CN116540259A

CN116540259A - Three-dimensional point cloud data acquisition device, crop phenotype acquisition method and device

Info

Publication number: CN116540259A
Application number: CN202310821096.7A
Authority: CN
Inventors: 郭新宇; 蔡双泽; 苟文博; 樊江川; 温维亮; 王传宇
Original assignee: Research Center of Information Technology of Beijing Academy of Agriculture and Forestry Sciences
Current assignee: Research Center of Information Technology of Beijing Academy of Agriculture and Forestry Sciences
Priority date: 2023-07-06
Filing date: 2023-07-06
Publication date: 2023-08-04
Anticipated expiration: 2043-07-06
Also published as: CN116540259B

Abstract

The invention provides a three-dimensional point cloud data acquisition device, a crop phenotype acquisition method and a crop phenotype acquisition device, which relate to the technical field of intelligent agriculture, and comprise the following steps: running gear, bearing structure, first slide rail, first slider, radar sensor, controller and communication module. According to the three-dimensional point cloud data acquisition device, the crop phenotype acquisition method and the device, the three-dimensional point cloud data acquisition device moves at fixed points, the radar sensor in the three-dimensional point cloud data acquisition device can acquire three-dimensional point cloud data of crops at multiple positions and multiple angles, the radar sensor is in a static state in the process of data acquisition, vibration and jolt generated when the three-dimensional point cloud data acquisition device moves can be prevented from affecting the accuracy of the three-dimensional point cloud data acquired by the radar sensor, and the three-dimensional point cloud data of crops of different types or crops of the same type in different growth periods can be acquired more flexibly and accurately.

Description

Three-dimensional point cloud data acquisition device, crop phenotype acquisition method and device

Technical Field

The invention relates to the technical field of intelligent agriculture, in particular to a three-dimensional point cloud data acquisition device, a crop phenotype acquisition method and a crop phenotype acquisition device.

Background

Typically, different kinds of crops or crops of the same kind with different growth periods have different growth states.

However, it is difficult to flexibly and accurately acquire three-dimensional point cloud data of different kinds of crops or crops of the same kind and different growth periods in the prior art.

Therefore, how to obtain three-dimensional point cloud data of different kinds of crops or crops of the same kind and different growth periods more flexibly and accurately is a technical problem to be solved in the field.

Disclosure of Invention

The invention provides a three-dimensional point cloud data acquisition device, a crop phenotype acquisition method and a crop phenotype acquisition device, which are used for solving the defect that the three-dimensional point cloud data of different types of crops or crops with the same type and different growth periods are difficult to flexibly and accurately acquire in the prior art, and realizing more flexible and more accurate acquisition of the three-dimensional point cloud data of the different types of crops or the crops with the same type and different growth periods.

The invention provides a three-dimensional point cloud data acquisition device, which comprises: the device comprises a travelling mechanism, a supporting structure, a first sliding rail, a first sliding block, a radar sensor, a controller and a communication module;

the travelling mechanism is connected with the supporting structure; the controller is arranged on the supporting structure; the travelling mechanism and the first sliding block are respectively and electrically connected with the controller; the controller is in communication connection with the user terminal through the communication module;

The first sliding rail is arranged on the supporting structure perpendicular to a first direction; the first sliding block is movably arranged on the first sliding rail along the extending direction of the first sliding rail;

the radar sensor is arranged on the first sliding block towards the vertical downward direction; the radar sensor is in communication connection with the controller and the communication module; the radar sensor is used for responding to the control of the controller, collecting point cloud data or stopping collecting point cloud data, and sending the collected original point cloud data to the user terminal through the communication module;

the controller is used for determining that the support structure reaches the target areaiUnder the condition of walking position, the first sliding block is controlled to drive the radar sensor to slide from one end of the first sliding rail to the other end at a constant speed, and the radar sensor is controlled to collect point cloud data in the process that the first sliding block slides from one end of the first sliding rail to the other end at a constant speed,

the controller is also used for controlling the travelling mechanism to drive the supporting structure to move from the first sliding rail under the condition that the first sliding block is determined to slide from one end of the first sliding rail to the other end at a constant speed iThe walking position moves to the first position in the target areai+1 walking position, the firsti+1 walking position is located at the first positioniThe walking position is at a preset distance in said first direction,isequentially taking 0,1,2 and …,I，Ithe 0 th walking position is the initial walking position of the target area and is a positive integer greater than 1IThe walking position is the termination walking position of the target area.

The three-dimensional point cloud data acquisition device provided by the invention further comprises: the second sliding rail and the second sliding block;

the second sliding rail is perpendicular to the first sliding rail and is arranged on the supporting structure; the second sliding block is movably arranged on the second sliding rail along the extending direction of the second sliding rail;

the second sliding block is connected with the first sliding rail; the second sliding block is in communication connection with the controller; the second sliding block is used for responding to the control of the controller and driving the first sliding rail to slide along the second sliding rail from the preset stay position where the first sliding rail is currently positioned to the other preset stay position;

the controller is further configured to, upon determining that the support structure reaches the first positioniUnder the condition that the first sliding rail is positioned at any preset stay position at the walking position, the first sliding block is controlled to drive the radar sensor to slide from one end of the first sliding rail to the other end at a constant speed, the radar sensor is controlled to acquire point cloud data in the process of sliding from one end of the first sliding rail to the other end at a constant speed,

The controller is also used for controlling the second sliding block to drive the preset stay position where the first sliding rail is currently positioned to slide to another preset stay position under the condition that the first sliding block is determined to slide from one end of the first sliding rail to the other end at a constant speed,

the controller is further configured to control the running mechanism to drive the support structure to move from a first position to a second position when it is determined that the first slide rail is located at each preset stop position and the first slide block has slid from one end of the first slide rail to the other end at a constant speediThe walking position moves to the firsti+1 walking position.

According to the three-dimensional point cloud data acquisition device provided by the invention, the number of the second sliding rails and the second sliding blocks is 2; the two second sliding rails are arranged in parallel at intervals; each second sliding block is movably arranged on each second sliding rail along the extending direction of each second sliding rail;

one end of the first sliding rail is arranged on one second sliding block, and the other end of the first sliding rail is arranged on the other second sliding block.

According to the three-dimensional point cloud data acquisition device provided by the invention, the travelling mechanism comprises four telescopic brackets, four wheels, four travelling motors and four steering motors; each telescopic bracket, each traveling motor and each steering motor are electrically connected with the controller;

Any wheel is rotationally connected with one end of a telescopic bracket;

the other end of any telescopic bracket is rotationally connected with the supporting structure;

any walking motor is electrically connected with one wheel, and is used for responding to the control of the controller to drive the wheel to rotate;

any steering motor is electrically connected with one telescopic bracket and is used for responding to the control of the controller to drive the telescopic bracket to rotate;

each of the telescoping supports is further adapted to extend or retract in response to control by the controller.

The three-dimensional point cloud data acquisition device provided by the invention further comprises: positioning equipment; the positioning device is arranged on the supporting structure; the positioning device is electrically connected with the controller;

the positioning device is used for acquiring real-time position information of the supporting structure and sending the real-time position information to the controller;

the controller is further configured to receive the real-time position information and determine whether the support structure reaches the first position based on the real-time position informationiAnd the walking position and the walking mechanism are controlled.

The three-dimensional point cloud data acquisition device provided by the invention further comprises: the user terminal.

According to the three-dimensional point cloud data acquisition device provided by the invention, the controller is further used for determining that the supporting structure is positioned at the first positionIAnd under the condition that the first sliding block at the walking position slides from one end of the first sliding rail to the other end at a constant speed, sending target information to the user terminal through the communication module, so that the user terminal can acquire the three-dimensional point cloud block corresponding to the target area based on all received original point cloud data under the condition that the user terminal receives the target information.

According to the three-dimensional point cloud data acquisition device provided by the invention, the user terminal acquires the three-dimensional point cloud block corresponding to the target area based on all received original point cloud data, and the three-dimensional point cloud data acquisition device comprises the following steps:

based on a preset speed and the preset distance, carrying out coordinate system conversion on all original point cloud data to obtain three-dimensional point cloud blocks corresponding to each running position, wherein the speed of the first sliding block, which slides from one end of the first sliding rail to the other end at a constant speed, is the preset speed;

after coarse registration is carried out on the three-dimensional point cloud blocks corresponding to each walking position, sequentially carrying out fine registration on the three-dimensional point cloud blocks corresponding to each two adjacent walking positions on the basis of an iterative nearest point algorithm according to the arrangement sequence of each walking position in the first direction, and obtaining the three-dimensional point cloud blocks corresponding to each walking position after registration;

And sequentially splicing the three-dimensional point cloud blocks corresponding to the registered walking positions according to the arrangement sequence of the walking positions in the first direction to obtain the three-dimensional point cloud blocks corresponding to the target area.

According to the three-dimensional point cloud data acquisition device provided by the invention, the user terminal is further used for acquiring the phenotype parameters of crops in the target area based on the three-dimensional point cloud block corresponding to the target area.

The invention also provides a crop phenotype acquisition method realized based on the three-dimensional point cloud data acquisition device, which comprises the following steps:

receiving original point cloud data;

under the condition that target information is received, acquiring a three-dimensional point cloud block corresponding to a target area based on all received original point cloud data;

and acquiring the phenotype parameters of crops in the target area based on the three-dimensional point cloud block corresponding to the target area.

According to the crop phenotype obtaining method provided by the invention, the three-dimensional point cloud block corresponding to the target area is obtained based on all received original point cloud data, and the method comprises the following steps:

acquiring a preset speed and a preset distance;

based on the preset speed and the preset distance, performing coordinate system conversion on all original point cloud data to obtain three-dimensional point cloud blocks corresponding to each travelling position;

The invention also provides a crop phenotype acquisition device, which comprises:

the data acquisition module is used for receiving the original point cloud data;

the point cloud processing module is used for acquiring a three-dimensional point cloud block corresponding to the target area based on all received original point cloud data under the condition of receiving the target information;

the phenotype acquisition module is used for acquiring phenotype parameters of crops in the target area based on the three-dimensional point cloud block corresponding to the target area.

The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the crop phenotype acquisition method as described in any one of the above when executing the program.

The invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which when executed by a processor implements a crop phenotype acquisition method as described in any of the above.

The invention also provides a computer program product comprising a computer program which when executed by a processor implements a crop phenotype acquisition method as described in any one of the above.

According to the three-dimensional point cloud data acquisition device, the crop phenotype acquisition method and the device, the three-dimensional point cloud data acquisition device moves at fixed points, the radar sensor in the three-dimensional point cloud data acquisition device can slide and acquire three-dimensional point cloud data in the direction perpendicular to the advancing direction of the three-dimensional point cloud data acquisition device, the three-dimensional point cloud data of crops can be acquired at multiple positions and multiple angles, the three-dimensional point cloud data acquisition device is in a static state in the process of three-dimensional point cloud data acquisition by the radar sensor, vibration and jolt of the three-dimensional point cloud data acquisition device during movement can be avoided, the accuracy of the three-dimensional point cloud data acquired by the radar sensor is influenced, the three-dimensional point cloud data of different types of crops or crops with different growth periods of the same type can be acquired more flexibly and accurately, data support can be provided for research and application such as crop growth phase monitoring and diagnosis, genetic breeding assistance and crop precision management, and the three-dimensional point cloud data acquisition device in the embodiment of the invention is suitable for greenhouse and large crops, and has the advantages of small volume, good moving, flexibility, high cross country and the like.

Drawings

In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic structural diagram of a three-dimensional point cloud data acquisition device provided by the invention;

FIG. 2 is a second schematic structural diagram of the three-dimensional point cloud data acquisition device according to the present invention;

FIG. 3 is a walking schematic diagram of the three-dimensional point cloud data acquisition device provided by the invention;

FIG. 4 is a schematic flow chart of a crop phenotype acquisition method provided by the invention;

FIG. 5 is a second flow chart of the crop phenotype acquisition method according to the present invention;

FIG. 6 is a schematic diagram of a crop phenotype acquisition apparatus provided by the invention;

fig. 7 is a schematic structural diagram of an electronic device provided by the present invention.

Reference numerals:

1: a support structure; 2: a walking mechanism; 3: a walking motor; 4: a wheel; 5: a second slide rail; 6: a second slider; 7: a radar sensor; 8: a power module; 9: positioning equipment; 10: a communication module; 11: a controller; 12: a first slider; 13: a steering motor; 14: a first slide rail; 15: and a telescopic bracket.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In the description of the invention, it should be noted that, unless explicitly stated and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.

It should be noted that the crop phenotype acquisition device is a device capable of intelligently acquiring crop phenotypes. The crop phenotype acquisition device commonly used in the greenhouse at present comprises: orbital crop phenotype acquisition device, mobile crop phenotype acquisition device, fixed point crop phenotype acquisition device, portable crop phenotype acquisition device, microscopic crop phenotype acquisition device, and the like.

The mobile crop phenotype acquisition device has received great attention in crop breeding and agricultural research due to the advantages of high flexibility, low cost and the like. How to use the movable crop phenotype acquisition device to acquire the crop phenotype rapidly, accurately and efficiently has important significance for researches and applications such as crop growth phase monitoring and diagnosis, genetic breeding assistance and screening, crop precise management and the like.

In the related art, a sensor that can be used to acquire three-dimensional information of a crop includes: binocular cameras, depth cameras, lidar, multi-view cameras, and the like. The three-dimensional information of the crops collected by the sensor can be used for obtaining the phenotypes of the crops based on principles of binocular stereoscopic vision, structured light, time of Flight (ToF), multi-view stereoscopic reconstruction (MVS) and the like. Therefore, the sensor is widely applied to a three-dimensional point cloud data acquisition device.

Typically, different kinds of crops or crops of the same kind with different growth periods have different growth states. However, the position of the sensor in the traditional three-dimensional information acquisition device is usually fixed, the visual angle is single when the traditional three-dimensional information acquisition device acquires three-dimensional information of crops, and the situation that the sensor is blocked by the sagging crop leaves easily occurs, so that the traditional three-dimensional information acquisition device is difficult to flexibly and accurately acquire three-dimensional information of different types of crops or crops of the same type in different growth periods, and further difficult to accurately acquire phenotype parameters of different types of crops or crops of the same type in different growth periods.

The three-dimensional information acquisition device is provided with a plurality of sensors with different visual angles, so that the three-dimensional information of different kinds of crops or crops with the same kind and different growth periods can be acquired more flexibly and accurately, but the three-dimensional information acquisition device is provided with a plurality of sensors with different visual angles, so that higher equipment cost is required, the subsequent calculation amount is larger, and the data processing is difficult.

In addition, the traditional three-dimensional information acquisition device generally acquires the three-dimensional information of crops in the walking process, but vibration and jolt generated when the traditional three-dimensional information acquisition device moves in a crop planting field can reduce the accuracy of the three-dimensional information of the crops acquired by the traditional three-dimensional information acquisition device, and cause problems such as loss or drift of slam construction drawings, so that the accuracy of crop phenotypes calculated based on the three-dimensional information of the crops can be reduced.

Aiming at the current situation that a three-dimensional information acquisition device with high stability, high flexibility, high accuracy and low cost is lacking in the related technology and the requirement of continuous dynamic monitoring of crop phenotypes, the invention provides the three-dimensional point cloud data acquisition device. The three-dimensional point cloud data acquisition device provided by the invention is suitable for acquiring three-dimensional point cloud data of greenhouse and field crops, has the advantages of small volume, flexible movement, good cross-country performance, high automation degree and the like, can acquire the three-dimensional point cloud data of different types of crops or crops of the same type in different growth periods more accurately and flexibly, further can acquire phenotypes of the different types of crops or the crops of the same type in different growth periods more accurately and flexibly, greatly reduces the workload of measuring staff, avoids the subjectivity of measurement, and can provide data support for the research and application of crop growth condition long-phase monitoring and diagnosis, genetic breeding assistance and screening, crop precise management and the like.

Fig. 1 is a schematic structural diagram of a three-dimensional point cloud data acquisition device provided by the invention. Fig. 2 is a second schematic structural diagram of the three-dimensional point cloud data acquisition device provided by the invention. The three-dimensional point cloud data acquisition device provided by the invention is described below with reference to fig. 1 and 2. As shown in fig. 1 and 2, the apparatus includes: the device comprises a travelling mechanism 2, a supporting structure 1, a first sliding rail 14, a first sliding block 12, a radar sensor 7, a controller 11 and a communication module 10;

the travelling mechanism 2 is connected with the supporting structure 1; the controller 11 is arranged on the supporting structure 1; the travelling mechanism 2 and the first sliding block 12 are respectively and electrically connected with the controller 11; the controller 11 and the user terminal are communicatively connected through the communication module 10.

Fig. 3 is a walking schematic diagram of the three-dimensional point cloud data acquisition device provided by the invention. As shown in fig. 3, the area to be measured is an acquisition object of the three-dimensional point cloud data acquisition device provided by the invention. The area to be measured may be a square area. A plurality of rows of crops are planted in the region to be detected, any two adjacent rows of crops in the region to be detected are parallel to each other, and a certain distance is reserved between any two adjacent rows of crops in the region to be detected, so that the three-dimensional point cloud data acquisition device provided by the invention can walk.

In the embodiment of the invention, any row of crops in the area to be detected can be determined as the target row of crops, and the area where the target row of crops is located is determined as the target area.

It will be appreciated that the target area is an elongate area.

In the embodiment of the invention, one boundary in the long-side direction of the target area can be determined as the initial walking position of the target area, and the other boundary in the long-side direction of the target area can be determined as the final walking position of the target area.

In the embodiment of the invention, a direction parallel to the long side of the target area and pointing from the start position to the end position of the target area is determined as a first direction.

It can be understood that, in the embodiment of the present invention, the walking direction of the three-dimensional point cloud data acquisition device is the first direction.

In the embodiment of the invention, the main material of the support structure 1 is stainless steel, and part of the main material is aluminum alloy so as to reduce the weight of the support structure 1. The whole weight of the three-dimensional point cloud data acquisition device in the embodiment of the invention is 200kg.

As an alternative embodiment, the running gear 2 comprises four telescopic brackets 15, four wheels 4, four running motors 3 and four steering motors 13; each telescopic bracket 15, each walking motor 3 and each steering motor 13 are electrically connected with the controller 11;

Any wheel 4 is rotatably connected with one end of a telescopic bracket 15;

the other end of any telescopic bracket 15 is rotationally connected with the supporting structure 1;

any one of the travel motors 3 is electrically connected with one wheel 4, and any one of the travel motors 3 is used for driving the wheel 4 to rotate in response to the control of the controller 11;

any steering motor 13 is electrically connected with one telescopic bracket 15, and any steering motor 13 is used for driving the telescopic bracket 15 to rotate in response to the control of the controller 11;

each telescoping support 15 also functions to extend or retract in response to control by the controller 11.

Specifically, in order to drive the support structure 1 to move smoothly, the running mechanism 2 adopts a four-wheel-drive symmetrical layout, and the running mechanism 2 includes four telescopic brackets 15, four wheels 4, four independent brushless motors (running motors 3) and four independent steering motors 13.

The support structure 1 in the embodiment of the invention may be square, and one end of any bracket may be rotatably connected to one corner of the support structure 1. Any tire is rotatably connected with the other end of one bracket.

Each wheel 4 is driven by a respective independent brushless motor, and the four supports are also driven by respective independent four steering motors 13, so as to drive the wheels 4 to steer.

The running mechanism 2 adopts two schemes of four-wheel steering and front wheel steering with the same direction of front wheels and rear wheels. The steering scheme can be switched through the remote controller according to different use scenes, when the front wheels and the rear wheels are used for steering in the same direction, the turning radius is smaller, the turning translation can be completed in a narrow space, and when the front wheels are used for steering, the turning radius is larger, and the direction of a vehicle body can be adjusted.

The running gear 2 may form a closed loop control system by means of an Arduino control board, four absolute value encoders and a controller 11. The controller 11 may control the steering of the four wheels 4 through an encoder, arduino control board. The controller 11 can also control the rotation of the four wheels by controlling the brushless motors of the four wheels 4.

And RS485 communication is adopted among the encoder, the driver, the Arduino control board and the controller 11.

Alternatively, the wheel 4 may be a solid tire. The material of the wheel 4 may be solid rubber. The wheel 4 may have an outer diameter of 660mm and a tire width of 35mm. The narrower tire width can enable the three-dimensional point cloud data acquisition device to flexibly pass through crops.

When the travelling mechanism 2 drives the supporting structure 1 to travel from the initial position to the final position of the target area, four wheels 4 in the travelling mechanism 2 are respectively positioned at two sides of the target row crops, and the four wheels 4 roll in gaps between the target row crops and the row crops at the two sides. The support structure 1 is located directly above the target area.

The running mechanism 2 is in four-wheel drive, the brushless motor can be a permanent magnet synchronous brushless direct current motor (BLDC), the rated power of the motor is 1.5kW, and the rated rotating speed is 3600r/min.

The steering motor 13 may be a brushed direct current motor (BDC) with a rated power of 0.95kW and a rated rotational speed of 1000r/min.

The dimensions of the support structure 1 and the height of the support in the running gear 2 can be determined from a priori knowledge and actual conditions, for example, the dimensions of the support structure 1 and the height of the support in the running gear 2 can be determined from the crop volume.

Alternatively, the dimensions of the support structure 1 and the running gear 2 in the embodiment of the invention are 2195mm×1900mm×2065mm, and the height of the lowest part of the support structure 1 from the ground is 1400mm.

The steering motor 13 and the travel motor 3 are electrically connected to the controller 11. The steering motor 13 may drive the tire to rotate in response to the control of the controller 11, thereby moving the support structure 1. The walking motor 3 can drive the support structure 1 to rotate in response to the control of the controller 11, so as to drive the support structure 1 to turn.

The first slide rail 14 is arranged on the support structure 1 perpendicular to the first direction; the first slider 12 is movably disposed on the first slide rail 14 along the extending direction of the first slide rail 14.

In the embodiment of the present invention, the vertical direction of the first direction is determined as the second direction. The first slide rail 14 is connected with the support structure 1 in the second direction.

The first slide rail 14 may be disposed at the bottom of the support structure 1 in the embodiment of the present invention.

It should be noted that the length of the first sliding rail 14 is not less than the width of the support structure 1 in the second direction.

Optionally, the first sliding rail 14 in the embodiment of the present invention is an electrically controlled sliding rail.

The radar sensor 7 is provided on the first slider 12 toward the vertically downward direction; the radar sensor 7 is in communication connection with the controller 11 and the communication module 10; the radar sensor 7 is configured to perform collection of point cloud data or collect stopping point cloud data in response to control of the controller 11, and send the collected original point cloud data to the user terminal through the communication module 10.

It can be understood that the laser radar is widely applied to the acquisition of three-dimensional information of crops due to the advantages of high precision, high measurement range and the like.

Alternatively, the radar sensor 7 in the embodiment of the present invention is a Velodyne VLP-16 laser radar, which has 16 scan lines of 360 degrees, the angular resolution measured in the horizontal direction is 0.1 ° to 0.4 °, the angular range measured in the vertical direction is 30 °, and the angular resolution is 2 °.

In the case where the dimensions of the support structure 1 and the travelling mechanism 2 are 2195mm×1900mm×2065mm and the height of the lowest part of the support structure 1 from the ground is 1400mm, the height from the ground after the laser radar is set on the first slider 12 is 1m.

Alternatively, the radar sensor 7 may be connected to the controller 11 via an ethernet network. The first slider 12 can be connected with the controller 11 through rs485 rotation usb.

It should be noted that, in the embodiment of the present invention, the raw point cloud data collected by the radar sensor 7 may include coordinates of each point in each frame under the radar coordinate system.

It should be noted that, the three-dimensional point cloud data acquisition device in the embodiment of the present invention may further include a power module 8, which may supply power to the travelling mechanism 2, the controller 11, the radar sensor 7, and other components.

Alternatively, the power module 8 may include two lead acid batteries and a power regulator.

It should be noted that, in the embodiment of the present invention, the communication module 10 may perform wireless communication with the user terminal. The user terminal is a terminal used by a user, such as a tablet computer, a notebook computer, and the like.

Optionally, the communication module 10 in the embodiment of the present invention may be a wireless router.

The controller 11 is used to determine that the support structure 1 reaches the target areaiWalking positionUnder the condition that the first slide block 12 is controlled to drive the radar sensor 7 to slide from one end of the first slide rail 14 to the other end at a constant speed, and the radar sensor 7 is controlled to collect point cloud data in the process that the first slide block 12 slides from one end of the first slide rail 14 to the other end at a constant speed, the controller 11 is further used for controlling the travelling mechanism 2 to drive the supporting structure 1 to slide from one end of the first slide rail 14 to the other end at a constant speed under the condition that the first slide block 12 is determined to slide from one end of the first slide rail 14 to the other endiThe walking position moves to the first position in the target areai+1 walking position, 1 sti+1 walking position is located at the firstiAt a preset distance of the walking position in the first direction,isequentially taking 0,1,2 and …,I，Ithe 0 th walking position is the initial walking position of the target area, which is a positive integer greater than 1IThe walking position is the termination walking position of the target area.

Specifically, in the embodiment of the present invention, the initial walking position of the target area may be determined as the 0 th walking position, and the walking position may be determined from the initial walking position of the target area at intervals of a preset distance in the first direction.

In the case of the first embodiment mThe distance between the walking position and the final walking position of the target area is smaller than the preset distance, the final walking position of the target area can be determined as the firstIWalking position and determiningI=m+1, a step of; wherein, the liquid crystal display device comprises a liquid crystal display device,mis a positive integer greater than 0,mis the quotient of the length of the long side of the target area and the preset distance.

It should be noted that the preset distance is predefined based on actual conditions and/or a priori knowledge. The value of the preset distance is not particularly limited in the embodiment of the present invention.

Optionally, in the embodiment of the present invention, the preset distance may have a value ranging from 180cm to 200cm, for example, the preset distance may have a value of 180cm, 190cm or 200cm.

Preferably, the preset distance has a value of 190cm.

For the firstiWalking positioniSequentially taking 0,1,2 and …,I) The controller 11 may determine that the support structure 1 is already located at the first position in a number of ways0 walking position, or whether it has been moved by the first position in the target areai-1 the walking position moves to the firstiThe walking position, for example, the controller 11 can be based on the position set at the first positioniSensor of the walking position, determining that the support structure 1 reaches the first mentionediA walking position; alternatively, the controller 11 may determine that the support structure 1 reaches the above-mentioned first using a positioning device provided on the support structure 1 iWalking position.

As an optional embodiment, the three-dimensional point cloud data acquisition device further includes: a positioning device 9; the positioning device is arranged on the supporting structure 1; the positioning device is electrically connected with the controller 11;

the positioning device is used for acquiring real-time position information of the supporting structure 1 and sending the real-time position information to the controller 11;

the controller 11 is also configured to receive the real-time position information and determine whether the support structure 1 reaches the first position based on the real-time position informationiThe traveling position and the traveling mechanism 2 are controlled.

It should be noted that the positioning device in the embodiment of the present invention may be a real-time dynamic differential positioning system (RTK-GPS). RTK-GPS can provide centimeter-level positioning accuracy, and the measurement frequency can reach 5Hz.

The RTK-GPS and the controller 11 adopt an RS232 communication mode.

The controller 11 determines the first time the support structure 1 reaches the target areaiUnder the condition of the walking position, the first sliding block 12 can be controlled to drive the radar sensor 7 to slide from one end of the first sliding rail 14 to the other end at the same speed through a control instruction.

It should be noted that, the first slider 12 drives the radar sensor 7 to slide from the current end of the first sliding rail 14 to the other end at a preset speed. The preset speed may be predefined based on actual conditions and/or a priori knowledge. The value of the preset speed is not particularly limited in the embodiment of the present invention.

The controller 11 can control the radar sensor 7 to start collecting the point cloud data through a control instruction while controlling the first sliding block 12 to start sliding from one end of the first sliding rail 14 to the other end at equal speedThe method comprises the steps of carrying out a first treatment on the surface of the Under the condition that the controller 11 determines that the first sliding block 12 has uniformly slid from the end where the first sliding rail 14 is currently located to the other end, the control instruction can control the collection of cloud data at the stopping point of the radar sensor 7, so that the radar sensor 7 can collect the first sliding block in the process of uniformly sliding from the end where the first sliding rail 14 is currently located to the other endiRaw point cloud data at walking locations.

It should be noted that, based on the feedback information returned by the first slider 12 and/or the first slide rail 14, the controller 11 may determine that the first slider 12 has slid from the end where the first slide rail 14 is currently located to the other end at equal speed.

The radar sensor 7 is acquiring the firstiThe first step can be performed after the original point cloud data at the walking positioniThe original point cloud data at the walking position is sent to the user terminal through the communication module 10, so that the user can look up the first point cloud data through the user terminaliOrigin cloud data at walking position and based on the above-mentioned first iThe raw point cloud data at the walk location is subjected to further data analysis.

The controller 11 can also control the running mechanism 2 to drive the supporting structure 1 to move from the first position in the target area through a control command under the condition that the first sliding block 12 is determined to slide from the current one end of the first sliding rail 14 to the other end at equal speediThe walking position moves to the firsti+1 walking position and when it is determined that the support structure 1 reaches the first positioniIn the case of the +1 travel position, the above-described control process is repeated for the first slider 12 and the radar sensor 7.

The radar sensor in the three-dimensional point cloud data acquisition device moves at fixed points, can uniformly and rapidly acquire three-dimensional point cloud data in the direction perpendicular to the advancing direction of the three-dimensional point cloud data acquisition device, can acquire three-dimensional point cloud data of crops at multiple positions and multiple angles, is in a static state in the process of three-dimensional point cloud data acquisition, can avoid vibration and jolt generated by the three-dimensional point cloud data acquisition device during movement, influences the accuracy of three-dimensional point cloud data acquired by the radar sensor, can acquire three-dimensional point cloud data of crops of different types or crops of the same type in different growth periods more flexibly and accurately, can provide data support for researches and applications such as crop growth phase monitoring and diagnosis, genetic breeding assistance and screening, crop precise management and the like, and the three-dimensional point cloud data acquisition device in the embodiment of the invention is suitable for three-dimensional point cloud data acquisition of greenhouse and field crops, and has the advantages of small volume, flexible movement, good performance, high degree of automation and the like.

Based on the content of the above embodiments, the three-dimensional point cloud data acquisition device further includes: a second slide rail 5 and a second slider 6;

the second slide rail 5 is arranged on the support structure 1 perpendicular to the first slide rail 14; the second sliding block 6 is movably arranged on the second sliding rail 5 along the extending direction of the second sliding rail 5;

the second sliding block 6 is connected with the first sliding rail 14; the second sliding block 6 is in communication connection with the controller 11; the second sliding block 6 is used for responding to the control of the controller 11 and driving the first sliding rail 14 to slide along the second sliding rail 5 from the preset stay position where the first sliding rail is currently positioned to the other preset stay position;

in particular, the second slide rail 5 may be arranged at the bottom of the support structure 1 in the first direction.

It should be noted that the length of the second slide rail 5 is not smaller than the width of the support structure 1 in the first direction.

Optionally, the second slide rail 5 in the embodiment of the present invention is an electrically controlled slide rail.

It should be noted that in the embodiment of the present invention, the number of the second sliding rails 5 may be one or more.

It should be noted that, in the embodiment of the present invention, a plurality of preset stay positions may be determined for the first slide rail 14 based on a priori knowledge and/or actual conditions.

Optionally, in the embodiment of the present invention, the first sliding rail 14 is located at one end of the second sliding rail 5, the first sliding rail 14 is located at the other end of the second sliding rail 5, and the first sliding rail 14 is located at the middle of the second sliding rail 5, which are determined as three preset stay positions.

As an alternative embodiment, the number of the second slide rails 5 and the second sliders 6 is 2; the two second sliding rails 5 are arranged in parallel at intervals; each second sliding block 6 is movably arranged on each second sliding rail 5 along the extending direction of each second sliding rail 5;

one end of the first sliding rail 14 is arranged on one second sliding block 6, and the other end of the first sliding rail 14 is arranged on the other second sliding block 6.

The controller 11 is also used to determine that the support structure 1 reaches the firstiUnder the condition that the first sliding rail 14 is positioned at any preset stay position at the walking position, the first sliding block 12 is controlled to drive the radar sensor 7 to slide from one end of the first sliding rail 14 to the other end at a constant speed, and the radar sensor 7 is controlled to collect point cloud data in the process of sliding from one end of the first sliding rail 14 to the other end at a constant speed, the controller 11 is further used for controlling the second sliding block 6 to drive the preset stay position of the first sliding rail 14 to slide from one end of the first sliding rail 14 to the other preset stay position under the condition that the first sliding block 12 is determined to slide from one end of the first sliding rail 14 to the other end at a constant speed, and the controller 11 is further used for controlling the walking mechanism 2 to drive the supporting structure 1 to slide from the first sliding rail 14 under the condition that the first sliding rail 14 is determined to be positioned at each preset stay position iThe walking position moves to the firsti+1 walking position.

It will be appreciated that after the support structure 1 reaches the firstiIn the case of the walking position, the first slide 14 is necessarily located at a certain preset resting position, and therefore, the controller 11 determines that the support structure 1 reaches the first positioniUnder the condition of the walking position, the first sliding block 12 can be controlled to drive the radar sensor 7 to slide from one end of the first sliding rail 14 to the other end at the same speed through a control instruction.

The controller 11 can control the first sliding block 12 to start sliding from one end of the first sliding rail 14 to the other end at the same time, and can also control the radar sensor 7 to start collecting the point cloud data through a control instruction; in the case that the controller 11 determines that the first sliding block 12 has slid from the end where the first sliding rail 14 is currently located to the other end at equal speed, the control instruction can control the radar sensor 7 to stop collecting cloud data, so that the radar sensor 7 can make the first sliding block slide from the end where the first sliding rail 14 is currently locatedIn the process that one end of the first slide rail 14 slides to the other end at the same speed, the first slide rail is collectediOriginal point cloud data at a preset stay position where the first slide rail 14 is currently located at the walking position.

The radar sensor 7 collects the first iThe first slide rail 14 can be moved to the first position after the original point cloud data at the preset stop positioniThe original point cloud data of the preset stay position where the first slide rail 14 of the walking position is currently located is sent to the user terminal through the communication module 10,

the controller 11 can also control the second slider 6 to drive the first slider 14 to slide from the preset stop position where the first slider 14 is currently located to another preset stop position through the control command under the condition that it is determined that the first slider 12 is at the current position and the first slider 14 has slid from the end where the first slider 14 is currently located to the other end at uniform speed, and repeat the above control process for the first slider 12.

It should be noted that, based on feedback information returned by the second slider 6 and/or the second slide rail 5, the controller 11 may determine that the first slide rail 14 has slid from the preset parking position where it is currently located to another preset parking position.

The controller 11 can also control the travelling mechanism 2 to drive the supporting structure 1 to move from the first position to the second position through a control command under the condition that the controller 11 determines that the first sliding rail 14 is positioned at each preset stop position and the first sliding block 12 is uniformly slid from one end of the first sliding rail 14 to the other end iThe walking position moves to the firsti+1 walking position and when it is determined that the support structure 1 reaches the first positioniIn the case of the +1 travel position, the above-described control process is repeated for the first slider 12, the second slider 6, and the radar sensor 7.

It can be appreciated that in the support structure 1, the first one is formed ofiThe walking position moves to the firstiBefore the +1 walking position, the user terminal can receive the firstiOriginal point cloud data of each preset stay position of the walking position, and the three-dimensional point cloud data of the plurality of groups of crops are the same as the first point cloud dataiThe walking positions correspond to each group of three-dimensional point cloud data of crops, and each group of three-dimensional point cloud data of the crops also corresponds to each preset stay position.

The first guide rail in the three-dimensional point cloud data acquisition device can also slide in the first direction, so that more positions and more visual angles can be provided for the radar sensor 7 to acquire three-dimensional point cloud data of crops, and the flexibility and accuracy for acquiring the three-dimensional point cloud data of different types of crops or crops with the same type and different growth periods can be further improved.

Optionally, the user terminal may also send a control instruction to the controller 11 through the communication module 10, so that the controller 11 controls the traveling direction of the travelling mechanism 2, the position of the first slider 12, or the position of the second slider 6 in response to the control instruction. The user terminal may also send a travel route plan to the controller 11 via the communication module 10, so that the controller 11 may control the travelling mechanism 2 based on the travel route plan.

Based on the above embodiments, the controller 11 is further configured to, when determining that the support structure 1 is located at the first positionIUnder the condition that the first sliding block 12 at the walking position slides from one end of the first sliding rail 14 to the other end at a constant speed, target information is sent to the user terminal through the communication module 10, so that the user terminal can acquire the three-dimensional point cloud block corresponding to the target area based on all the received original point cloud data under the condition that the target information is received.

Specifically, the controller 11 determines that the support structure 1 is located at the first positionIThe first sliding block 12 at the walking position slides from one end of the first sliding rail 14 to the other end at a constant speed, and the acquisition target information can be sent to the user terminal through the communication module 10. The target information indicates that the three-dimensional point cloud data acquisition device has completed acquisition of the three-dimensional point cloud block corresponding to the target area.

Under the condition that the user terminal receives the target information, a three-dimensional point cloud block corresponding to the target area can be obtained through a data processing mode based on all received original point cloud data. All original point cloud data received by the user terminal include original point cloud data at 0 th walking position, original point cloud data at 1 st walking position, original point cloud data at 2 nd walking position …, and the like mRaw point cloud data at walk location and itemIWalking positionRaw point cloud data at the location.

As an alternative embodiment, the controller 11 is further configured to, upon determining that the first slider 12 is positioned at the first position of the support structure 1IThe first sliding rail 14 is located at each preset stay position, and when the first sliding block 12 slides from one end of the first sliding rail 14 to the other end at a constant speed, the communication module 10 sends target information to the user terminal, so that the user terminal obtains a three-dimensional point cloud block corresponding to the target area based on all received original point cloud data under the condition that the user terminal receives the target information.

Specifically, the controller 11 determines that the support structure 1 is located at the first positionIThe first sliding rail 14 is located at each preset stop position at the walking position, and the first sliding block 12 is slid from one end of the first sliding rail 14 to the other end at a constant speed, so that the acquisition target information can be sent to the user terminal through the communication module 10. The target information indicates that the three-dimensional point cloud data acquisition device has completed acquisition of the three-dimensional point cloud block corresponding to the target area.

Under the condition that the user terminal receives the target information, a three-dimensional point cloud block corresponding to the target area can be obtained through a data processing mode based on all received original point cloud data. All original point cloud data received by the user terminal, including original point cloud data at each preset stay position of the 0 th walking position, original point cloud data at each preset stay position of the 1 st walking position, original point cloud data at each preset stay position of the 2 nd walking position, …, and the 2 nd walking position mOriginal point cloud data at each preset stay position of walking position and the first point cloud dataIRaw point cloud data at each preset stay position of the walking position.

Based on the content of each embodiment, the user terminal obtains the three-dimensional point cloud block corresponding to the target area based on all the received original point cloud data, and the method comprises the following steps: based on the preset speed and the preset distance, carrying out coordinate system conversion on all original point cloud data to obtain three-dimensional point cloud blocks corresponding to each walking position, wherein the speed of the first sliding block 12 from one end of the first sliding rail 14 to the other end at a constant speed is the preset speed;

specifically, since the radar sensor 7 performs three-dimensional point cloud data acquisition in the uniform sliding process, the relative pose between frames does not need to be estimated, and the pose translation between frames can be acquired at a preset speed.

Therefore, in the case where the preset speed is Amm/s and the speed of three-dimensional point cloud data acquisition by the radar sensor 7 is 15 frames per second, for the P-th point of the k-th frame in all the above-described original point cloud data, the coordinates of the P-th point in the world coordinate system can be obtained based on the coordinates P of the P-th point in the radar coordinate system by 。

（1）

Wherein:

（2）

coordinates P to coordinatesThe conversion process of (1) is expressed as->；

Wherein the transformation matrix T can be expressed as:

（3）

since the radar sensor 7 is slid in the second direction or the opposite direction of the second direction, in the formula (3):

based on the formulas (1) to (3), the coordinates of each point of each frame in the above all original point cloud data in the world coordinate system can be obtained.

After the coordinates of each point of each frame in the original point cloud data in the world coordinate system are obtained, the coordinates of each point of any two adjacent frames in the world coordinate system can be spliced in a uniform-speed superposition mode, and a three-dimensional point cloud block corresponding to each walking position is obtained.

After coarse registration is performed on the three-dimensional point cloud blocks corresponding to each walking position, sequentially performing fine registration on the three-dimensional point cloud blocks corresponding to each two adjacent walking positions based on an iterative closest point algorithm according to the arrangement sequence of each walking position in the first direction, and obtaining the three-dimensional point cloud blocks corresponding to each walking position after registration.

Specifically, after the three-dimensional point cloud block for each walking position is obtained, the first walking position may be sequentially aligned according to the arrangement sequence of the first walking position based on the iterative closest point algorithm (Iterative Closest Point, ICP) iThree-dimensional point cloud block and third corresponding to walking positioni+1 three-dimensional point cloud block corresponding to walking positioniSequentially taking 0,1,2 and …,I-1) And performing fine registration to obtain a three-dimensional point cloud block corresponding to the target area.

Because the iterative nearest point algorithm has high requirements on initial positions of the registration point cloud and the reference point cloud, if the initial positions of the two point clouds are very different, the algorithm is easy to generate a local optimal condition after registration.

Therefore, in the embodiment of the invention, the iterative closest point algorithm pair is basediThree-dimensional point cloud block and third corresponding to walking positioni+1, before the three-dimensional point cloud block corresponding to the walking position is precisely registered, firstly, the first three-dimensional point cloud block is registerediAnd performing coarse registration on the three-dimensional point cloud blocks corresponding to the walking positions.

Due to the firstiWalking position and the firsti+1 the distance between the walking positions is a preset distance (the preset distance is 190cm in the embodiment of the invention), andiwalking position and the firsti+1 has a 30% overlap between walking positions, thus, will beiThe three-dimensional point cloud block corresponding to the walking position moves 190 ℃ to the first directioni-1) cm, can be finishedBecome the firstiCoarse registration of three-dimensional point cloud blocks corresponding to walking positions.

The iterative nearest point algorithm is based on a least squares method, and finds nearest neighbors according to certain constraints to calculate the optimal registration parameters, i.e. by changing the rotation matrix R and the translation vector t such that the value of the error function E (R, t) is minimized. The error function E (R, t) can be expressed by the following formula:

(4)

Where n is the number of nearest neighbor pairs,is a point in the target point cloud P, +.>Is the AND +.>The corresponding closest point, R, is the rotation matrix and t is the translation vector. The implementation steps of the iterative closest point algorithm comprise:

step S11, selecting a point set from the target point cloud P∈P；

Step S12, finding out in the source point cloud QCorresponding Point set->Satisfy->E Q and let ∈ ->The value of (2) is the smallest;

step S13, obtaining a rotation matrix R and a translation vector t by calculating the relation between the corresponding point sets, so that the error is causedFunction ofMinimum;

step S14, R and t are calculated according to the step S3 by matchingTransforming to obtain->Is a set of transformation points of (a)；

Step S15, calculatingAnd->Corresponding Point set->An average distance d of (2); the calculation formula of the distance d is as follows:

(5)

step S16, if d is smaller than the predefined threshold value or the iteration number is larger than the predefined iteration number threshold value, determining that the convergence condition is met, otherwise, repeatedly executing the steps S12 to S16 until the convergence condition is met.

In the embodiment of the invention, the first step can beiDetermining a three-dimensional point cloud block corresponding to the walking position as a source point cloud Q, and determining the third point cloud block as the source point cloud Qi+1, determining the three-dimensional point cloud block corresponding to the walking position as the target point cloud P, and further realizing the first step based on the steps S11 to S16 iThree-dimensional point cloud block and third corresponding to walking positioni+And 1, fine registration of the three-dimensional point cloud block corresponding to the walking position.

At the position ofiSequentially taking 0,1,2 and …,I-1, can realize three-dimensional corresponding to every two adjacent walking positions in the first directionAnd (3) performing fine registration on the point cloud blocks to obtain three-dimensional point cloud blocks corresponding to each walking position after registration.

Specifically, after the three-dimensional point cloud block corresponding to each travel position after registration is obtained, the first registered point cloud block after registration can be splicediThree-dimensional point cloud block corresponding to walking position and post-registration thirdiA three-dimensional point cloud block corresponding to the +1 walking position,isequentially taking 0,1,2 and …,I-1。

and sequentially splicing the three-dimensional point cloud blocks corresponding to the walking positions after registration according to the arrangement sequence of the walking positions in the first direction, so as to obtain the three-dimensional point cloud blocks corresponding to the target area.

According to the user terminal provided by the embodiment of the invention, after the three-dimensional point cloud blocks corresponding to each walking position are roughly registered, the three-dimensional point cloud blocks corresponding to each two adjacent walking positions are finely registered based on an iterative nearest point algorithm according to the arrangement sequence of each walking position in the first direction, so that the three-dimensional point cloud blocks corresponding to each walking position after registration are obtained, and then the three-dimensional point cloud blocks corresponding to each walking position after registration are spliced according to the arrangement sequence of each walking position in the first direction, so that the three-dimensional point cloud blocks corresponding to a target area can be more accurately obtained, and data support can be provided for the crop phenotype of the target area.

Based on the content of each embodiment, the user terminal is further configured to obtain the phenotype parameter of the crop in the target area based on the three-dimensional point cloud block corresponding to the target area.

It should be noted that, the point cloud data collected by the laser sensor may have uneven density, and may generate sparse outliers, thereby reducing accuracy of the point cloud data.

Therefore, in the embodiment of the invention, the Statistical Outlier Removal filter is used for removing the outlier points or the rough difference points caused by the measurement errors in the three-dimensional point cloud block corresponding to the target area, so as to obtain the three-dimensional point cloud block corresponding to the denoised target area.

The filtering process of the Statistical Outlier Removal filter comprises the following steps: a statistical analysis is performed on the neighborhood of each point to calculate its average distance to all nearby points. Assuming that the result is a gaussian distribution whose shape is determined by the mean and standard deviation, points whose mean distance is outside of the standard range (defined by the global distance mean and variance) can be defined as outliers and removed from the data.

After the three-dimensional point cloud block corresponding to the denoised target area is obtained, the ground is required to be detected, and the ground point cloud in the three-dimensional point cloud block corresponding to the denoised target area is removed, so that the three-dimensional point cloud block corresponding to crops in the target area is obtained.

Therefore, in the embodiment of the invention, a Random sample consensus (RANSAC) algorithm is applied to distinguish the ground point cloud in the three-dimensional point cloud block corresponding to the denoised target area from the crop point cloud, so that the ground point cloud is prevented from being erroneously detected as the crop in subsequent processing.

When the algorithm is designed, firstly setting the iteration number of the algorithm to be 1, the distance error threshold to be delta T1, the total point number in the three-dimensional point cloud block corresponding to the denoised target area to be N, randomly selecting 3 points to form the ground to be fitted at the initial time, and setting the three-dimensional coordinates of the 3 points to be (X ₁ ，Y ₁ ，Z ₁ )、(X ₂ ，Y ₂ ，Z ₂ )、(X ₃ ，Y ₃ ，Z ₃ ) The fitted plane model is

(6)

Wherein the method comprises the steps of

A=(Y ₂ －Y ₁ )·(Z ₃ －Z ₁ )－(Z ₂ －Z ₁ )·(Y ₃ －Y ₁ )

B=(Z ₂ －Z ₁ )·(X ₃ －X ₁ )－(X ₂ －X ₁ )·(Z ₃ －Z ₁ )

C=(X ₂ －X ₁ )·(Y ₃ －Y ₁ )－(Y ₂ －Y ₁ )·(X ₃ －X ₁ )

D=－(AX ₁ +BY ₁ +CZ ₁ )

Then any point in space (X ₀ ，Y ₀ ，Z ₀ ) The distance L to the plane is

（7）

If the distance L between a certain point and the assumed plane is less than or equal to deltaT 1, the point is the point in the model. And traversing other N-3 points except the initial sampled 3 points in sequence, and recording the number of the inner points of the model. And randomly sampling 3 points to construct a plane model, and obtaining the number of inner points of the model according to the same method. The method of random sampling is iterated 1 time. The probability of producing a reasonable result increases with increasing iteration number. And finally, voting based on the number of the internal points of each model, and selecting the ground model with the largest number of the internal points as the best fitting result.

After the three-dimensional point cloud blocks corresponding to the crops in the target area are obtained, first, each plant of the three-dimensional point cloud blocks corresponding to the crops can be separated by using an European clustering algorithm.

The essence of European clustering is to classify points that are close in distance into one class. Assuming that n points exist in the point cloud C, defining Euclidean distance as the affinity and sparseness degree of two points, and taking the distance between adjacent points as a standard, thereby realizing the point cloud clustering segmentation. The specific process of segmentation comprises the following steps: for the preprocessed point cloud data set P, determining a query point Pi, setting a distance threshold r, and finding n nearest neighbor points Pj (j=1, 2, ⋯, n) nearest to the query point Pi through KD-Tree; calculating Euclidean distances dj from n adjacent points to the query point according to a formula (8); comparing the distance dj with a distance threshold r, and classifying points smaller than r into a class M until the number of points in the class M is not increased any more, and completing the segmentation.

(8)

The algorithm can only divide single plants under the condition that leaves between two basins are not overlapped, leaves of some crops grow very large, and the arrangement distance between the two basins is too short, so that European clustering can not divide the single plants under the condition, and then the K-means clustering algorithm is adopted for dividing the crops under the condition. The K-means algorithm belongs to a partitional clustering algorithm, in which the center of each cluster is represented by the mean of all objects in the cluster. Inputs are the number of clusters (K) and a dataset (D) containing n objects; the output is a set of K clusters.

The algorithm flow comprises

And 21, arbitrarily selecting K objects from the D as initial clustering centers.

Step 22, each object is allocated to the most similar cluster according to the average value of the objects in the cluster.

And step 23, updating the cluster mean, namely recalculating the mean of the objects in each cluster.

Step 24, loop through steps 22 and 23 until no more changes in the clusters occur.

The characteristic of K-mean clustering method is that the cluster number and the initial cluster center are determined in advance.

Based on the mode, the single plant three-dimensional point cloud corresponding to each crop in the target area can be obtained. Based on the single three-dimensional point cloud of each crop, the plant height and the phenotype parameter of the maximum crown amplitude of each crop in the target area can be obtained.

Based on the single three-dimensional point cloud of each crop, the method for acquiring the plant height of each crop in the target area specifically comprises the following steps: because the point cloud data acquired by the laser radar is inconsistent with the xyz coordinate axis direction of the world coordinate system, the whole point cloud is subjected to horizontal plane calibration before the plant height is extracted.

Because the ground point cloud and the crop point cloud in the target area are distinguished in the early stage, the ground plane equation can be estimated through the separated ground point cloud 。

The rotation matrix R is obtained through the normal vector a (a, b, c) before the horizontal calibration of the ground and the vector b (0, 1) vertically upwards of the radar point cloud coordinate system, the original point cloud is multiplied by the rotation matrix R to obtain the point cloud after the horizontal calibration, the difference between the z value of the highest point of crops and the z value of the ground plane is calculated, and the known flowerpot height is subtracted, so that the plant height is obtained. The plant height calculation formula is

(9)

Based on the single three-dimensional point cloud of each crop, the maximum crown width of each crop in the target area is obtained, and the method specifically comprises the following steps: the extraction of the maximum crown amplitude for each crop is essentially to find the furthest point pair in the blade Ping Miandian cloud, which can take advantage of the geometry.

Firstly, calculating projections of the single-plant point cloud in the vertical direction, namely enabling Z coordinate values of all points in the single-plant point cloud to be 0, so that the single-plant point cloud is changed from a three-dimensional point cloud to a two-dimensional point cloud on an xy plane; then, the convex polygon outline of the two-dimensional Ping Miandian cloud can be extracted, after the convex hull is obtained, the furthest point pair of the Ping Miandian cloud convex hull is calculated by a rotation stuck-hull algorithm, and the algorithm flow comprises:

step 31, calculating the end point of the convex polygon in the y direction, which is called y _min And y _max ；

Step 32, by y _min And y _max Constructing two horizontal tangents, calculating the distance between them and maintaining the distance as a current maximum value because the two horizontal tangents are already a pair of butts;

step 33, rotating two lines simultaneously until one of the lines coincides with one side of the polygon;

step 34, a new butt point is generated for this time. Calculating a new distance, comparing the new distance with the current maximum value, and updating if the new distance is larger than the current maximum value;

step 35, the process of steps 33 and 34 is repeated until the pair of heels (y _min ,y _max ) ；

Step 36, outputting the pair of heels determining the maximum diameter.

The algorithm time complexity is O (n), and the butt point pair with the maximum diameter calculated by the algorithm is the maximum crown width of each crop.

After removing noise and ground point clouds in a three-dimensional point cloud block corresponding to a target area based on a Statistical Outlier Removal filter and a RANSAC algorithm, the user terminal in the embodiment of the invention cuts out single plant three-dimensional point clouds of each crop in the target area based on European clusters and K-means clusters, so that phenotype parameters of each crop in the target area can be extracted more accurately.

Fig. 4 is a schematic flow chart of a crop phenotype acquisition method provided by the invention. The crop phenotype acquisition method provided by the invention is realized based on the three-dimensional point cloud data acquisition device. The crop phenotype acquisition method of the present invention is described below with reference to fig. 4. As shown in fig. 4, the method includes: step 401, receiving original point cloud data;

It should be noted that, in the embodiment of the present invention, the execution body is a crop phenotype acquisition device, and the crop phenotype acquisition device is a user terminal in the three-dimensional point cloud data acquisition device in each embodiment.

Step 402, under the condition of receiving target information, acquiring a three-dimensional point cloud block corresponding to a target area based on all received original point cloud data;

as an optional embodiment, based on all the received original point cloud data, obtaining a three-dimensional point cloud block corresponding to the target area includes: acquiring a preset speed and a preset distance;

based on the preset speed and the preset distance, performing coordinate system conversion on all original point cloud data to obtain three-dimensional point cloud blocks corresponding to each walking position;

Step 403, obtaining the phenotype parameters of the crops in the target area based on the three-dimensional point cloud block corresponding to the target area.

It should be noted that, the specific steps of the user terminal for executing the crop phenotype acquiring method provided by the present invention to acquire the phenotype parameter as the target area may be referred to the content of each embodiment, which is not described in detail in the embodiments of the present invention.

In order to facilitate understanding of the crop phenotype acquisition method provided by the present invention, the crop phenotype acquisition method provided by the present invention is described below by way of one example. FIG. 5 is a second flow chart of the crop phenotype acquisition method according to the present invention. As shown in fig. 5, after the radar sensor 7 acquires the original point cloud data, the original point cloud data is transmitted to the user terminal;

under the condition that the user terminal receives the target information, acquiring a three-dimensional point cloud block corresponding to each walking position in the target area based on all received original point cloud data;

the user terminal registers and splices the three-dimensional point cloud blocks corresponding to each walking position in the target area to obtain the three-dimensional point cloud blocks corresponding to the target area;

the user terminal eliminates noise and ground point clouds in the three-dimensional point cloud block corresponding to the target area, and obtains the three-dimensional point cloud block corresponding to crops in the target area;

The user terminal performs single-plant segmentation on the three-dimensional point cloud blocks corresponding to the crops in the target area to obtain single-plant three-dimensional point clouds corresponding to each crop in the target area;

the user terminal can extract the plant height and the maximum crown amplitude of each crop in the target area based on the single three-dimensional point cloud corresponding to each crop in the target area.

According to the embodiment of the invention, after the three-dimensional point cloud block corresponding to the target area is obtained based on the original point cloud data acquired by the radar sensor, the phenotype parameters of crops in the target area can be more accurately and more efficiently acquired based on the three-dimensional point cloud block corresponding to the target area, and data support can be provided for researches and applications such as crop growth vigor long-phase monitoring and diagnosis, genetic breeding assistance and screening, crop precise management and the like.

Fig. 6 is a schematic structural view of a crop phenotype acquisition apparatus provided by the invention. The crop phenotype acquiring apparatus provided by the present invention will be described below with reference to fig. 6, and the crop phenotype acquiring apparatus described below and the crop phenotype acquiring method provided by the present invention described above may be referred to correspondingly. As shown in fig. 6, a data acquisition module 601, a point cloud processing module 602, and a phenotype acquisition module 603.

A data acquisition module 601, configured to receive original point cloud data;

the point cloud processing module 602 is configured to, when receiving the target information, obtain a three-dimensional point cloud block corresponding to the target area based on all the received original point cloud data;

the phenotype acquisition module 603 is configured to acquire a phenotype parameter of the crop in the target area based on the three-dimensional point cloud block corresponding to the target area.

Specifically, the data acquisition module 601, the point cloud processing module 602, and the phenotype acquisition module 603 are electrically connected.

It should be noted that, the crop phenotype acquiring device in the embodiment of the present invention is a user terminal in the three-dimensional point cloud data acquiring device in each embodiment.

According to the crop phenotype acquisition device, after the three-dimensional point cloud block corresponding to the target area is acquired based on the original point cloud data acquired by the radar sensor, the phenotype parameters of crops in the target area can be acquired more accurately and more efficiently based on the three-dimensional point cloud block corresponding to the target area, and data support can be provided for researches and applications such as crop growth vigor long-phase monitoring and diagnosis, genetic breeding assistance and screening, crop precise management and the like.

Fig. 7 illustrates a physical schematic diagram of an electronic device, as shown in fig. 7, which may include: processor 710, communication interface (Communications Interface) 720, memory 730, and communication bus 740, wherein processor 710, communication interface 720, memory 730 communicate with each other via communication bus 740. Processor 710 may invoke logic instructions in memory 730 to perform a crop phenotype acquisition method comprising: receiving original point cloud data; under the condition that target information is received, acquiring a three-dimensional point cloud block corresponding to a target area based on all received original point cloud data; and acquiring the phenotype parameters of the crops in the target area based on the three-dimensional point cloud block corresponding to the target area.

Further, the logic instructions in the memory 730 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In another aspect, the present invention also provides a computer program product comprising a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program, when executed by a processor, being capable of performing the crop phenotype acquisition method provided by the methods described above, the method comprising: receiving original point cloud data; under the condition that target information is received, acquiring a three-dimensional point cloud block corresponding to a target area based on all received original point cloud data; and acquiring the phenotype parameters of the crops in the target area based on the three-dimensional point cloud block corresponding to the target area.

In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which when executed by a processor is implemented to perform the crop phenotype acquisition method provided by the methods described above, the method comprising: receiving original point cloud data; under the condition that target information is received, acquiring a three-dimensional point cloud block corresponding to a target area based on all received original point cloud data; and acquiring the phenotype parameters of the crops in the target area based on the three-dimensional point cloud block corresponding to the target area.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-dimensional point cloud data acquisition device, comprising: the device comprises a travelling mechanism, a supporting structure, a first sliding rail, a first sliding block, a radar sensor, a controller and a communication module;

the controller is also used for controlling the travelling mechanism to drive the supporting structure to move from the first sliding rail under the condition that the first sliding block is determined to slide from one end of the first sliding rail to the other end at a constant speediThe walking position moves to the first position in the target areai+1 walking position, the firsti+1 walking position is located at the first positioniPresetting of the walking position in the first direction At the distance from each other,isequentially taking 0,1,2 and …,I，Ithe 0 th walking position is the initial walking position of the target area and is a positive integer greater than 1IThe walking position is the termination walking position of the target area.

2. The three-dimensional point cloud data acquisition device of claim 1, further comprising: the second sliding rail and the second sliding block;

3. The three-dimensional point cloud data acquisition device according to claim 2, wherein the number of the second sliding rails and the second sliding blocks is 2; the two second sliding rails are arranged in parallel at intervals; each second sliding block is movably arranged on each second sliding rail along the extending direction of each second sliding rail;

4. The three-dimensional point cloud data acquisition device of claim 1, wherein the travelling mechanism comprises four telescopic brackets, four wheels, four travelling motors and four steering motors; each telescopic bracket, each traveling motor and each steering motor are electrically connected with the controller;

Any wheel is rotationally connected with one end of a telescopic bracket;

5. The three-dimensional point cloud data acquisition device of claim 1, further comprising: positioning equipment; the positioning device is arranged on the supporting structure; the positioning device is electrically connected with the controller;

the controller is also configured to receive theReal-time position information and determining whether the support structure reaches the first position based on the real-time position informationiAnd the walking position and the walking mechanism are controlled.

6. The three-dimensional point cloud data acquisition apparatus according to any one of claims 1 to 5, further comprising: the user terminal.

7. The three-dimensional point cloud data collection device of claim 6, wherein said controller is further configured to, upon determining that said support structure is located at a first locationIAnd under the condition that the first sliding block at the walking position slides from one end of the first sliding rail to the other end at a constant speed, sending target information to the user terminal through the communication module, so that the user terminal can acquire the three-dimensional point cloud block corresponding to the target area based on all received original point cloud data under the condition that the user terminal receives the target information.

8. The three-dimensional point cloud data acquisition device according to claim 7, wherein the user terminal acquires the three-dimensional point cloud block corresponding to the target area based on all the received original point cloud data, and the method comprises the following steps:

9. The three-dimensional point cloud data acquisition device of claim 8, wherein the user terminal is further configured to acquire a phenotype parameter of a crop in the target area based on the three-dimensional point cloud block corresponding to the target area.

10. A crop phenotype acquisition method implemented based on the three-dimensional point cloud data acquisition device according to claim 7, comprising:

receiving original point cloud data;

11. The method for obtaining a crop phenotype according to claim 10, wherein the obtaining a three-dimensional point cloud block corresponding to the target area based on all the received original point cloud data includes:

acquiring a preset speed and a preset distance;

12. A crop phenotype acquisition apparatus comprising: