CN114312853A - System, method, apparatus and storage medium for object detection - Google Patents
System, method, apparatus and storage medium for object detection Download PDFInfo
- Publication number
- CN114312853A CN114312853A CN202111472456.4A CN202111472456A CN114312853A CN 114312853 A CN114312853 A CN 114312853A CN 202111472456 A CN202111472456 A CN 202111472456A CN 114312853 A CN114312853 A CN 114312853A
- Authority
- CN
- China
- Prior art keywords
- laser radar
- door
- point cloud
- radar device
- boundary information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 110
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000003860 storage Methods 0.000 title claims abstract description 7
- 230000008569 process Effects 0.000 claims abstract description 33
- 238000005070 sampling Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 7
- 238000003825 pressing Methods 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000009434 installation Methods 0.000 description 10
- 238000004140 cleaning Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 239000000428 dust Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 239000004071 soot Substances 0.000 description 3
- 230000004888 barrier function Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- PCTMTFRHKVHKIS-BMFZQQSSSA-N (1s,3r,4e,6e,8e,10e,12e,14e,16e,18s,19r,20r,21s,25r,27r,30r,31r,33s,35r,37s,38r)-3-[(2r,3s,4s,5s,6r)-4-amino-3,5-dihydroxy-6-methyloxan-2-yl]oxy-19,25,27,30,31,33,35,37-octahydroxy-18,20,21-trimethyl-23-oxo-22,39-dioxabicyclo[33.3.1]nonatriaconta-4,6,8,10 Chemical compound C1C=C2C[C@@H](OS(O)(=O)=O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2.O[C@H]1[C@@H](N)[C@H](O)[C@@H](C)O[C@H]1O[C@H]1/C=C/C=C/C=C/C=C/C=C/C=C/C=C/[C@H](C)[C@@H](O)[C@@H](C)[C@H](C)OC(=O)C[C@H](O)C[C@H](O)CC[C@@H](O)[C@H](O)C[C@H](O)C[C@](O)(C[C@H](O)[C@H]2C(O)=O)O[C@H]2C1 PCTMTFRHKVHKIS-BMFZQQSSSA-N 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Landscapes
- Optical Radar Systems And Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
The application discloses a system, a method and a device for target detection and a storage medium, and belongs to the technical field of intelligent monitoring. The system comprises: installing support, driving motor, laser radar device, the installing support includes fixing base and swing arm. The fixed seat is arranged at the top end designated position of the inner side of the shielding door on the train platform, and the inner side is close to one side of the track of the mobile traffic train. The driving motor is positioned on the fixed seat. A rotating shaft of the driving motor is fixedly connected with the swinging arm so as to drive the swinging arm to do circular arc swinging in a gap area between the shielding door and the train door during rotation, wherein the tangential direction of a circular arc track of the circular arc swinging is perpendicular to the traveling direction of the moving traffic train. The laser radar device is positioned on the inner wall of the swing arm and used for detecting a target in the swinging process of the swing arm. Realize target detection through the laser radar device, compare in through preventing pressing from both sides the baffle, the detection validity of this system is higher, can avoid as far as possible missing and examine.
Description
Technical Field
The present application relates to the field of intelligent monitoring technologies, and in particular, to a system, a method, an apparatus, and a storage medium for target detection.
Background
At present, mobile traffic trains such as subway trains, high-speed trains and motor trains are becoming popular choices for people to go out. In some scenarios, to ensure passenger safety, a screen door is typically set up on the train platform. Certain clearance exists between the shielding door and the train door, and the phenomenon that people and objects are clamped needs to be avoided in the process of passengers getting on and off the train.
In the related art, in order to prevent the screen door from being pinched, an anti-pinch baffle plate having a height of about 60 cm and a width of about 15 cm is generally installed below the inner side of the screen door. If a passenger is clamped between the shielding door and the train door, when the shielding door is closed, the anti-clamping baffle can be blocked by the legs of the passenger, so that the shielding door cannot be closed.
However, if a passenger is detected by bypassing the anti-pinch barrier and is squeezed out of the train door from the end of the moving traffic train, a pinched event may be initiated between the gaps. In this regard, the solution, while simple and low cost, is prone to cause missed inspections.
Disclosure of Invention
The embodiment of the application provides a system, a method and a device for target detection and a storage medium. The problem that detection is easy to cause missed detection through an anti-pinch baffle in the related technology can be solved. The technical scheme is as follows:
in a first aspect, a system for object detection is provided, the system comprising: the laser radar device comprises a mounting bracket, a driving motor and a laser radar device, wherein the mounting bracket comprises a fixed seat and a swing arm;
the fixed seat is arranged at the designated position of the top end of the inner side of the shielding door on the train platform, and the inner side is close to one side of the track of the mobile traffic train;
the driving motor is positioned on the fixed seat;
the rotating shaft of the driving motor is fixedly connected with the swinging arm so as to drive the swinging arm to do circular arc swinging in a gap area between the shielding door and the train door during rotation, wherein the tangential direction of a circular arc track of the circular arc swinging is vertical to the traveling direction of the moving traffic train;
the laser radar device is located on the inner wall of the swing arm and used for detecting a target in the swing process of the swing arm.
In a second aspect, based on the system provided in the first aspect, the present application provides a method for target detection, where the lidar device projects a light curtain downwards along the direction of travel, and the method includes:
in the process of target detection of the laser radar device, constructing a three-dimensional scene according to first detection data of the laser radar device and first swing angles of the laser radar device at various swing positions to obtain a first three-dimensional scene;
determining a point cloud set of the clearance area from the point clouds of the first three-dimensional scene according to target boundary information, wherein the target boundary information is used for indicating the boundary of the train door, the boundary of the screen door and the boundary of the clearance area;
and if the gap area has the target according to the point cloud set of the gap area, carrying out early warning prompt.
As an example of the present application, the determining the gap area existence target according to the point cloud set of the gap area includes:
clustering the point cloud sets in the gap area to obtain at least one point cloud block;
determining three-dimensional dimensions of each of the at least one point cloud block, the three-dimensional dimensions including a length, a width, and a height;
if a corresponding point cloud block with a three-dimensional size meeting a target size condition exists in the at least one point cloud block, determining that the target exists in the gap area, wherein the fact that the three-dimensional size meets the target size condition means that: a length of the three-dimensional dimension is greater than or equal to a length dimension threshold, and/or a width of the three-dimensional dimension is greater than or equal to a width dimension threshold, and/or a height of the three-dimensional dimension is greater than or equal to a height dimension threshold.
As an example of the present application, the determining manner of the target boundary information includes:
controlling the laser radar device to swing within a preset angle range at a specified speed;
in the process of rotating the laser radar device, constructing a three-dimensional scene according to second detection data of the laser radar device and second swing angles of the laser radar device at each swing position to obtain a second three-dimensional scene;
and extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the gap area from the second three-dimensional scene to obtain the target boundary information.
As an example of the present application, the constructing the three-dimensional scene according to the second detection data of the lidar device and the second swing angle of the lidar device at each swing position includes:
establishing a rectangular coordinate system by taking a light emitting point of the laser radar device as an origin of a coordinate system, taking a vertical downward direction as a Z axis, taking a traveling direction of the mobile traffic train as an X axis, and taking a direction perpendicular to the traveling direction of the mobile traffic train as a Y axis;
determining three-dimensional coordinate information of each ranging point in each frame in the rectangular coordinate system according to the ranging value of each ranging point in each frame, the second swing angle corresponding to each frame and the scanning angle resolution of the laser radar device;
and constructing a three-dimensional scene according to the three-dimensional coordinate information of each ranging point in each frame.
As an example of the application, the extracting boundary information of the train door, boundary information of the screen door, and boundary information of the gap area from the second three-dimensional scene to obtain the target boundary information includes:
determining a normal vector of each point cloud in the second three-dimensional reconstruction scene;
clustering according to the directionality of the normal vector of each point cloud to obtain a plurality of point cloud sets, wherein the plurality of point cloud sets comprise the point cloud set of the train door, the point cloud set of the shield door and the point cloud set of the gap area;
and extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the clearance area according to the plurality of point cloud sets to obtain the target boundary information.
As an example of the application, the extracting, according to the plurality of point cloud sets, the boundary information of the train door, the boundary information of the screen door, and the boundary information of the gap area to obtain the target boundary information includes:
determining the minimum y component value and the maximum z component value in the point cloud set of the train door as the boundary information of the train door;
determining the maximum y component value and the maximum z component value in the point cloud set of the screen door as the boundary information of the screen door;
and carrying out random plane sampling on the point cloud set of the gap area, and determining an obtained z-component value of a sampling plane as boundary information of the gap area.
As an example of the present application, the method further comprises:
according to the target boundary information, respectively determining a first angle range corresponding to the shielding door, a second angle range corresponding to the gap area and a third angle range corresponding to the train door;
control the lidar device is in with first predetermined speed swing in the first angle scope with the swing of second predetermined speed in the second angle scope, and the swing of third predetermined speed in the third angle scope, first predetermined speed with the third predetermined speed all is greater than the second predetermined speed.
As an example of the present application, the method further comprises:
in an initial state, controlling the laser radar device to stop at a specified position, wherein the specified position enables a detection view field of the laser radar device to be a third angle range corresponding to the train door;
detecting the state of the train door when the detection value of the laser radar device changes;
and if the train door is switched from the opening state to the closing state, controlling the laser radar device to start swinging, and starting target detection by the laser radar device in the swinging process.
As an example of the present application, after detecting the state of the train door, the method further includes:
and sending a notification message to the screen door according to the state of the train door, wherein the notification message is used for enabling the screen door and the train door to be synchronously opened or closed.
In a third aspect, an apparatus for object detection is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method according to any one of the second aspect when executing the computer program.
In a fourth aspect, a computer-readable storage medium is provided, having instructions stored thereon, which when executed by a processor, implement the method of any of the second aspects.
In a fifth aspect, there is provided a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of the second aspect described above.
The technical scheme provided by the embodiment of the application has the following beneficial effects:
and performing target detection based on the provided system, and in the detection process of the laser radar device, constructing a three-dimensional scene according to first detection data of the laser radar device and first swing angles of the laser radar device at all swing positions to obtain a first three-dimensional scene. And determining a point cloud set of the gap area from the point clouds of the first three-dimensional scene according to target boundary information, wherein the target boundary information is used for indicating the boundary of the train door, the boundary of the screen door and the boundary of the gap area. If the target exists in the gap area according to the point cloud set of the gap area, it is indicated that the gap area has the invader, and therefore early warning prompt is carried out. Also this system passes through the laser radar device and realizes target detection, compares in through preventing pressing from both sides the baffle, and this system's detection validity is higher, can avoid as far as possible missing the detection.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram illustrating a system for object detection in accordance with an exemplary embodiment;
FIG. 2 is a schematic diagram of a system for object detection according to another exemplary embodiment;
FIG. 3 is a flow diagram illustrating a method of object detection in accordance with an exemplary embodiment;
FIG. 4 is a schematic diagram illustrating an application scenario in accordance with an illustrative embodiment;
FIG. 5 is a schematic illustration of a scan angle of a light ray according to an exemplary embodiment;
FIG. 6 is a schematic illustration of a scanning effect of a lidar apparatus according to an exemplary embodiment;
FIG. 7 is a schematic illustration of a scanning effect of a lidar apparatus according to another exemplary embodiment;
FIG. 8 is a schematic illustration of a range of swing angles shown in accordance with an exemplary embodiment;
FIG. 9 is a schematic illustration of a range of swing angles according to another exemplary embodiment;
FIG. 10 is a flow chart illustrating a method of object detection in accordance with another exemplary embodiment;
FIG. 11 is a schematic illustration of the detection effect of a mobile traffic train after arrival in accordance with an exemplary embodiment;
fig. 12 is a schematic diagram illustrating a structure of an apparatus for object detection according to an exemplary embodiment.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
It should be understood that reference to "a plurality" in this application means two or more. In the description of the present application, "/" indicates an OR meaning, for example, A/B may indicate A or B; "and/or" herein is merely an association describing an associated object, and means that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, for the convenience of clearly describing the technical solutions of the present application, the terms "first", "second", and the like are used to distinguish the same items or similar items having substantially the same functions and actions. Those skilled in the art will appreciate that the terms "first," "second," etc. do not denote any order or quantity, nor do the terms "first," "second," etc. denote any order or importance.
First, a system for object detection provided in the embodiments of the present application will be described.
Referring to fig. 1, fig. 1 is a schematic diagram illustrating a system for object detection according to an exemplary embodiment. The system comprises: installing support 1, driving motor 2, laser radar device 3. Wherein the mounting bracket 1 comprises a fixed base 11 and a swing arm 12.
As an example of the present application, the fixed seat 11 is installed at a top end designated position of an inner side of the screen door 4 on the train platform, wherein the inner side refers to a side close to a track of the mobile traffic train, and the top end designated position can be set according to actual requirements. The driving motor 2 is located on the fixed seat 11, a rotating shaft of the driving motor 2 is fixedly connected with the swing arm 12, so that the swing arm 12 is driven to do circular arc swing in a gap area between the screen door and the train door when rotating, wherein the tangential direction of a circular arc track of the circular arc swing is perpendicular to the traveling direction of the moving traffic train. The laser radar device 3 is located on the inner wall of the oscillating arm 12 for target detection during oscillation of the oscillating arm 12. That is, under the driving of the driving motor 2, the laser radar apparatus 3 swings within a certain swing angle range, and performs target detection during the swing.
In one embodiment, the driving motor 2 may be a rotary motor.
In one example, the laser radar device 3 is located at a preset height above a gap area between the screen door 4 and the train door 5. The preset height can be set according to actual requirements. Illustratively, the preset height may range from 2.5 meters to 2.8 meters.
In one example, the lidar device 3 includes a target detection device with a processor that can send control commands to the drive motor 2 according to actual needs. In this way, the driving motor 2 drives the laser radar device 3 to move repeatedly within a certain swing angle range as required according to the control command. In the process, the laser radar device 3 throws the light curtain downwards along the advancing direction of the mobile traffic train so as to obtain the point cloud of the position space where the light curtain is located in real time, and thus target detection is carried out according to the point cloud.
As an example of the present application, the laser radar apparatus 3 includes a single line laser radar to perform target detection by the single line laser radar.
It is worth mentioning that if the single-line laser radar is directly arranged above the gap area between the train door 5 and the shield door 4, the gap area is a narrow space, and some train doors 5 are inward-opening type and some are outward-opening type, if the single-line laser radar is not installed properly, the train door 5 is easily judged mistakenly as an invader in the gap area, and thus the alarm system is triggered mistakenly. This brings serious challenge to the installation and debugging of the single-line laser radar, and the installation and debugging work is time-consuming and labor-consuming. Similarly, if directly lay multi-thread lidar (for example four-wire lidar) in the clearance region's between train door 5 and shield door 4 top, then need make in installation and debugging that one light scanning in four-wire lidar's the both sides light is on the shield door, and another light scanning is on train door, and two middle light are perpendicular downwards, are parallel with the shield door, so also have higher requirements for the installation accuracy. In comparison, the system provided by the application detects the target in the swinging process through the laser radar device, the swinging angle range of the laser radar device is adjustable and controllable, namely the detection range can be adjusted according to the size of the gap area, so that the requirement on the installation pose of the laser radar device is not strict, only the light curtain of the laser radar device 3 is needed to be thrown downwards along the advancing direction of the mobile traffic train, and the problem that whether light scans the door boundary is not strictly required during installation, so that the requirement on the pose of the single-line laser radar during installation is simplified, the installation efficiency is improved, and the installation difficulty is reduced.
In addition, if the multi-line laser radar is used for detection, some light beams may hit the region outside the detection range, thereby causing resource waste. The system provided by the embodiment of the application can effectively detect the detection range through the single-line laser radar, so that the cost can be saved, and unnecessary waste is reduced.
As an example of the present application, please refer to FIG. 2, the system further comprises a soot cleaning device 6, the soot cleaning device 6 comprises a fixing member 61 and a soot cleaning brush 62. The fixing part 61 is located on the outer wall of the swing arm 12, the ash cleaning brush 62 is located on the inner wall of the swing arm 12, one end, not provided with a brush, of the ash cleaning brush 62 penetrates through the side wall of the swing arm 12 and is fixedly connected to the fixing part 61, one end, provided with a brush, of the ash cleaning brush 62 is in contact with the light emitting part of the laser radar device 3, and the fixing part 61 is used for controlling the ash cleaning brush 62 to clean dust for the light emitting part of the laser radar device 3.
In practice, the part of the lidar device 3 with the light exit position rotates along a rotation axis, which may be arranged on the lidar device 3 and perpendicular to the direction of travel of the moving traffic train. In the rotating process, the dust cleaning brush 62 in the dust cleaning device 6 can clean dust for the light emitting part of the laser radar device 3.
Based on the system shown in fig. 1 or fig. 2, the method for object detection provided by the embodiment of the present application will be described in detail below. For ease of understanding, the debug phase is first described, and the main purpose of the debug phase is to determine the target information. The target information includes boundary information of train doors, boundary information of screen doors, and boundary information of gap areas. Referring to fig. 3, fig. 3 is a flow chart illustrating a method for object detection according to an exemplary embodiment, which may be applied to an apparatus for object detection, which may be configured in the lidar apparatus 3 by way of example and not limitation, or may be a separate entity device that establishes communication connections with the lidar apparatus and the driving motor, respectively. The method may include the following:
step 301: and controlling the laser radar device to swing within a preset angle range at a specified speed.
In implementation, the driving motor drives the laser radar device to swing in the rotating process, so that the swinging speed of the laser radar device can be controlled by controlling the rotating speed of the driving motor.
Wherein, the designated speed can be set according to actual requirements. Illustratively, the designated speed may be a minimum rotational speed of the drive motor.
In a similar way, every time the driving motor rotates by an angle, the laser radar device swings by a certain angle under the driving of the driving motor, so that the swing angle range of the laser radar device and the swing angle of the laser radar device at each swing position can be controlled by the rotation angle of the driving motor.
Wherein, predetermine the angle scope and can set up according to actual demand. In one embodiment, the preset angle range may be a maximum swing angle range of the laser radar apparatus. For example, referring to fig. 4, it is assumed that the zero-degree angle of the laser radar apparatus is vertically downward, the side of the zero-degree angle close to the shield door is a negative direction, and the side of the zero-degree angle close to the train door is a positive direction. The maximum swing angle range of the laser radar device is (-45 degrees, 45 degrees).
In general, in order to effectively determine the boundaries of the sections of the screen door, the train door and the gap region in the debugging stage, the laser radar device may be controlled to swing within the maximum swing angle range at the minimum rotation speed, for example, the swing angle range corresponding to the swing start angle and the swing end angle may be (-45 °, 45 °) to obtain the maximum scene scanning range.
It should be noted that the swing angle range of the laser radar apparatus may be controlled according to actual requirements, that is, the swing start angle and the swing end angle of the laser radar apparatus may be set within the maximum swing angle range as required. In one example, if the clearance between the train door and the screen door is small, the swing angle range can be set to be small, for example, the swing angle range corresponding to the swing start angle and the swing end angle can be adjusted to (-5 °, 5 °). In yet another example, if the train door is of an outward opening type, the swing angle range may need to be smaller, for example, the swing angle range corresponding to the swing start angle and the swing end angle may be adjusted to (-5 °, 2 °). In another example, if there are two screen doors for each train door and one system is required to cover the detection ranges of the two screen doors, the swing angle range corresponding to the swing start angle and the swing end angle can be adjusted to (-5 °, 1.2 °).
Step 302: and in the process of swinging the laser radar device, constructing a three-dimensional scene according to second detection data of the laser radar device and second swinging angles of the laser radar device at each swinging position to obtain a second three-dimensional scene.
The second detection data includes ranging values for respective ranging points within each of the plurality of frames, wherein a ranging point may be understood as a point scanned by a light line of the lidar device. Each frame corresponds to a second swing angle.
The second swing angle of the laser radar device at each swing position is an angle between a light curtain plane and a zero angle when the laser radar device is at each swing position.
That is, the laser radar apparatus throws the light curtain continuously during the swing of the laser radar apparatus to perform the target detection in real time. And the driving motor feeds back the second swing angle of the laser radar device at each swing position in real time. In addition, every time the laser radar device swings to a second swing angle, the laser radar device completes one frame of detection, namely, each second swing angle corresponds to one frame of scanning data, so that the laser radar device swings for a detection period, namely, after swinging from the leftmost end to the rightmost end of the angle swing range, the laser radar device correspondingly detects to obtain multi-frame scanning data. For ease of understanding and distinction, the multiple frames of scan data of the debug phase are referred to herein as second probe data.
And constructing a three-dimensional scene according to the second detection data and the second swing angle of the laser radar device at each swing position after each detection period. In one embodiment, a specific implementation of constructing a three-dimensional scene may include: a rectangular coordinate system is established by taking a light emitting point of the laser radar device as the origin of a coordinate system, taking the vertical downward direction as the Z axis, taking the advancing direction of the mobile traffic train as the X axis and taking the direction vertical to the advancing direction of the mobile traffic train as the Y axis. And determining the three-dimensional coordinate information of each ranging point in each frame in the rectangular coordinate system according to the ranging value of each ranging point in each frame, the second swing angle corresponding to each frame and the scanning angle resolution of the laser radar device. And constructing a three-dimensional scene according to the three-dimensional coordinate information of each ranging point in each frame.
That is, in the implementation, coordinate transformation is performed on each ranging point, so that each ranging point is projected to the same rectangular coordinate system, and three-dimensional reconstruction is realized. In one example, each ranging point may be coordinate-converted by the following equation (1):
wherein x isi，jX-axis coordinate value, y, of the ith ranging point in the jth frame of the lidar devicei，jA y-axis coordinate value, z, of the ith ranging point in the jth frame of the lidar devicei，jIs the z-axis coordinate value of the ith ranging point in the jth frame of the lidar device. li，jFor the j frame of the lidar deviceDistance measurement value of the ith distance measurement point, alphaiFor the scanning angle, beta, of the ith range point in each frame of the lidar apparatusjAnd scanning a second swing angle of the laser radar device corresponding to the j frame scanning data of the laser radar device.
It should be noted that, for the single line lidar, the scanning angle α of each ranging point in each scanning isiIs determined in relation to the scanning angle resolution and the range point sequence number. Taking fig. 5 as an example, a frame of scanning data is formed from 0 degree to 180 degrees in each scanning period of the single line laser radar, all light rays emitted in each scanning form a 2D plane, an included angle between two adjacent light rays is 0.1 degree, that is, a scanning angle resolution is 0.1 degree, and then a scanning angle α of each ranging point is obtainedi0.1 ═ i. For example, the scanning angle corresponding to the 0 th light ray is 0 degree, that is, the scanning angle of the ranging point corresponding to the 0 th light ray is 0 degree. For example, the scanning angle of the 360 th light ray is 36 degrees, that is, the scanning angle of the ranging point corresponding to the 360 th light ray is 36 degrees.
It is worth mentioning that in a common multiline laser radar, the track of the light curtain of non-0 degree is a conical surface, as shown in fig. 6, the center of light of the multiline laser radar is at point O, OMNS is the track of the light curtain of 2 degree, the light curtain is a conical surface body, and the intersecting line formed by the light curtain and the ground train door, the screen door ABCD is a curve LKH. For the single-line laser radar, under the action of the driving motor, the light curtain has no conical surface characteristic, as shown in fig. 7, the light curtain under each swing angle is a plane, and the intersecting line of the light curtain and the train door and the shield door is a straight line (such as a straight line L2K 2). Compared with a multi-line laser radar, the single-line laser radar has high scanning point cloud quality under the swing action, and the reconstructed three-dimensional scene is more regular and more beneficial to searching boundary information of each part.
Step 303: and extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the gap area from the second three-dimensional scene to obtain target boundary information.
As an example of the present application, the specific implementation of step 303 may include: and determining a normal vector of each point cloud in the second three-dimensional reconstruction scene. And clustering according to the directionality of the normal vector of each point cloud to obtain a plurality of point cloud sets, wherein the plurality of point cloud sets comprise a point cloud set of a train door, a point cloud set of a shield door and a point cloud set of a gap area. And extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the gap area according to the plurality of point cloud sets to obtain target boundary information.
All ranging points in the second three-dimensional scene constitute a point cloud. In implementation, normal vector calculation is performed according to the three-dimensional coordinate information of each point cloud to obtain a normal vector of each point cloud. The normal vector of the point cloud of the same object has the same directivity, for example, the point cloud of train doors has the same directivity, and the point cloud of shield doors has the same directivity and the point cloud of gap areas has the same directivity. Therefore, clustering processing can be performed according to the directionality of the normal vector of each point cloud, and the obtained multiple point cloud sets comprise a point cloud set of a train door, a point cloud set of a screen door and a point cloud set of a gap area. And then, extracting target boundary information according to the obtained plurality of point cloud sets.
As an example of the present application, the specific implementation of extracting boundary information of train doors, boundary information of screen doors, and boundary information of gap areas from a plurality of point cloud sets may include: and determining the minimum y component value and the maximum z component value in the point cloud set of the train door as the boundary information of the train door. And determining the maximum y component value and the maximum z component value in the point cloud set of the screen door as the boundary information of the screen door. And carrying out random plane sampling on the point cloud set of the gap area, and determining the obtained z-component value of the sampling plane as the boundary information of the gap area.
For example, referring to fig. 8, based on the rectangular coordinate system described above, the barrier door is located on the negative y-axis side of the rectangular coordinate system, the train door is located on the positive y-axis side of the rectangular coordinate system, and the gap area is located in the middle of the rectangular coordinate system. According to the position space distribution characteristics of each part in the rectangular coordinate system, the corresponding boundary information can be extracted from the point cloud set corresponding to each part based on the coordinate information.
In one example, the boundary information of the screen door is extracted from the point cloud set of the screen door by the following formula (2), assuming that the screen doorIs recorded as { y0，z0And then:
in one example, the boundary information of the train door may be extracted from the point cloud set of the train door by the following formula (3), assuming that the boundary information of the train door is noted as { y }1，z1And then:
in one example, the point cloud concentration in the gap area is randomly sampled in a plane, and the resulting sampling plane may be referred to as a bottom plane, and referring to fig. 1, the bottom plane may be understood as a plane where a straight line 00 is located, that is, a platform plane inside the screen door. Determining the z-component value of the sampling plane as the boundary information of the gap area, and recording the boundary information of the gap area as z2。
In one example, after the above steps, the extracted boundary and boundary information are as shown in fig. 9: wherein O is the luminous center of the single-line laser radar, the left rectangular frame is the shielding door part, and the requirements are that
The middle filling rectangle is a gap area, and the conditions are met
The right rectangular frame is a train door part, and the conditions are met
In the embodiment of the application, after a three-dimensional scene is built, boundary information of train doors, a gap area and a shield door is respectively established, so that the swing angle range corresponding to each part can be determined, as shown in fig. 8, a point O is an installation position of a laser radar device, a point angle AOB corresponds to the shield door part, a point angle BOC corresponds to the gap area part, and a point angle COD corresponds to the train door part. Subsequently, according to the boundary information of each part, the scanning data of the gap area is screened out to judge whether the gap area has a target or not, and in addition, different swing speeds can be set in the swing angle ranges corresponding to different parts to improve the detection efficiency. See in particular the examples below.
Based on the data provided by the system shown in fig. 1 or fig. 2 and the embodiment shown in fig. 3, the application process (i.e., the process of object detection) will be described next. Referring to fig. 10, fig. 10 is a flowchart illustrating a method for object detection according to an exemplary embodiment. The method may be applied to an apparatus for object detection, which may be configured in the lidar apparatus 3, or may be a separate entity device that establishes communication connections with the lidar apparatus and the drive motor, respectively. The method may include the following:
step 1001: in the process of target detection of the laser radar device, a three-dimensional scene is constructed according to first detection data of the laser radar device and first swing angles of the laser radar device at all swing positions, and a first three-dimensional scene is obtained.
In one embodiment, the laser radar apparatus detects the same duration of one frame as the laser radar apparatus swings through an angle. Illustratively, assuming that the lidar device takes 20ms to detect a frame, the time period for the lidar device to reach from one angle to another is also 20 ms. In this way, the detection of the lidar means can be synchronized with the oscillation of the lidar means.
As an example of the present application, in an initial state, the laser radar apparatus is controlled to stop at a specified position such that a detection field of view of the laser radar apparatus is a third angular range corresponding to the train door. When the detection value of the laser radar device changes, the state of the train door is detected. And if the train door is switched from the opening state to the closing state, controlling the laser radar device to start swinging, and starting target detection by the laser radar device in the swinging process.
In the initial state, the laser radar device does not swing, namely in the initial state, the position and pose of the laser radar device always face the position of a train door to be detected by controlling the driving motor so as to detect whether a mobile traffic train enters the station or not. As shown in fig. 11, if the mobile traffic train does not arrive at the station, the scanning distance of the laser radar device is long, and the ranging value is large. For example, each scanning point in a single frame forms a straight line as shown in CD in fig. 11. If the mobile traffic train enters the station, the detection value of the laser radar device changes due to the shielding of the mobile traffic train, for example, each scanning point in a single frame forms a straight line as shown by AB in fig. 11, and the distance measurement value becomes smaller, for example, smaller than the distance threshold value, at this time, it can be determined that the mobile traffic train enters the station.
Wherein, the distance threshold value can be set according to actual requirements.
After the train door of the mobile traffic train is determined to enter the station, the state of the train door is continuously detected, as an example of the application, when the state of the train door is detected, a certain specified object can be used as a reference object, and the specified object can be set according to actual requirements. For example, a prescribed object (an object such as a rope) is provided at a position inside the train door, or a fixed object inside the train door is used as the prescribed object. And when the specified object is determined to be detected according to the detection data of the laser radar device, determining that the train door is in an open state, otherwise, determining that the train door is in a closed state if the specified object is not detected.
If the train door is in an open state, the passengers normally get on or off the train after the train door is opened, so that detection is not needed at the moment. When the train door is detected to be switched from the open state to the closed state, in order to avoid the situation that people clamp objects, target detection needs to be carried out on the gap area, therefore, the laser radar device is controlled to start swinging, and in the swinging process, the laser radar device starts to carry out target detection.
And constructing a three-dimensional scene according to the first detection data of the laser radar device and the first swing angles of the laser radar device at each swing position every time a detection period passes, so as to obtain a first three-dimensional scene. In one example, a specific implementation of constructing the first three-dimensional scene may include: a rectangular coordinate system is established by taking a light emitting point of the laser radar device as the origin of a coordinate system, taking the vertical downward direction as the Z axis, taking the advancing direction of the mobile traffic train as the X axis and taking the direction vertical to the advancing direction of the mobile traffic train as the Y axis. And determining three-dimensional coordinate information of each ranging point in each frame in the rectangular coordinate system according to the ranging value of each ranging point in each frame in the first detection data, the first swing angle corresponding to each frame and the scanning angle resolution of the laser radar device. And constructing a first three-dimensional scene according to the three-dimensional coordinate information of each ranging point in each frame. For specific implementation, reference may be made to the implementation process for constructing the second three-dimensional scene in the embodiment in fig. 3, and details are not repeated here.
As an example of the present application, after detecting the state of the train door, a notification message may be sent to the screen door according to the state of the train door, where the notification message is used to open or close the screen door and the train door synchronously.
In some scenes, a communication device does not exist between the shield door and the train door, and in order to ensure that the train door and the shield door are synchronously opened and closed, when the train door is detected to be opened according to detection data of the laser radar device, a notification message can be sent to the shield door, for example, the notification message can be directly sent, or the notification message can be forwarded by means of third-party equipment so as to indicate that the shield door is opened. Similarly, when the train door is detected to be closed according to the detection data of the laser radar device, a notification message can be sent to the shield door to indicate that the shield door is closed. Therefore, the train door and the shield door are opened and closed synchronously. Therefore, the synchronous opening and closing of the train door and the shield door can be realized without increasing a communication channel between the shield door and the train door, and the increase of the transformation cost is avoided.
In another embodiment, if the shielding door and the train door have communication channels at both ends thereof and the ground control end can detect the state of the train door, the laser radar apparatus may not detect the state of the train door but only wait for the ground control end to input the state of the train door. In this case, it is also possible to synchronize the opening and closing of the train door and the screen door without the aid of a laser radar device, for example, via a communication channel.
Step 1002: and determining a point cloud set of the clearance area from the point clouds of the first three-dimensional scene according to target boundary information, wherein the target boundary information is used for indicating the boundary of the train door, the boundary of the screen door and the boundary of the clearance area.
As described above, the target boundary information includes the boundary information of the train door, the boundary information of the screen door, and the boundary information of the gap area, and thus, the point cloud set belonging to the train door, the point cloud set belonging to the screen door, and the point cloud set belonging to the gap area can be determined from the first three-dimensional scene according to the target boundary information. In one embodiment, the point clouds in the first three-dimensional scene can be filtered, and the filtered point cloud set is determined as a point cloud set of the gap area. Or screening out a point cloud set of the gap area from the point cloud of the first three-dimensional scene.
Illustratively, the target boundary information includes { y }0，z0}、{y1，z1}、z2. Then, according to the target boundary information, the pair satisfies
Conditions and
and filtering the point cloud set of the conditions, wherein the reserved point cloud set is the point cloud set of the gap area. Or, according to the target boundary information, screening out the point cloud in the first three-dimensional scene to satisfy the requirement
And (4) carrying out point cloud collection on the conditions to obtain a point cloud collection of the gap area.
Step 1003: and if the target exists in the gap area according to the point cloud set of the gap area, carrying out early warning prompt.
As an example of the present application, determining a specific implementation of the gap region existence target according to the point cloud of the gap region may include: clustering the point cloud sets in the gap area to obtain at least one point cloud block, and determining the three-dimensional size of each point cloud block in the at least one point cloud block, wherein the three-dimensional size comprises length, width and height. If a corresponding point cloud block with a three-dimensional size meeting the target size condition exists in at least one point cloud block, determining that a target exists in the gap area, wherein the fact that the three-dimensional size meets the target size condition means that: a length in the three-dimensional dimension is greater than or equal to a length dimension threshold, and/or a width in the three-dimensional dimension is greater than or equal to a width dimension threshold, and/or a height in the three-dimensional dimension is greater than or equal to a height dimension threshold.
Wherein, length dimension threshold, width dimension threshold, height dimension threshold all can set up according to actual demand.
Each point cloud block in at least one point cloud block corresponds to an object, and if the three-dimensional size of a certain point cloud block in at least one point cloud block meets the target size condition, it can be shown that the object corresponding to the point cloud block is an invader, such as a luggage, a person, a child anti-loss rope, and the like, and it can also be determined that a target exists in the gap area.
If the target exists in the gap area determined according to the point cloud set of the gap area, that is, if the target exists in the gap area after the train door is closed, the condition of people clamping or object clamping can occur, and in order to avoid the condition, early warning prompt can be performed, for example, early warning prompt can be performed through ringing, voice broadcast and other modes. In one embodiment, after the warning prompt, the screen door can be controlled by the screen door control device, such as to reopen the screen door to prevent people from clamping objects.
After the moving traffic train leaves the station, the laser radar device is controlled to be in an initial state, for example, the position and posture of the laser radar device are always opposite to the third angle range of the train door again by controlling the driving motor.
As an example of the present application, a first angle range corresponding to the screen door, a second angle range corresponding to the gap area, and a third angle range corresponding to the train door may be determined according to the target boundary information. And controlling the laser radar device to swing at a first preset speed in a first angle range, at a second preset speed in a second angle range and at a third preset speed in a third angle range, wherein the first preset speed and the third preset speed are both greater than the second preset speed.
The laser radar device has the maximum swing speed, and the first preset speed, the second preset speed and the third preset speed are all less than or equal to the maximum swing speed and can be specifically set according to actual requirements.
That is, the real-time swing speed of the lidar apparatus may be configured as desired, such as when a higher detection rate for smaller objects is desired, the swing speed may be set to be smaller. Higher detection rates for smaller targets, such as child-resistant lanyards, are often required in the gap area. Also generally need carry out accurate survey to the target in the clearance region to avoid lou examining, for this reason, in the second angular range that the clearance region corresponds, can adopt slower swing speed, also the second is preset speed and is set up less, so that laser radar device catches the more cloud in clearance region, thereby promote the some cloud details of clearance region invading thing, and then improve the rate of accuracy of surveying.
It should be noted that the preset speed corresponding to each swing angle range can be configured to the driving motor in advance, so that the driving motor rotates according to the configuration in the rotating process, and the laser radar device is driven to swing at different preset speeds in different swing angle ranges.
It is worth mentioning that the swing angle range and the swing speed of the laser radar device are configurable, so that the laser radar device can flexibly adapt to different detection ranges, thereby being suitable for various different gap areas and train doors which are inwards sunken and outwards opened, and improving the applicability.
It should be noted that, the above description is only given by taking an example that the first preset speed and the third preset speed are both greater than the second preset speed. In another embodiment, the first preset speed, the second preset speed, and the third preset speed may be the same, which is not limited in this application.
In the embodiment of the application, target detection is performed based on the provided system, and in the detection process of the laser radar device, a three-dimensional scene is constructed according to first detection data of the laser radar device and first swing angles of the laser radar device at various swing positions, so that a first three-dimensional scene is obtained. And determining a point cloud set of the gap area from the point clouds of the first three-dimensional scene according to target boundary information, wherein the target boundary information is used for indicating the boundary of the train door, the boundary of the screen door and the boundary of the gap area. If the target exists in the gap area according to the point cloud set of the gap area, it is indicated that the gap area has the invader, and therefore early warning prompt is carried out. Also this system passes through the laser radar device and realizes target detection, compares in through preventing pressing from both sides the baffle, and this system's detection validity is higher, can avoid as far as possible missing the detection.
In addition, the laser radar device is used for swing detection, boundaries can be flexibly determined in any swing angle range, so that a point cloud set of a gap area can be screened out according to the boundaries, then target detection is carried out on the gap area, and therefore the applicability is improved.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
Based on the method for object detection provided in the foregoing embodiments, fig. 12 is a schematic structural diagram illustrating an object detection apparatus according to an exemplary embodiment, which may be implemented by software, hardware, or a combination of the two as part of a lidar apparatus. The apparatus may include: comprising a memory 1210, a processor 1220, and a computer program 1230 stored in said memory 1210 and executable on said processor 1220, said processor when executing said computer program for performing:
in the process of target detection of the laser radar device, constructing a three-dimensional scene according to first detection data of the laser radar device and first swing angles of the laser radar device at various swing positions to obtain a first three-dimensional scene;
determining a point cloud set of the clearance area from the point clouds of the first three-dimensional scene according to target boundary information, wherein the target boundary information is used for indicating the boundary of the train door, the boundary of the screen door and the boundary of the clearance area;
and if the gap area has the target according to the point cloud set of the gap area, carrying out early warning prompt.
As an example of the present application, the processor 1220 is configured to:
clustering the point cloud sets in the gap area to obtain at least one point cloud block;
determining three-dimensional dimensions of each of the at least one point cloud block, the three-dimensional dimensions including a length, a width, and a height;
if a corresponding point cloud block with a three-dimensional size meeting a target size condition exists in the at least one point cloud block, determining that the target exists in the gap area, wherein the fact that the three-dimensional size meets the target size condition means that: a length of the three-dimensional dimension is greater than or equal to a length dimension threshold, and/or a width of the three-dimensional dimension is greater than or equal to a width dimension threshold, and/or a height of the three-dimensional dimension is greater than or equal to a height dimension threshold.
As an example of the present application, the processor 1220 is configured to:
controlling the laser radar device to swing within a preset angle range at a specified speed;
in the process of swinging of the laser radar device, constructing a three-dimensional scene according to second detection data of the laser radar device and second swinging angles of the laser radar device at each swinging position to obtain a second three-dimensional scene;
and extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the gap area from the second three-dimensional scene to obtain the target boundary information.
As an example of the present application, the second detection data includes ranging values of respective ranging points in each frame of a plurality of frames, each frame corresponding to a second swing angle, and the processor 1220 is configured to:
establishing a rectangular coordinate system by taking a light emitting point of the laser radar device as an origin of a coordinate system, taking a vertical downward direction as a Z axis, taking a traveling direction of the mobile traffic train as an X axis, and taking a direction perpendicular to the traveling direction of the mobile traffic train as a Y axis;
determining three-dimensional coordinate information of each ranging point in each frame in the rectangular coordinate system according to the ranging value of each ranging point in each frame, the second swing angle corresponding to each frame and the scanning angle resolution of the laser radar device;
and constructing a three-dimensional scene according to the three-dimensional coordinate information of each ranging point in each frame.
As an example of the present application, the processor 1220 is configured to:
determining a normal vector of each point cloud in the second three-dimensional reconstruction scene;
clustering according to the directionality of the normal vector of each point cloud to obtain a plurality of point cloud sets, wherein the plurality of point cloud sets comprise the point cloud set of the train door, the point cloud set of the shield door and the point cloud set of the gap area;
and extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the clearance area according to the plurality of point cloud sets to obtain the target boundary information.
As an example of the present application, the processor 1220 is configured to:
determining the minimum y component value and the maximum z component value in the point cloud set of the train door as the boundary information of the train door;
determining the maximum y component value and the maximum z component value in the point cloud set of the screen door as the boundary information of the screen door;
and carrying out random plane sampling on the point cloud set of the gap area, and determining an obtained z-component value of a sampling plane as boundary information of the gap area.
As an example of the present application, the processor 1220 is further configured to:
according to the target boundary information, respectively determining a first angle range corresponding to the shielding door, a second angle range corresponding to the gap area and a third angle range corresponding to the train door;
control the lidar device is in with first predetermined speed swing in the first angle scope with the swing of second predetermined speed in the second angle scope, and the swing of third predetermined speed in the third angle scope, first predetermined speed with the third predetermined speed all is greater than the second predetermined speed.
As an example of the present application, the processor 1220 is further configured to:
in an initial state, controlling the laser radar device to stop at a specified position, wherein the specified position enables a detection view field of the laser radar device to be a third angle range corresponding to the train door;
detecting the state of the train door when the detection value of the laser radar device changes;
and if the train door is switched from the opening state to the closing state, controlling the laser radar device to start swinging, and starting target detection by the laser radar device in the swinging process.
As an example of the present application, the processor 1220 is further configured to:
and sending a notification message to the screen door according to the state of the train door, wherein the notification message is used for enabling the screen door and the train door to be synchronously opened or closed.
In the embodiment of the application, target detection is performed based on the provided system, and in the detection process of the laser radar device, a three-dimensional scene is constructed according to first detection data of the laser radar device and first swing angles of the laser radar device at various swing positions, so that a first three-dimensional scene is obtained. And determining a point cloud set of the gap area from the point clouds of the first three-dimensional scene according to target boundary information, wherein the target boundary information is used for indicating the boundary of the train door, the boundary of the screen door and the boundary of the gap area. If the target exists in the gap area according to the point cloud set of the gap area, it is indicated that the gap area has the invader, and therefore early warning prompt is carried out. Also this system passes through the laser radar device and realizes target detection, compares in through preventing pressing from both sides the baffle, and this system's detection validity is higher, can avoid as far as possible missing the detection.
It should be noted that, for the information interaction, execution process, and other contents between the above-mentioned devices/units, the specific functions and technical effects thereof are based on the same concept as those of the embodiment of the method of the present application, and specific reference may be made to the part of the embodiment of the method, which is not described herein again.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (12)
1. A system for object detection, the system comprising: the laser radar device comprises a mounting bracket, a driving motor and a laser radar device, wherein the mounting bracket comprises a fixed seat and a swing arm;
the fixed seat is arranged at the designated position of the top end of the inner side of the shielding door on the train platform, and the inner side is close to one side of the track of the mobile traffic train;
the driving motor is positioned on the fixed seat;
the rotating shaft of the driving motor is fixedly connected with the swinging arm so as to drive the swinging arm to do circular arc swinging in a gap area between the shielding door and the train door during rotation, wherein the tangential direction of a circular arc track of the circular arc swinging is vertical to the traveling direction of the moving traffic train;
the laser radar device is located on the inner wall of the swing arm and used for detecting a target in the swing process of the swing arm.
2. A method of target detection implemented based on the system of claim 1, wherein the lidar means projects a light curtain downwardly along the direction of travel, the method comprising:
in the process of target detection of the laser radar device, constructing a three-dimensional scene according to first detection data of the laser radar device and first swing angles of the laser radar device at various swing positions to obtain a first three-dimensional scene;
determining a point cloud set of the clearance area from the point clouds of the first three-dimensional scene according to target boundary information, wherein the target boundary information is used for indicating the boundary of the train door, the boundary of the screen door and the boundary of the clearance area;
and if the gap area has the target according to the point cloud set of the gap area, carrying out early warning prompt.
3. The method of claim 2, wherein determining the gap region presence target from the point cloud set of gap regions comprises:
clustering the point cloud sets in the gap area to obtain at least one point cloud block;
determining three-dimensional dimensions of each of the at least one point cloud block, the three-dimensional dimensions including a length, a width, and a height;
if a corresponding point cloud block with a three-dimensional size meeting a target size condition exists in the at least one point cloud block, determining that the target exists in the gap area, wherein the fact that the three-dimensional size meets the target size condition means that: a length of the three-dimensional dimension is greater than or equal to a length dimension threshold, and/or a width of the three-dimensional dimension is greater than or equal to a width dimension threshold, and/or a height of the three-dimensional dimension is greater than or equal to a height dimension threshold.
4. The method of claim 2, wherein the target boundary information is determined in a manner comprising:
controlling the laser radar device to swing within a preset angle range at a specified speed;
in the process of swinging of the laser radar device, constructing a three-dimensional scene according to second detection data of the laser radar device and second swinging angles of the laser radar device at each swinging position to obtain a second three-dimensional scene;
and extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the gap area from the second three-dimensional scene to obtain the target boundary information.
5. The method of claim 4, wherein the second detection data comprises ranging values for respective ranging points in each of a plurality of frames, each frame corresponding to a second pan angle, and wherein constructing the three-dimensional scene based on the second detection data for the lidar device and the second pan angle for the lidar device at the respective pan positions comprises:
establishing a rectangular coordinate system by taking a light emitting point of the laser radar device as an origin of a coordinate system, taking a vertical downward direction as a Z axis, taking a traveling direction of the mobile traffic train as an X axis, and taking a direction perpendicular to the traveling direction of the mobile traffic train as a Y axis;
determining three-dimensional coordinate information of each ranging point in each frame in the rectangular coordinate system according to the ranging value of each ranging point in each frame, the second swing angle corresponding to each frame and the scanning angle resolution of the laser radar device;
and constructing a three-dimensional scene according to the three-dimensional coordinate information of each ranging point in each frame.
6. The method of claim 5, wherein the extracting boundary information of the train door, boundary information of the screen door, and boundary information of the gap area from the second three-dimensional scene to obtain the target boundary information comprises:
determining a normal vector of each point cloud in the second three-dimensional reconstruction scene;
clustering according to the directionality of the normal vector of each point cloud to obtain a plurality of point cloud sets, wherein the plurality of point cloud sets comprise the point cloud set of the train door, the point cloud set of the shield door and the point cloud set of the gap area;
and extracting the boundary information of the train door, the boundary information of the shield door and the boundary information of the clearance area according to the plurality of point cloud sets to obtain the target boundary information.
7. The method of claim 6, wherein extracting boundary information of the train door, boundary information of the screen door, and boundary information of the gap area from the plurality of point clouds to obtain the target boundary information comprises:
determining the minimum y component value and the maximum z component value in the point cloud set of the train door as the boundary information of the train door;
determining the maximum y component value and the maximum z component value in the point cloud set of the screen door as the boundary information of the screen door;
and carrying out random plane sampling on the point cloud set of the gap area, and determining an obtained z-component value of a sampling plane as boundary information of the gap area.
8. The method of claim 2, further comprising:
according to the target boundary information, respectively determining a first angle range corresponding to the shielding door, a second angle range corresponding to the gap area and a third angle range corresponding to the train door;
control the lidar device is in with first predetermined speed swing in the first angle scope with the swing of second predetermined speed in the second angle scope, and the swing of third predetermined speed in the third angle scope, first predetermined speed with the third predetermined speed all is greater than the second predetermined speed.
9. The method of claim 2, further comprising:
in an initial state, controlling the laser radar device to stop at a specified position, wherein the specified position enables a detection view field of the laser radar device to be a third angle range corresponding to the train door;
detecting the state of the train door when the detection value of the laser radar device changes;
and if the train door is switched from the opening state to the closing state, controlling the laser radar device to start swinging, and starting target detection by the laser radar device in the swinging process.
10. The method of claim 9, wherein after detecting the status of the train door, further comprising:
and sending a notification message to the screen door according to the state of the train door, wherein the notification message is used for enabling the screen door and the train door to be synchronously opened or closed.
11. An apparatus for object detection comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 2 to 10 when executing the computer program.
12. A computer readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the method of any of claims 2 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111472456.4A CN114312853B (en) | 2021-12-03 | 2021-12-03 | System, method, device and storage medium for target detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111472456.4A CN114312853B (en) | 2021-12-03 | 2021-12-03 | System, method, device and storage medium for target detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114312853A true CN114312853A (en) | 2022-04-12 |
| CN114312853B CN114312853B (en) | 2023-11-14 |
Family
ID=81049305
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111472456.4A Active CN114312853B (en) | 2021-12-03 | 2021-12-03 | System, method, device and storage medium for target detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114312853B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117465483A (en) * | 2023-12-27 | 2024-01-30 | 江苏明伟万盛科技有限公司 | Visual safety detection device for gap of rail transit platform |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015186948A1 (en) * | 2014-06-02 | 2015-12-10 | 한국교통연구원 | Method for sensing objects around platform screen doors |
| JP2017019380A (en) * | 2015-07-09 | 2017-01-26 | Ｊｒ東日本メカトロニクス株式会社 | Method, device, system and program for generating detection area |
| CN106864463A (en) * | 2017-01-26 | 2017-06-20 | 广州地铁设计研究院有限公司 | Overhead type foreign matter detection system between a kind of subway platform door and car door |
| JP2018131071A (en) * | 2017-02-15 | 2018-08-23 | 日本信号株式会社 | System and method for object detection |
| CN108490447A (en) * | 2018-03-21 | 2018-09-04 | 交控科技股份有限公司 | The anti-clipping system and method for gate and train |
| CN109709564A (en) * | 2018-12-05 | 2019-05-03 | 交控科技股份有限公司 | A kind of shield door anti-clipping system and method based on the detection of laser radar single line |
| CN110217244A (en) * | 2019-05-10 | 2019-09-10 | 森思泰克河北科技有限公司 | Control method, control device and the radar of shield door |
| CN110481601A (en) * | 2019-09-04 | 2019-11-22 | 深圳市镭神智能系统有限公司 | A kind of rail detection system |
| CN111815776A (en) * | 2020-02-04 | 2020-10-23 | 山东水利技师学院 | Three-dimensional building fine geometric reconstruction method integrating airborne and vehicle-mounted three-dimensional laser point clouds and streetscape images |
| CN111986115A (en) * | 2020-08-22 | 2020-11-24 | 王程 | Accurate elimination method for laser point cloud noise and redundant data |
| CN112343448A (en) * | 2019-08-09 | 2021-02-09 | 北醒（北京）光子科技有限公司 | Shielding door protection system based on 3D laser radar |
| CN113311443A (en) * | 2020-02-26 | 2021-08-27 | 保定市天河电子技术有限公司 | Anti-pinch detection method and system for bus shielding door |
-
2021
- 2021-12-03 CN CN202111472456.4A patent/CN114312853B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015186948A1 (en) * | 2014-06-02 | 2015-12-10 | 한국교통연구원 | Method for sensing objects around platform screen doors |
| JP2017019380A (en) * | 2015-07-09 | 2017-01-26 | Ｊｒ東日本メカトロニクス株式会社 | Method, device, system and program for generating detection area |
| CN106864463A (en) * | 2017-01-26 | 2017-06-20 | 广州地铁设计研究院有限公司 | Overhead type foreign matter detection system between a kind of subway platform door and car door |
| JP2018131071A (en) * | 2017-02-15 | 2018-08-23 | 日本信号株式会社 | System and method for object detection |
| CN108490447A (en) * | 2018-03-21 | 2018-09-04 | 交控科技股份有限公司 | The anti-clipping system and method for gate and train |
| CN109709564A (en) * | 2018-12-05 | 2019-05-03 | 交控科技股份有限公司 | A kind of shield door anti-clipping system and method based on the detection of laser radar single line |
| CN110217244A (en) * | 2019-05-10 | 2019-09-10 | 森思泰克河北科技有限公司 | Control method, control device and the radar of shield door |
| CN112343448A (en) * | 2019-08-09 | 2021-02-09 | 北醒（北京）光子科技有限公司 | Shielding door protection system based on 3D laser radar |
| CN110481601A (en) * | 2019-09-04 | 2019-11-22 | 深圳市镭神智能系统有限公司 | A kind of rail detection system |
| CN111815776A (en) * | 2020-02-04 | 2020-10-23 | 山东水利技师学院 | Three-dimensional building fine geometric reconstruction method integrating airborne and vehicle-mounted three-dimensional laser point clouds and streetscape images |
| CN113311443A (en) * | 2020-02-26 | 2021-08-27 | 保定市天河电子技术有限公司 | Anti-pinch detection method and system for bus shielding door |
| CN111986115A (en) * | 2020-08-22 | 2020-11-24 | 王程 | Accurate elimination method for laser point cloud noise and redundant data |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117465483A (en) * | 2023-12-27 | 2024-01-30 | 江苏明伟万盛科技有限公司 | Visual safety detection device for gap of rail transit platform |
| CN117465483B (en) * | 2023-12-27 | 2024-04-09 | 江苏明伟万盛科技有限公司 | Visual safety detection device for gap of rail transit platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114312853B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11072354B2 (en) | Anti-pinch system and method for platform screen door and train | |
| CN107256019B (en) | A kind of paths planning method of clean robot | |
| CN104408932B (en) | A kind of drunk driving vehicle detecting system based on video monitoring | |
| CN208225167U (en) | Testimony of a witness ticket veritifies gate automatically | |
| CN114312853A (en) | System, method, apparatus and storage medium for object detection | |
| US20210225091A1 (en) | Auto-calibrating range sensing gate | |
| CN103400422B (en) | Type free stream Electronic Toll Collection for Lane System and method | |
| CN107355161A (en) | Safety guard for all-high shield door | |
| CN109709564A (en) | A kind of shield door anti-clipping system and method based on the detection of laser radar single line | |
| CN112193957A (en) | Elevator door detection method based on TOF light field equipment | |
| JP2021127251A (en) | User detection system of elevator | |
| JPH11203589A (en) | Traffic image pickup device and traffic monitoring device | |
| CN112343448A (en) | Shielding door protection system based on 3D laser radar | |
| JP6031908B2 (en) | Image processing apparatus, object detection method, and object detection program | |
| CN208076713U (en) | Parking systems and its people's vehicle are strayed into detecting system | |
| CN214823230U (en) | Platform door safety inspection device based on multi-line detection formula laser radar | |
| CN110874551A (en) | Image processing method for access control gate passage logic control | |
| JP2009288055A (en) | Method of calculating position information of object | |
| CN103481911A (en) | Rim tread image collecting system and train wheel anomaly detection system | |
| CN207908967U (en) | A kind of laser radar formula auxiliary AGV uses lift appliance | |
| US3946578A (en) | Detection of defective knitting needles of a knitting machine | |
| JP7108115B2 (en) | Vehicle door open/close detector | |
| CN107881951B (en) | A kind of freeway tunnel prevents automobile second accident control system and method | |
| JP2013052738A (en) | Detector for rushing-into-train | |
| CN114940428B (en) | Elevator protection method and device based on TOF image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |