detection rates. For example, light curtains or photoelectric sensors cannot detect a person's presence if they are not moving or blocking the light curtain or sensor within the passageway. They cannot accurately determine the number of people entering the passageway, and cannot distinguish between multiple people entering and a single person blocking the light curtain for an extended period. If a person enters the passageway simultaneously with a vehicle, the light curtain is blocked by the vehicle, placing the person entering the passageway with the vehicle in a blind spot and making them undetectable. For example, I / O laser radars cannot distinguish between people and vehicles when they enter the passageway simultaneously. Furthermore, these systems cannot accurately determine the exact size of objects entering the passageway, making it impossible to accurately locate them.

[0041] Furthermore, channel detection methods in related technologies often only detect objects and provide alarm control for a single channel type. This results in the need to design and develop separate channel detection solutions for multiple channel types, wasting significant manpower and resources.

[0042] In light of this, this disclosure proposes a channel detection method that can accurately detect the type of objects entering different channel types and generate an alarm when necessary, thereby improving the security and protection of channel detection and control. Furthermore, since there is no need to develop separate channel detection algorithms for different channel types, this method saves manpower and material resources.

[0043] Figure 1 is a flow chart of a channel detection method according to some embodiments of the present disclosure. As shown in Figure 1 , the channel detection method includes steps S110 to S150.

[0044] In step S110 , point cloud data is received from a scanning laser radar disposed on at least one side of the channel.

[0045] In some embodiments, the channel detection method is performed by a channel detection device.

[0046] In some embodiments, a scanning laser radar (also known as an area laser radar or a laser radar scanner) is provided on one side of the channel. For example, one or more laser radars are provided on the left side of the channel.

[0047] In some embodiments, scanning laser radars are provided on both sides of the channel. For example, a first laser radar is provided on one side of the channel and a second laser radar is provided on the other side of the channel, and the first laser radar and the second laser radar are provided in an alternating manner.

[0048] For example, the first laser radar and the second laser radar can be staggered in the following manner: the first laser radar and the second laser radar are staggered in the horizontal direction (that is, not facing each other).

[0049] In the disclosed embodiments, by installing scanning LiDARs on both sides of the channel, blind spots can be reduced and detection accuracy can be improved. Furthermore, by staggering the LiDARs on both sides, for example by rotating the LiDARs' yaw angles so that their detection planes form a certain angle, signal interference between the two LiDARs can be reduced or avoided, further helping to improve the accuracy of object classification detection.

[0050] In some embodiments, the detection angle of the scanning laser radar is set to be greater than 180 degrees, thereby improving the coverage of the channel.

[0051] In some embodiments of the present disclosure, different numbers of LiDARs are deployed for different types of passageways, taking into account the varying detection accuracy requirements. For example, one LiDAR is installed on one side of a pedestrian passageway, while two or more LiDARs are installed on each side of a vehicle passageway. This allows the detection accuracy requirements of different passageways to be met at a lower cost.

[0052] In some embodiments, in step S110, point cloud data is received from a scanning laser radar set up on one side of the channel.

[0053] In some embodiments, in step S110, point cloud data is received from scanning laser radars disposed on both sides of the channel. For example, first point cloud data is received from a first laser radar disposed on one side of the channel, and second point cloud data is received from a second laser radar disposed on the other side of the channel.

[0054] It can be understood that one side or both sides of the channel refers to one side or both sides in the direction in which the object to be detected travels in the channel.

[0055] In step S120 , the outline of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel are determined based on the point cloud data.

[0056] In some embodiments, the outline of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel are determined based on first point cloud data received from a first laser radar set on one side of the channel and second point cloud data received from a second laser radar set on the other side of the channel. For example, the first point cloud data and the second point cloud data are aligned to obtain aligned point cloud data; based on the aligned point cloud data, the outline of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel are determined. For example, the first point cloud data and the second point cloud data are analyzed separately, and the outline of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel are determined based on the analysis results.

[0057] In some examples, the point cloud data includes the coordinates of the point cloud point, the distance from the point cloud point to the scanning lidar, and the acquisition timestamp corresponding to the point cloud point.

[0058] In some embodiments, in step S120, background point cloud data is removed from the point cloud data to obtain a point cloud point set of the object to be detected; based on the point cloud point set of the object to be detected, the outline of the object to be detected and the distance from the object to be detected to the scanning laser radar are determined; based on the distance from the object to be detected to the scanning laser radar and the scanning angle of the scanning laser radar, the distance from the object to be detected to at least one side of the channel is determined.

[0059] When there's no object to be detected in the channel, the point cloud data acquired by the LiDAR is considered background point cloud data. When there's an object to be detected in the channel, the point cloud data acquired by the LiDAR includes both background point cloud data and point cloud data of the object to be detected. The background point cloud data includes point cloud data obtained by the LiDAR scanning the other side of the channel.

[0060] When no object enters the channel, the vertical distance from one side of the channel to the side of the channel where the laser radar is located, obtained by measuring the laser radar, is a fixed value (i.e., the channel width); and when an object enters the channel, the vertical distance from the object to the side of the channel where the laser radar is located, obtained by measuring the laser radar, will become smaller. Therefore, in some embodiments, the background point cloud data is removed from the point cloud data based on the vertical distance from the point cloud point to the side of the channel where the laser radar is located, so as to obtain a point cloud point set of the object to be detected. For example, the point cloud point whose vertical distance to the side of the channel where the laser radar is located is less than a set value is used as the point cloud point set of the object to be detected. In some examples, the set value can be determined according to the width of the channel. For example, let the set value be equal to the width of the channel or equal to half the width of the channel. In some examples, the set value can be determined based on the statistical value of the width of the object passing through in the actual application scenario.

[0061] In some examples, after obtaining a point cloud set of the object to be detected, a contour detection algorithm is used to determine the contour of the object to be detected; the distance from the point cloud points of the object to be detected to the scanning laser radar is used to determine the distance; and based on the distance from the object to be detected to the scanning laser radar and the scanning angle of the scanning laser radar, the distance from the object to be detected to at least one side of the channel is determined.

[0062] In some examples, a first laser radar is positioned on one side of a channel, and a second laser radar is positioned on the other side of the channel. In these examples, the distance from the object to be detected to the first laser radar and the scanning angle of the first laser radar is used to determine the distance from the object to be detected to the side of the channel where the first laser radar is located; and the distance from the object to be detected to the second laser radar and the scanning angle of the second laser radar is used to determine the distance from the object to be detected to the side of the channel where the second laser radar is located.

[0063] In the embodiment of the present disclosure, the positioning of the object to be detected is achieved by determining the distance from the object to be detected to the scanning laser radar based on point cloud data; and the accurate estimation of the width, height and other dimensions of the object to be detected is achieved by determining the outline of the object to be detected based on point cloud data.

[0064] In some embodiments, step S120 further includes: determining, based on the point cloud data, the residence time of the object to be detected in the designated area within the channel.

[0065] For example, based on the coordinate information of the point cloud points of the object to be detected and the corresponding acquisition timestamp information, the time when the object to be detected enters the designated area in the channel and the time when the object to be detected leaves the designated area in the channel are determined; based on the time when the object to be detected leaves the designated area in the channel and the time when the object to be detected enters the designated area in the channel, the residence time of the object to be detected in the designated area in the channel is determined.

[0066] In step S130 , the category of the object to be detected is determined according to the outline of the object to be detected and the distance between the object to be detected and at least one side of the channel.

[0067] In the embodiment of the present disclosure, it is taken into account that objects of different categories have differences in outline and distance from both sides of the channel. For example, the outline of a person is different from the outline of a car. When a person and a car walk along the center line of the channel, the distances between the person and the car and the two sides of the channel are also different. Therefore, the category of the object to be detected can be determined based on these two factors.

[0068] Furthermore, considering that in some cases, the category of the object to be detected is determined only based on the distance or outline from the object to be detected to the channel, there may be a problem of inaccurate detection results. For example, when a person follows a car into a channel, if the object category is determined only based on the distance from the object to be detected to both sides of the channel, it is likely that the person and the car will not be treated as one object, which will lead to inaccurate category detection results. For example, when a person follows a car into a channel and the person is stationary, as shown in Figure 3b, the outline of the person determined by the point cloud data collected by the lidar will be elongated and distorted, resulting in inaccurate object category determined only based on the outline of the object to be detected. In view of this, in the embodiment of the present disclosure, the outline information is combined with the distance information to more accurately determine the object category.

[0069] In some embodiments, the outline of the object to be detected includes the width of the object to be detected. The width of the object to be detected refers to the length of the object to be detected along the front-to-back direction of the channel, and / or the length along the left-to-right direction of the channel. In these embodiments, the width of the object to be detected is compared with the width value range of objects of known categories, and a first candidate category of the object to be detected is determined based on the comparison result; a second candidate category of the object to be detected is determined based on the comparison result of the distance from the object to be detected to at least one side of the channel and the distance value range of objects of known categories to at least one side of the channel; and the category of the object to be detected is determined based on the first candidate category and the second candidate category.

[0070] In some embodiments, different width ranges and distance ranges are set for different types of objects. For example, the width of a small hatchback is set to 1.5 to 1.7 meters, and the distance to both sides of the passage is set to 0.65 to 0.75 meters. The width of a mid-size sedan is set to 1.7 to 1.8 meters, and the distance to both sides of the passage is set to 0.6 to 0.65 meters. The width of a large sedan is set to 1.8 to 2 meters, and the distance is set to 0.5 to 0.6 meters.

[0071] In some examples, when the first candidate category and the second candidate category are the same, the first candidate category is used as the category of the object to be detected. When the first candidate category and the second candidate category are different, the category of the object to be detected is determined by combining other auxiliary factors. For example, the category of the object to be detected is determined by combining factors such as the residence time of the object to be detected in the channel.

[0072] For example, assuming the width of the object to be detected is 0.5 meters, the width of the objects in the first category ranges from 1.8 meters to 2 meters, and the width of the objects in the second category ranges from 0.4 meters to 0.6 meters. If the width of the object to be detected falls within the width range of the objects in the second category, then the first candidate category is determined to be the second category. assuming the distance from the object to be detected to the left side of the channel is 0.95 meters, the distance from the objects in the first category ranges from 0.2 meters to 0.3 meters, and the distance from the objects in the second category ranges from 0.9 meters to 1 meter, then the second candidate category of the object to be detected is determined to be the second category. Furthermore, the category of the object to be detected is determined to be the second category.

[0073] In some embodiments, the outline of the object to be detected includes the width and height of the object to be detected. In these embodiments, the first candidate category of the object to be detected is determined based on the comparison results of the width and height by comparing the width of the object to be detected with the width value range of objects of known categories, and comparing the height of the object to be detected with the height value range of objects of known categories; the second candidate category of the object to be detected is determined based on the comparison results of the distance from the object to be detected to at least one side of the channel and the distance value range of objects of known categories to at least one side of the channel; and the category of the object to be detected is determined based on the first candidate category and the second candidate category.

[0074] In some embodiments, in step S130 , the category of the object to be detected is determined based on the outline of the object to be detected, the distance from the object to be detected to at least one side of the channel, and the residence time of the object to be detected in a designated area in the channel.

[0075] In some examples, the category of the object to be detected is determined as follows: a first candidate category of the object to be detected is determined based on a comparison result of the width of the object to be detected and a range of width values ​​of objects of known categories; a second candidate category of the object to be detected is determined based on a comparison result of the distance from the object to be detected to at least one side of the channel and a range of distance values ​​of objects of known categories to at least one side of the channel; a third candidate category of the object to be detected is determined based on a comparison result of the residence time of the object to be detected in a specified area in the channel and a range of residence time of objects of known categories in a specified area in the channel; the category of the object to be detected is determined based on the first candidate category, the second candidate category, and the third candidate category of the object to be detected.

[0076] In some embodiments, different dwell time value ranges are set for different categories of objects.

[0077] In some embodiments, the dwell time range is determined based on the length and speed of different types of objects. For example, if a vehicle is 20 meters long, travels at 3.6 km / h, and takes 20 seconds to pass through the entrance, the dwell time range can be set to 20 to 25 seconds.

[0078] In some examples, when at least two candidate categories among the first candidate category, the second candidate category, and the third candidate category are the same, the same candidate category is used as the category of the object to be detected.

[0079] In some examples, when the first candidate category, the second candidate category, and the third candidate category are all different, a detection failure prompt message is output.

[0080] For example, if the first, second, and third candidate categories are different, a detection failure prompt message is sent to the management terminal, or the detection failure prompt message is displayed through voice, text, lighting, etc. By outputting a detection failure prompt message when the object category cannot be determined, operation and maintenance personnel can be notified of the detection failure in a timely manner and intervene in a timely manner, thereby minimizing missed detections and improving safety protection.

[0081] In the embodiment of the present disclosure, the category of the object to be detected is determined based on multiple factors, such as the outline of the object to be detected, the distance from the object to be detected to at least one side of the channel, and the length of time the object to be detected stays in a specified area in the channel. Compared with determining the object category based on a single or two factors, this can alleviate the problem of inaccurate category detection results caused by outline distortion due to factors such as the object being stationary, and thus help improve the accuracy of category identification of the object to be detected.

[0082] In step S140 , a restricted object category corresponding to the channel type is acquired.

[0083] In some embodiments, the channel type and the corresponding restricted object category information are pre-stored. When the channel type is known, the restricted object category corresponding to the channel type is obtained from the storage module.

[0084] In some embodiments, when the channel type is a pedestrian channel, the corresponding restricted object category includes vehicles; when the channel type is a vehicle channel, the corresponding restricted object category includes people; when the channel type is a channel accessible to both people and vehicles, the corresponding restricted object category is the situation where both people and vehicles enter at the same time.

[0085] In step S150 , when the category of the object to be detected includes a restricted object category corresponding to the type of the channel, an alarm prompt message is output.

[0086] In some embodiments, when the channel type is a pedestrian channel, if the category of the object to be detected includes a vehicle, an alarm prompt information is output; otherwise, no alarm prompt information is output; when the channel type is a vehicle channel, if the category of the object to be detected includes a person, an alarm prompt information is output; otherwise, no alarm prompt information is output.

[0087] In some embodiments, when the channel type is a vehicle channel, if a person is detected outside the vehicle (such as the side, front, or rear of the vehicle), or only a person is detected, an alarm prompt message is output; otherwise, no alarm prompt message is output.

[0088] In some embodiments, in step S150, the alarm prompt information is output according to at least one of the following methods: sending the alarm prompt information to the management terminal; displaying the alarm prompt information through voice, text, lighting, etc.

[0089] In the disclosed embodiments, the above method can accurately detect the type of objects entering different types of channels, distinguish between people and vehicles, and generate an alarm when necessary, thereby helping to improve the safety and protection of channel detection and control. Furthermore, since there is no need to develop separate channel detection algorithms for different types of channels, human and material resources are saved.

[0090] Figure 2 is a schematic diagram of the layout of a scanning laser radar according to some embodiments of the present disclosure. As shown in Figure 2, in some embodiments, scanning laser radars are set on both sides of the channel, including a first scanning laser radar 201 and a second scanning laser radar 202.

[0091] In some embodiments, the first scanning laser radar 201 and the second scanning laser radar 202 are arranged in an alternating manner.

[0092] In some embodiments, the detection surfaces of the first scanning laser radar 201 and the second scanning laser radar 202 are perpendicular or nearly perpendicular to the ground 205 , thereby helping to improve the coverage of the channel detection area.

[0093] In some embodiments, the detection angles of the first scanning laser radar 201 and the second scanning laser radar 202 exceed 180 degrees to improve the coverage of the detection area.

[0094] When pedestrian 203 and vehicle 204 enter the lane together, vehicle 204 may obstruct pedestrian 203. With only the first scanning LiDAR 201 installed, pedestrian 203 entering the lane cannot be identified. Installing two or more scanning LiDARs allows for complementary detection, helping to eliminate blind spots and improve the accuracy of subsequent object detection.

[0095] In the embodiment of the present disclosure, a scanning laser radar as shown in Figure 2 is set up, and the point cloud data collected by the scanning laser radar is read in real time by a channel detection device. The outer contour of the object to be detected and the distance from the object to be detected to at least one side of the channel are determined according to the point cloud data, and the category of the object entering the channel is determined accordingly, thereby realizing personnel detection, vehicle detection, etc.

[0096] Figure 3a is a schematic diagram of a scanning result according to some embodiments of the present disclosure. When multiple people follow a vehicle into a passage, point cloud data is collected by a scanning laser radar and analyzed to obtain a schematic diagram of a scanning result as shown in Figure 3a.

[0097] Figure 3b is a schematic diagram of scanning results according to some embodiments of the present disclosure. When a person enters a passageway with a vehicle and stands still for a long time beside the vehicle, point cloud data is collected by a scanning lidar and analyzed to obtain the schematic diagram of the scanning results shown in Figure 3b.

[0098] In some examples, when a person enters a passageway along with a vehicle, the passageway detection device analyzes the point cloud data to determine that there are two objects to be detected and obtains the width of each object. The category of the object to be detected is determined by comparing the width of each object to a range of width values ​​for objects of known categories. To further improve detection accuracy, the width of the object to be detected can be combined with the distance from the object to at least one side of the passageway and the time the object remains in a designated area within the passageway to determine the category of the object to be detected.

[0099] In actual application scenarios, there may also be situations where a person enters the vehicle lane alone, such as a single person entering without stopping, a single person entering after stopping, multiple people entering side by side without stopping, multiple people entering side by side after stopping, etc., as well as vehicles entering the pedestrian lane. By collecting and analyzing point cloud data, the category of the object entering the lane can be accurately identified.

[0100] Figure 4 is a flow chart of a channel detection method according to some embodiments of the present disclosure. As shown in Figure 4, the channel detection method includes steps S401 to S409.

[0101] In step S401, in response to the scanning laser radar activation condition being met, the scanning laser radar is controlled to be activated.

[0102] In some embodiments, the channel detection method shown in FIG4 is performed by a channel detection device.

[0103] In some embodiments, the activation conditions of the scanning laser radar include at least one of the following: the current time period is a set channel detection time period; and the presence of an object to be detected is determined based on an image captured by a camera set corresponding to the channel.

[0104] For example, after reaching the set channel detection period, the scanning laser radar set on the channel side is turned on at a fixed time.

[0105] In some examples, different channel types are set for different detection periods. For example, from 9:00 AM to 11:00 AM, the channel type is set to a vehicle channel; from 2:00 PM to 3:00 PM, the channel type is set to a pedestrian channel. By setting the channel type to different types during different detection periods, channel reusability can be improved and costs can be reduced.

[0106] For example, after determining that an object has entered the channel through the image collected by the camera, the scanning laser radar installed on the channel side is turned on.

[0107] In the embodiment of the present disclosure, the energy consumption required for channel detection can be reduced by turning on the laser radar after the turning-on conditions are met and not turning on the laser radar when the turning-on conditions are not met.

[0108] In step S402, point cloud data is received from a scanning laser radar installed on at least one side of the channel.

[0109] In some embodiments, scanning laser radars are provided on both sides of the passage. In these embodiments, point cloud data is received from the scanning laser radars provided on both sides of the passage.

[0110] In step S403 , the outline of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel are determined based on the point cloud data.

[0111] In some embodiments, in step S403, background point cloud data is removed from the point cloud data to obtain a point cloud point set of the object to be detected; based on the point cloud point set of the object to be detected, the outline of the object to be detected and the distance from the object to be detected to the scanning laser radar are determined; based on the distance from the object to be detected to the scanning laser radar and the scanning angle of the scanning laser radar, the distance from the object to be detected to at least one side of the channel is determined.

[0112] In some embodiments, in addition to determining the outline of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel based on the point cloud data, the residence time of the object to be detected in a specified area in the channel is also determined based on the point cloud data.

[0113] In step S404, the category of the object to be detected is determined according to the outline of the object to be detected and the distance between the object to be detected and at least one side of the channel.

[0114] Regarding how to determine the category of the object to be detected, please refer to the above description.

[0115] In step S405 , the restricted object category corresponding to the channel type is acquired.

[0116] In some embodiments, the channel type and the corresponding restricted object category information are pre-stored. When the channel type is known, the restricted object category corresponding to the channel type is obtained from the storage module.

[0117] In some embodiments, information such as channel identification, corresponding channel type, and restricted object category is pre-stored. In step S405, the channel type and corresponding restricted object category are obtained from the storage module according to the channel identification.

[0118] In some embodiments, when the type of the passage is a pedestrian passage, the corresponding restricted object category includes vehicles; when the type of the passage is a vehicle passage, the corresponding restricted object category includes people.

[0119] In some embodiments, the channel detection method further includes receiving a request to modify the channel type, and modifying the channel type information stored in the storage module based on the request. For example, after a user submits a request to modify the channel type through a human-computer interaction interface, the channel detection device modifies the channel type information stored in the database based on the request. Supporting the channel type modification function facilitates channel reuse and reduces costs.

[0120] In step S406 , it is determined whether the category of the object to be detected includes a restricted object category.

[0121] If the category of the object to be detected includes the restricted object category, step S407 and step S408 are executed; otherwise, step S409 is executed.

[0122] In some embodiments, when the channel type is a pedestrian channel, the restricted object category includes vehicles. In these embodiments, it is determined whether the object to be detected includes a vehicle.

[0123] In some embodiments, when the channel type is a vehicle channel, the restricted object category includes people. In these embodiments, it is determined whether the object to be detected includes people.

[0124] In some embodiments, when the channel type is a channel that is passable by both people and vehicles, the restricted object category includes people and vehicles that exist at the same time. In these embodiments, it is determined whether the object to be detected includes both people and vehicles.

[0125] In step S407, an alarm prompt message is output.

[0126] In step S408, the X-ray scanning device is controlled to be turned off.

[0127] In some embodiments, step S407 is performed first and then step S408. In addition, in specific implementations, step S407 and step S408 can also be performed in parallel, or step S408 can be performed first and then step S407.

[0128] In some embodiments, for all channel types, when it is determined that the category of the object to be detected includes a restricted object category, step S407 and step S408 are performed.

[0129] In some embodiments, for some channel types, when it is determined that the category of the object to be detected includes the restricted object category, steps S407 and S408 are executed; for another part of the channel types, when it is determined that the category of the object to be detected includes the restricted object category, only step S407 is executed.

[0130] For example, the channel detection method can be applied to a vehicle security inspection scenario based on an X-ray scanner. In this scenario, if the channel type is a vehicle channel and the category of the object to be detected includes people, the X-ray scanner is controlled to be turned off.

[0131] In the embodiment of the present disclosure, when the category of the object to be detected includes the restricted object category, not only an alarm prompt message is output but also the X-ray scanning device is controlled to be turned off, thereby improving the safety protection effect.

[0132] In step S409, the process ends.

[0133] In the disclosed embodiments, the above process enables accurate detection of the type of object entering a channel for each type of channel and generates an alarm when necessary, thereby improving the security and protection of channel detection and control. Furthermore, since there is no need to develop separate channel detection algorithms for different types of channels, this saves manpower and material resources.

[0134] FIG5 is a schematic diagram of the structure of a channel detection device according to some embodiments of the present disclosure. As shown in FIG5 , the channel detection device 500 includes a receiving module 501 , a first determining module 502 , a second determining module 503 , an acquiring module 504 , and an output module 505 .

[0135] The receiving module 501 is configured to receive point cloud data from a scanning laser radar disposed on at least one side of the channel.

[0136] In some embodiments, a scanning laser radar (also referred to as an area laser radar or a laser radar scanner) is provided on one side of the channel. In these embodiments, point cloud data is received from the scanning laser radar provided on one side of the channel.

[0137] In some embodiments, scanning LiDARs are located on both sides of the passageway. For example, a first LiDAR is located on one side of the passageway, and a second LiDAR is located on the other side of the passageway, with the first and second LiDARs interleaved. In these embodiments, point cloud data is received from the scanning LiDARs located on both sides of the passageway.

[0138] The first determination module 502 is configured to determine the outline of the object to be detected entering the channel and the distance between the object to be detected and at least one side of the channel according to the point cloud data.

[0139] In some embodiments, the first determination module 502 is configured to: remove background point cloud data from the point cloud data to obtain a point cloud point set of the object to be detected; determine the outline of the object to be detected and the distance from the object to be detected to the scanning laser radar based on the point cloud point set of the object to be detected; determine the distance from the object to be detected to at least one side of the channel based on the distance from the object to be detected to the scanning laser radar and the scanning angle of the scanning laser radar.

[0140] In some embodiments, the first determination module 502 is further configured to determine, based on the point cloud data, the residence time of the object to be detected in the specified area of ​​the channel.

[0141] For example, based on the coordinate information of the point cloud points of the object to be detected and the corresponding acquisition timestamp information, the time when the object to be detected enters the designated area in the channel and the time when the object to be detected leaves the designated area in the channel are determined; based on the time when the object to be detected leaves the designated area in the channel and the time when the object to be detected enters the designated area in the channel, the residence time of the object to be detected in the designated area in the channel is determined.

[0142] The second determination module 503 is configured to determine the category of the object to be detected according to the outline of the object to be detected and the distance between the object to be detected and at least one side of the channel.

[0143] In some embodiments, the outline of the object to be detected includes the width of the object to be detected. Wherein, the width of the object to be detected refers to the length of the object to be detected along the front-to-back direction of the channel, and / or the length along the left-to-right direction of the channel. In these embodiments, the second determination module 503 is configured to: compare the width of the object to be detected with the width value range of objects of known categories, and determine the first candidate category of the object to be detected based on the comparison result; determine the second candidate category of the object to be detected based on the comparison result of the distance from the object to be detected to at least one side of the channel and the distance value range of objects of known categories to at least one side of the channel; and determine the category of the object to be detected based on the first candidate category and the second candidate category.

[0144] In some embodiments, the second determination module 503 is configured to determine the category of the object to be detected based on the outline of the object to be detected, the distance from the object to be detected to at least one side of the channel, and the residence time of the object to be detected in a specified area in the channel.

[0145] In some examples, the second determination module 503 determines the category of the object to be detected as follows: determining a first candidate category of the object to be detected based on a comparison result of the width of the object to be detected with a range of width values ​​of objects of a known category; determining a second candidate category of the object to be detected based on a comparison result of the distance from the object to be detected to at least one side of the channel with a range of distance values ​​of objects of a known category to at least one side of the channel; determining a third candidate category of the object to be detected based on a comparison result of the residence time of the object to be detected in a specified area in the channel with a range of residence time of objects of a known category in a specified area in the channel; determining the category of the object to be detected based on the first candidate category, the second candidate category, and the third candidate category of the object to be detected.

[0146] The acquisition module 504 is configured to acquire a restricted object category corresponding to the channel type.

[0147] In some embodiments, the channel type and the corresponding restricted object category information are pre-stored. When the channel type is known, the acquisition module 504 acquires the restricted object category corresponding to the channel type from the storage module.

[0148] In some embodiments, when the type of the passage is a pedestrian passage, the corresponding restricted object category includes vehicles; when the type of the passage is a vehicle passage, the corresponding restricted object category includes people.

[0149] The output module 505 is configured to output an alarm prompt message when the category of the object to be detected includes a restricted object category corresponding to the type of the channel.

[0150] In some embodiments, the channel detection device further includes a first control module. The first control module is configured to control the X-ray scanning device to be turned off when the channel type is a vehicle channel and the category of the object to be detected includes a person.

[0151] In some embodiments, the channel detection device further includes a second control module configured to control the scanning laser radar to turn on in response to a turning-on condition of the scanning laser radar being met.

[0152] In some embodiments, the activation conditions of the scanning laser radar include at least one of the following: the current time period is a set channel detection time period; and the presence of an object to be detected is determined based on an image captured by a camera set corresponding to the channel.

[0153] In the disclosed embodiments, the above device can accurately detect the type of objects entering different types of channels and generate an alarm when necessary, thereby improving the security and protection of channel detection and control. Furthermore, since there is no need to develop separate channel detection algorithms for different types of channels, human and material resources are saved.

[0154] FIG6 is a schematic structural diagram of a channel detection device according to other embodiments of the present disclosure.

[0155] As shown in FIG6 , a channel detection apparatus 600 includes a memory 601 and a processor 602 coupled to the memory 601. The memory 601 is configured to store instructions for executing the channel detection method according to the embodiments. The processor 602 is configured to execute the channel detection method according to any of the embodiments of the present disclosure based on the instructions stored in the memory 601.

[0156] FIG7 is a schematic diagram of the components of a channel detection system according to some embodiments of the present disclosure. As shown in FIG7 , a channel detection system 700 includes a channel detection device 701 and a scanning laser radar 702 .

[0157] The scanning laser radar 702 is arranged on at least one side of the channel and is used to collect point cloud data.

[0158] In some embodiments, the scanning laser radar 702 includes a first laser radar arranged on one side of the channel and a second laser radar arranged on the other side of the channel, and the first laser radar and the second laser radar are arranged in an alternating manner.

[0159] The channel detection device 701 is configured to execute the channel detection method as described above.

[0160] In some embodiments, the channel detection system 700 further includes a camera for capturing images of the object to be detected. For example, one or more cameras are positioned near the channel. The channel detection device 701 is configured to activate a scanning laser radar when, based on the image, it is determined that an object to be detected is about to enter the channel.

[0161] In the disclosed embodiments, the above system can accurately detect the type of objects entering different types of channels and generate an alarm when necessary, thereby improving the security and protection of channel detection and control. Furthermore, since there is no need to develop separate channel detection algorithms for different types of channels, human and material resources are saved.

[0162] FIG8 is a schematic diagram of the structure of a computer system according to some embodiments of the present disclosure.

[0163] As shown in Figure 8, a computer system 800 may be implemented as a general-purpose computing device and includes a memory 801, a processor 802, and a bus 803 connecting various system components.

[0164] The memory 801 may include, for example, a system memory, a non-volatile storage medium, and the like. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs. The system memory may include a volatile storage medium, such as a random access memory (RAM) and / or a cache memory. The non-volatile storage medium may store, for example, instructions for executing at least one corresponding embodiment of the channel detection method. The non-volatile storage medium may include, but is not limited to, a disk memory, an optical memory, a flash memory, and the like.

[0165] The processor 802 can be implemented as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, or as discrete hardware components such as discrete gates or transistors. Accordingly, each module, such as the receiving module, the first determination module, the second determination module, the acquisition module, and the output module, can be implemented by a central processing unit (CPU) executing instructions in a memory for executing corresponding steps, or by dedicated circuits for executing corresponding steps.

[0166] The bus 803 may use any of a variety of bus architectures, including, but not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, and a Peripheral Component Interconnect (PCI) bus.

[0167] The computer system 800 interfaces 804, 805, and 806, as well as the memory 801 and the processor 802, can be connected via a bus 803. The input / output interface 804 provides a connection interface for input / output devices such as a display, mouse, and keyboard. The network interface 805 provides a connection interface for various networked devices. The storage interface 806 provides a connection interface for external storage devices such as floppy disks, USB flash drives, and SD cards.

[0168] Here, various aspects of the present disclosure are described with reference to flowcharts and / or block diagrams of methods, devices, and computer program products according to embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks, can be implemented by computer-readable program instructions.

[0169] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable device to produce a machine, so that the processor executes the instructions to produce means for implementing the functions specified in one or more blocks in the flowcharts and / or block diagrams.

[0170] These computer-readable program instructions may also be stored in a computer-readable memory, which cause the computer to operate in a specific manner to produce an article of manufacture, including instructions for implementing the functions specified in one or more blocks in the flowcharts and / or block diagrams.

[0171] The present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

[0172] The channel detection method, device, and system in the above-mentioned embodiments can improve the safety protection effect of channel detection and control while saving the manpower and material costs of channel detection solution design and development.

[0173] The channel detection method, device, and system according to the present disclosure have been described in detail. To avoid obscuring the concepts of the present disclosure, some details known in the art have been omitted. Based on the above description, those skilled in the art will fully understand how to implement the technical solutions disclosed herein.

Claims

1. A channel detection method, comprising: receiving point cloud data from a scanning laser radar disposed on at least one side of the passage; Determine, according to the point cloud data, the contour of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel; Determining the category of the object to be detected according to the outline of the object to be detected and the distance from the object to be detected to at least one side of the channel; Acquire a restricted object category corresponding to the type of the channel; In the case where the category of the object to be detected includes a restricted object category corresponding to the type of the channel, an alarm prompt message is output.

2. The channel detection method according to claim 1, wherein: Determining the category of the object to be detected according to the contour of the object to be detected and the distance from the object to be detected to at least one side of the channel includes: The category of the object to be detected is determined according to the outline of the object to be detected, the distance from the object to be detected to at least one side of the channel, and the residence time of the object to be detected in a designated area in the channel.

3. The channel detection method according to claim 2, wherein: The outline of the object to be detected includes the width of the object to be detected, and determining the category of the object to be detected according to the outline of the object to be detected, the distance from the object to be detected to at least one side of the channel, and the residence time of the object to be detected in a designated area of ​​the channel includes: Determining a first candidate category of the object to be detected according to a comparison result between the width of the object to be detected and a width value range of objects of known categories; Determining a second candidate category of the object to be detected according to a comparison result of a distance from the object to be detected to at least one side of the channel and a range of distances from objects of known categories to at least one side of the channel; Determine a third candidate category of the object to be detected according to a comparison result of the residence time of the object to be detected in the designated area of ​​the channel with a value range of the residence time of objects of known categories in the designated area of ​​the channel; The category of the object to be detected is determined according to the first candidate category, the second candidate category, and the third candidate category of the object to be detected.

4. The channel detection method according to claim 3, wherein: Determining the category of the object to be detected according to the first candidate category, the second candidate category, and the third candidate category of the object to be detected includes: When at least two candidate categories among the first candidate category, the second candidate category, and the third candidate category are the same, the same candidate category is used as the category of the object to be detected.

5. The channel detection method according to claim 3 further comprises: When the first candidate category, the second candidate category, and the third candidate category are all different, a detection failure prompt message is output.

6. The channel detection method according to any one of claims 1 to 5, wherein: The type of the channel is a vehicle channel or a pedestrian channel. The restricted object categories corresponding to the vehicle channel include people, and the restricted object categories corresponding to the pedestrian channel include vehicles.

7. The channel detection method according to claim 6, wherein: The channel detection method is applied to a vehicle security inspection scenario based on an X-ray scanning device, and the channel detection method further includes: When the type of the channel is a vehicle passage and the category of the objects to be detected includes people, the X-ray scanning device is controlled to be turned off.

8. The channel detection method according to any one of claims 1 to 7, further comprising: Before receiving point cloud data from a scanning laser radar arranged on at least one side of the channel, in response to satisfying a start condition of the scanning laser radar, the scanning laser radar is controlled to be turned on.

9. The channel detection method according to claim 8, wherein: The activation condition of the scanning laser radar includes at least one of the following: The current period is the set channel detection period; The existence of the object to be detected is determined based on an image captured by a camera arranged corresponding to the channel.

10. The channel detection method according to any one of claims 1 to 9, wherein: The scanning laser radar arranged on at least one side of the channel includes a first laser radar arranged on one side of the channel and a second laser radar arranged on the other side of the channel, and the first laser radar and the second laser radar are arranged alternately.

11. The channel detection method according to any one of claims 1 to 10, wherein: Determining the contour of the object to be detected entering the channel and the distance from the object to be detected to at least one side of the channel according to the point cloud data includes: Removing background point cloud data from the point cloud data to obtain a point cloud point set of the object to be detected; Determine the contour of the object to be detected and the distance from the object to be detected to the scanning laser radar according to the point cloud point set of the object to be detected; The distance between the object to be detected and at least one side of the channel is determined according to the distance between the object to be detected and the scanning laser radar and the scanning angle of the scanning laser radar.

12. A channel detection device, comprising: A receiving module is configured to receive point cloud data from a scanning laser radar disposed on at least one side of the channel; A first determination module is configured to determine, based on the point cloud data, a contour of an object to be detected entering the channel and a distance from the object to be detected to at least one side of the channel; A second determination module is configured to determine the category of the object to be detected according to the outline of the object to be detected and the distance from the object to be detected to at least one side of the channel; An acquisition module, configured to acquire a restricted object category corresponding to the type of the channel; The output module is configured to output an alarm prompt message when the category of the object to be detected includes a restricted object category corresponding to the type of the channel.

13. A channel detection device, comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the channel detection method according to any one of claims 1 to 11 based on instructions stored in the memory.

14. A channel detection system, comprising: The channel detection device as claimed in claim 12 or 13; A scanning laser radar is arranged on at least one side of the channel to collect point cloud data.

15. The channel detection system according to claim 14, further comprising: The camera is used to collect images of the object to be detected. 16 . A computer-readable storage medium having computer program instructions stored thereon, wherein the instructions are executed by a processor to implement the channel detection method according to any one of claims 1 to 11.

17. A computer program comprising: Instructions, when executed by a processor, cause the processor to perform the channel detection method according to any one of claims 1 to 11.

Images