CN109708578A - Device, method and system for measuring plant phenotype parameters - Google Patents
Device, method and system for measuring plant phenotype parameters Download PDFInfo
- Publication number
- CN109708578A CN109708578A CN201910135784.1A CN201910135784A CN109708578A CN 109708578 A CN109708578 A CN 109708578A CN 201910135784 A CN201910135784 A CN 201910135784A CN 109708578 A CN109708578 A CN 109708578A
- Authority
- CN
- China
- Prior art keywords
- plant
- point cloud
- tested
- image
- camera
- 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
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention discloses a kind of plant phenotype parameter measuring apparatus, method and system.Measuring device includes: scanning head, test chassis, test platform and computer；Scanning head is fixed in test chassis, and test chassis can move up and down；Test platform is for bearing plant to be measured；Scanning head is with the plant image of the moving sweep of test chassis plant different height to be measured, and by plant image transmitting to computer, computer is used to determine the phenotypic parameter of plant to be measured according to plant image；Scanning head includes multiple video cameras and two linear laser transmitters, and multiple positions for video camera are in first level face, and multiple video camera equidistantly distributeds are on virtual circular arc；Two linear laser transmitters are located at the second horizontal plane below first level face.There is blind area when can be avoided data splicing in the present invention, improves the precision of plant phenotype parameter measurement.
Description
Technical field
The present invention relates to plant growth parameter measurement fields, more particularly to a kind of plant phenotype parameter measuring apparatus, side
Method and system.
Background technique
The core ideas of precision agriculture is to obtain inside plants and external information, guidance by advanced measurement means
It irrigates, the process of fertilising, physiology and high crop yield, efficient, degeneration-resistant crop production theoretical system.Inside plants letter
Breath is also known as plant physiology information, the wherein plant phenotype parameter in plant physiology information, be not all accurately measured always with
A specifications of surveys is formed, thus the phenotypic parameter of accurate quick obtaining plant is of great significance.
Traditional measurement plant phenotype parameter mainly obtains data by manual type, but this method damages plant life
Long and inefficiency.Currently, the method for research plant phenotype parameter mainly has based on image processing method and three-dimensional measurement side
Method, wherein optical three-dimensional measuring method becomes the method for three-dimensional measurement of mainstream.For example, passive vision measurement method and actively view
Feel that measurement method has been largely used to three-dimensional reconstruction and the phenotypic parameter measurement of plant.Passive vision measurement method has equipment
Simply, the features such as environmental restrictions are small, and human-computer interaction is few, but there are the problems such as computationally intensive, precision is low.Active vision measurement side
Method includes Depth Imaging method, laser ranging method and structural light measurement method.Depth Imaging method and laser ranging method can be obtained accurately
Take plant three-dimensional information, but there are equipment it is expensive, data volume is big, environmental disturbances the problems such as.Currently based on the three-dimensional of line-structured light
Measurement method has the characteristics that scanning real-time, high-precision, high stability is applied to plant three-dimensional measurement field.Feng Qingchun
It needs to be acquired after combining necessary rotating device to rotate plant to different angle Deng based on line-structured light measuring system,
Finally by the point cloud matching of each angle acquisition and splicing, the method has collection period long and acquisition blind zone problem, in addition by
Plant is stationary state in actual production, is only suitable for potted plant in this approach.Liu Jie etc. passes through twin camera and knot
Structure light integrated system is available to arrive the higher three-dimensional point cloud of precision, but can only collect the data of object visible face, and plant
The branches and leaves infall of object has more noise spot.
Summary of the invention
The object of the present invention is to provide a kind of plant phenotype parameter measuring apparatus, method and system, to improve phenotypic parameter
Measurement accuracy, reduce the complexity of measuring device.
To achieve the above object, the present invention provides following schemes:
A kind of plant phenotype parameter measuring apparatus, comprising: scanning head, test chassis, test platform and computer；Institute
It states scanning head to be fixed in the test chassis, the test chassis can move up and down；The test platform for bear to
Measure plant；The scanning head is for plant different height to be measured described in the moving sweep with the test chassis
Plant image, and by the plant image transmitting to the computer, the computer is used to be scanned according to the scanning head
Plant image determine the phenotypic parameter of the plant to be measured；
The scanning head includes multiple video cameras and two linear laser transmitters, and multiple positions for video camera are in the first water
Plane, multiple video camera equidistantly distributeds are on virtual circular arc；Two linear laser transmitters are located at described first
The second horizontal plane below horizontal plane, two linear laser transmitters tops respectively correspond two, virtual circular arc ends
The video camera of point；The third that the laser that two linear laser transmitters issue is located parallel to the test platform is horizontal
On face, the third horizontal plane and the first level face and second horizontal plane are parallel；The shooting angle of multiple video cameras
Degree is diagonally downward.
Optionally, further includes: motion control card and stepper motor, the test chassis and the stepper motor are and institute
Motion control card connection is stated, the motion state of the test chassis is adjusted by the stepper motor.
Optionally, further includes: multiple narrow band filters, the central wavelength of the narrow band filter are 532nm, Mei Gesuo
It states and is correspondingly arranged the narrow band filter on the camera lens of video camera.
A kind of plant phenotype measurement method of parameters, the plant phenotype measurement method of parameters are applied to plant table above-mentioned
Shape parameter measuring device, the plant phenotype measurement method of parameters include:
Obtain the plant image to be measured of each video camera shooting；
Laser stripe center line is extracted using the grey scale centre of gravity method of striation skeleton；
According to the calibrating parameters of each video camera, the three-dimensional point cloud under each camera angles is obtained；
The unification of all three-dimensional point clouds is spliced to global coordinate system, obtains initial plant three-dimensional point cloud to be measured；
The initial plant three-dimensional point cloud to be measured is projected to the OXY plane of the global coordinate system, plant to be measured is obtained
Strain two-dimensional projection point image；
By the stem edge in plant to be measured two-dimensional projection point image described in manual interaction, stem contour area is obtained；
According to the corresponding relationship of subpoint in three-dimensional point cloud and the global coordinate system, initial plant to be measured three is obtained
The stem point cloud sector domain of plant to be measured in dimension point cloud；
According to the Euclidean distance between the initial plant three-dimensional point cloud to be measured, the initial plant to be measured is determined
Multiple blade point cloud sectors domain of plant to be measured in three-dimensional point cloud；
The stem point cloud sector domain of the plant to be measured and blade point cloud sector domain are carried out using K mean cluster algorithm to be measured
The organ point cloud segmentation of plant obtains corresponding cloud sector of each organ of plant to be measured domain；
According to corresponding cloud sector of each organ of plant to be measured domain, the ginseng of each organ of plant to be measured is calculated
Number, obtains the phenotypic parameter of the plant to be measured.
Optionally, the plant image to be measured for obtaining each video camera shooting, before further include:
Internal reference and outer ginseng to each video camera are demarcated, and the calibrating parameters of each video camera are obtained；
Optic plane equations where the laser of linear laser transmitter transmitting are demarcated；
It, will be under the local coordinate system unification to global coordinate system of each video camera using global calibration method.
Optionally, the grey scale centre of gravity method using striation skeleton extracts laser stripe center line, specifically includes:
Using threshold filter method and gaussian filtering method is removed dryness to the plant image progress image to be measured and image binaryzation
Processing, obtains binary image；
Skeletal extraction is carried out to the binary image using morphology, determines the normal direction of every, skeleton；
Grey scale centre of gravity is calculated in each normal direction；
Grey scale centre of gravity corresponding in each normal direction is connected, the center line of laser stripe is obtained.
Optionally, described to splice the unification of all three-dimensional point clouds to global coordinate system, obtain initial plant to be measured
Three-dimensional point cloud, later further include:
The denoising of internal high frequency point cloud is carried out to the initial plant three-dimensional point cloud to be measured, the plant to be measured after being denoised
Strain three-dimensional point cloud.
A kind of plant phenotype parameter measurement system, the plant phenotype parameter measurement system are applied to plant table above-mentioned
Shape parameter measuring device, the plant phenotype parameter measurement system include:
Plant image collection module to be measured, for obtaining the plant image to be measured of each video camera shooting；
Laser stripe central line pick-up module, for extracting laser stripe center using the grey scale centre of gravity method of striation skeleton
Line；
Three-dimensional point cloud obtains module and obtains under each camera angles for the calibrating parameters according to each video camera
Three-dimensional point cloud；
Splicing module obtains initial plant to be measured for splicing the unification of all three-dimensional point clouds to global coordinate system
Strain three-dimensional point cloud；
Projection module is put down for projecting the initial plant three-dimensional point cloud to be measured to the OXY of the global coordinate system
Face obtains plant two-dimensional projection to be measured point image；
Manual interaction module, for passing through the stem edge in plant two-dimensional projection to be measured point image described in manual interaction,
Obtain stem contour area；
The stem point cloud sector domain of plant to be measured obtains module, for throwing according in three-dimensional point cloud and the global coordinate system
The corresponding relationship of shadow point obtains the stem point cloud sector domain of plant to be measured in initial plant three-dimensional point cloud to be measured；
The blade point cloud sector domain of plant to be measured obtains module, for according to the initial plant three-dimensional point cloud to be measured it
Between Euclidean distance, determine multiple blade point cloud sectors domain of plant to be measured in the initial plant three-dimensional point cloud to be measured；
Organ point cloud segmentation module, for using K mean cluster algorithm to the stem point cloud sector domain of the plant to be measured and
Blade point cloud sector domain carries out the organ point cloud segmentation of plant to be measured, obtains described corresponding cloud sector of each organ of plant to be measured
Domain；
Phenotypic parameter computing module, for according to corresponding cloud sector of each organ of plant to be measured domain, described in calculating
The parameter of each organ of plant to be measured, obtains the phenotypic parameter of the plant to be measured.
Optionally, further includes:
Camera calibration module, for each video camera internal reference and outer ginseng demarcate, obtain each video camera
Calibrating parameters；
Optic plane equations demarcating module, optic plane equations where laser for emitting linear laser transmitter into
Rower is fixed；
Coordinate system unified modules are unified to complete by the local coordinate system of each video camera for utilizing global calibration method
Under office's coordinate system.
Optionally, the laser stripe central line pick-up module, specifically includes:
Image processing unit, for carrying out figure to the plant image to be measured using threshold filter method and gaussian filtering method
It is handled as removing dryness with image binaryzation, obtains binary image；
Skeletal extraction unit determines the every point of skeleton for carrying out skeletal extraction to the binary image using morphology
Normal direction；
Grey scale centre of gravity computing unit, for calculating grey scale centre of gravity in each normal direction；
The center line acquiring unit of laser stripe is obtained for connecting grey scale centre of gravity corresponding in each normal direction
The center line of laser stripe.
The specific embodiment provided according to the present invention, the invention discloses following technical effects:
By the cooperation of multiple cameras and laser emitter, comprehensive scanning is carried out to plant from multiple angles, is avoided
There is blind area when splicing in data, which carries out that 3-D scanning is more efficient, strong applicability, at low cost to plant, and
Point cloud data is more complete, and it is also more preferable to rebuild effect.
Since green blade has absorption peak in feux rouges and blue wave band, so laser scanning is carried out using green (light) laser,
And in order to enhance the contrast of laser stripe, to obtain higher precision, central wavelength added on the camera lens of each camera
For the narrow band filter of 532nm, cable architecture Light stripes center extraction essence is extracted and improved which simplify line-structured light strip area
Degree.
In order to complete plant three dimensional structure separation, three-dimensional point is projected and generates two-dimensional projection's point and is clicked through based on two dimension
The network forming of row triangle and lighting process generate two-dimensional projection image, are partitioned into stem region based on two-dimensional projection image, then be based on two
The corresponding relationship of subpoint and three-dimensional point is tieed up, it is automatic to obtain stem point cloud sector domain and blade point cloud sector domain in three-dimensional point cloud, due to
Blade three-dimensional point cloud is segmented into different blade point cloud sector domains according to Euclidean distance, is then based on each separated region and adopts
Each of final cloud sector domain is obtained with K-means cluster segmentation.Three-dimensional computations are converted to two dimension and are calculated by whole process, drop
It is low direct to divide the complexity for carrying out organ separation using three-dimensional point cloud.
Three-dimensional point cloud segmentation is on the basis of obtaining based on stem three-dimensional point cloud region, to carry out different leaves point cloud sector domain minute
It cuts, then K-means cluster segmentation is carried out based on each point cloud sector domain divided.Entire three-dimensional point cloud segmentation is compared to without first
The point cloud segmentation of knowledge is tested, segmentation is more efficient, divides more acurrate.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be in embodiment
Required attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some realities of the invention
Example is applied, it for those of ordinary skill in the art, without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the structural schematic diagram of plant phenotype parameter measuring apparatus of the present invention；
Fig. 2 is the flow diagram of plant phenotype measurement method of parameters of the present invention；
Fig. 3 is the structural schematic diagram of plant phenotype parameter measurement system of the present invention；
Fig. 4 is angles of corn plant leaves piece length and width measurement figure in present invention specific implementation case.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
His embodiment, shall fall within the protection scope of the present invention.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
Fig. 1 is the structural schematic diagram of plant phenotype parameter measuring apparatus of the present invention.As shown in Figure 1, the plant phenotype ginseng
Number measuring device includes with flowering structure: scanning head, test chassis 3, test platform 4 and computer (not identifying in figure).It is described
Scanning head is fixed in the test chassis 3, and the test chassis 3 can move up and down；The test platform 4 is for bearing
Plant to be measured；The scanning head is for plant different height to be measured described in the moving sweep with the test chassis 3
Plant image, and by the plant image transmitting to the computer, the computer according to the scanning head for sweeping
The plant image retouched determines the phenotypic parameter of the plant to be measured.
The scanning head includes multiple video cameras and two linear lasers transmitter 2-1 and 2-2, multiple positions for video camera
In first level face, multiple video camera equidistantly distributeds are on virtual circular arc.In figure by taking 3 video cameras as an example, including take the photograph
Camera 1-1, video camera 1-2 and video camera 1-3.
Two linear laser transmitters are located at the second horizontal plane below the first level face, and two described linear sharp
The video camera of two endpoints of virtual circular arc is respectively corresponded above optical transmitting set, it is right above linear laser emitter 2-1 in figure
Answer video camera 1-1, corresponding video camera 1-3 above linear laser transmitter 2-2.The laser that two linear laser transmitters issue
It is located parallel on the third horizontal plane of the test platform, the third horizontal plane and the first level face and described the
Two horizontal planes are parallel.The shooting angle of multiple video cameras diagonally downward so that when what two linear laser transmitters issued swashs
When light is located at the top of plant to be measured, each video camera can take the image on plant top to be measured.Since green blade exists
Feux rouges and blue wave band have absorption peak, and linear laser transmitter can carry out laser scanning using green (light) laser.
The plant phenotype parameter measuring apparatus of the present embodiment further include: motion control card and stepper motor, the test machine
Frame 3 and the stepper motor are connect with the motion control card, adjust the test chassis 3 by the stepper motor
Motion state.The plant phenotype parameter measuring apparatus of the present embodiment can also include: multiple narrow band filters, the narrowband filter
The central wavelength of mating plate is 532nm, is correspondingly arranged the narrow band filter on the camera lens of each video camera.
Fig. 2 is the flow diagram of plant phenotype measurement method of parameters of the present invention.As shown in Fig. 2, the plant phenotype ginseng
Number measurement methods the following steps are included:
Step 100: obtaining the plant image to be measured of each video camera shooting.
Step 200: extracting laser stripe center line using the grey scale centre of gravity method of striation skeleton.
Step 300: according to the calibrating parameters of each video camera, obtaining the three-dimensional point cloud under each camera angles.
Step 400: the unification of all three-dimensional point clouds being spliced to global coordinate system, obtains initial plant to be measured three
Dimension point cloud.
Step 500: the initial plant three-dimensional point cloud to be measured being projected to the OXY plane of the global coordinate system, is obtained
To plant two-dimensional projection to be measured point image.
Step 600: by the stem edge in plant to be measured two-dimensional projection point image described in manual interaction, obtaining stem
Contour area.
Step 700: according to the corresponding relationship of subpoint in three-dimensional point cloud and the global coordinate system, obtain it is initial to
Survey the stem point cloud sector domain of plant to be measured in plant three-dimensional point cloud.
Step 800: according to the Euclidean distance between the initial plant three-dimensional point cloud to be measured, determining described initial
Multiple blade point cloud sectors domain of plant to be measured in plant three-dimensional point cloud to be measured.
Step 900: using K mean cluster algorithm to the stem point cloud sector domain and blade point cloud sector domain of the plant to be measured
The organ point cloud segmentation for carrying out plant to be measured obtains corresponding cloud sector of each organ of plant to be measured domain.
Step 1000: according to corresponding cloud sector of each organ of plant to be measured domain, it is each to calculate the plant to be measured
The parameter of organ obtains the phenotypic parameter of the plant to be measured.
It needs to demarcate video camera and optic plane equations parameter before step 100, detailed process is as follows:
Internal reference and outer ginseng to each video camera are demarcated, and the calibrating parameters of each video camera are obtained；
Optic plane equations where the laser of linear laser transmitter transmitting are demarcated；
It, will be under the local coordinate system unification to global coordinate system of each video camera using global calibration method.
Step 200 specifically includes:
Using threshold filter method and gaussian filtering method is removed dryness to the plant image progress image to be measured and image binaryzation
Processing, obtains binary image；
Skeletal extraction is carried out to the binary image using morphology, determines the normal direction of every, skeleton；
Grey scale centre of gravity is calculated in each normal direction；
Grey scale centre of gravity corresponding in each normal direction is connected, the center line of laser stripe is obtained.
After step 400 further include:
The denoising of internal high frequency point cloud is carried out to the initial plant three-dimensional point cloud to be measured, the plant to be measured after being denoised
Strain three-dimensional point cloud.
Fig. 3 is the structural schematic diagram of plant phenotype parameter measurement system of the present invention.As shown in figure 3, the plant phenotype ginseng
Number measuring system includes with flowering structure:
Plant image collection module 301 to be measured, for obtaining the plant image to be measured of each video camera shooting；
Laser stripe central line pick-up module 302, for being extracted in laser stripe using the grey scale centre of gravity method of striation skeleton
Heart line；
Three-dimensional point cloud obtains module 303 and obtains each camera angles for the calibrating parameters according to each video camera
Under three-dimensional point cloud；
Splicing module 304, for splicing to global coordinate system by all three-dimensional point clouds are unified, obtain it is initial to
Survey plant three-dimensional point cloud；
Projection module 305, for projecting the initial plant three-dimensional point cloud to be measured to the global coordinate system
OXY plane obtains plant two-dimensional projection to be measured point image；
Manual interaction module 306, for passing through the stem side in plant two-dimensional projection to be measured point image described in manual interaction
Edge obtains stem contour area；
The stem point cloud sector domain of plant to be measured obtains module 307, for according in three-dimensional point cloud and the global coordinate system
The corresponding relationship of subpoint obtains the stem point cloud sector domain of plant to be measured in initial plant three-dimensional point cloud to be measured；
The blade point cloud sector domain of plant to be measured obtains module 308, for according to the initial plant three-dimensional point cloud to be measured
Between Euclidean distance, determine multiple blade point cloud sectors domain of plant to be measured in the initial plant three-dimensional point cloud to be measured；
Organ point cloud segmentation module 309, for the stem point cloud sector using K mean cluster algorithm to the plant to be measured
Domain and blade point cloud sector domain carry out the organ point cloud segmentation of plant to be measured, obtain the corresponding point of each organ of plant to be measured
Cloud sector domain；
Phenotypic parameter computing module 3010, for calculating according to corresponding cloud sector of each organ of plant to be measured domain
The parameter of each organ of plant to be measured, obtains the phenotypic parameter of the plant to be measured.
Wherein, the laser stripe central line pick-up module 302 specifically includes:
Image processing unit, for carrying out figure to the plant image to be measured using threshold filter method and gaussian filtering method
It is handled as removing dryness with image binaryzation, obtains binary image；
Skeletal extraction unit determines the every point of skeleton for carrying out skeletal extraction to the binary image using morphology
Normal direction；
Grey scale centre of gravity computing unit, for calculating grey scale centre of gravity in each normal direction；
The center line acquiring unit of laser stripe is obtained for connecting grey scale centre of gravity corresponding in each normal direction
The center line of laser stripe.
Above system further include:
Camera calibration module, for each video camera internal reference and outer ginseng demarcate, obtain each video camera
Calibrating parameters；
Optic plane equations demarcating module, optic plane equations where laser for emitting linear laser transmitter into
Rower is fixed；
Coordinate system unified modules are unified to complete by the local coordinate system of each video camera for utilizing global calibration method
Under office's coordinate system.
The solution of the present invention is further illustrated below with reference to a specific implementation case.
This specific implementation case is used to measure the phenotypic parameter of plant.Measuring device is by 3 CCD industrial cameras, 3
A optical filter, 2 linear laser transmitters, test chassis, test platform, a computer composition.Using green (light) laser into
Row laser scanning, and in order to enhance the contrast of laser stripe, to obtain higher precision, add on the camera lens of each camera
The narrow band filter that central wavelength is 532nm is gone up.
Linear laser and video camera are fixed as scanning head, relative position remains constant, equidistantly solid respectively
Dingan County is mounted in 3 positions of rack.The linear laser that 2 lasers issue is maintained at same plane and is parallel to test platform, phase
Machine, which is tilted down, is kept fixed angle with laser.Rack is linked with motion control card, and stepper motor and motion control card connect
It connects, realizes that rack moves up and down by labview programming Control.
It before device starting, needs first to demarcate scanning head, the three-dimensional point under corresponding camera perspective is obtained with this
Coordinate information, while the overall situation is carried out to all three-dimensional point clouds by the positional relationship between camera and is calibrated under uniform coordinate, it completes
The splicing of entire three-dimensional point cloud.Specific calibration process is as follows:
1, camera (video camera) is demarcated
Camera internal reference calibration and usage plane gridiron pattern target carries out corresponding operation.The detailed process of camera calibration is as follows:
(1) scaling board placed into different location, different angle, in different positions shot, shoot 10 calibration in total
Image.
(2) to each calibration picture, sub-pix angle point information is extracted；(angle point is extracted, approximating method is then based on and asks
Sub-pixel location is taken, this is conventional existing method, is achieved that using computer vision kit calling)
(3) after getting chessboard calibration figure angle point image coordinate, according to camera calibration principle, calculate camera internal reference and
Outer parameter；
(4) the camera inside and outside parameter being calculated is advanced optimized, obtains optimal camera inside and outside parameter.Optimized
Journey is the prior art: since the camera inside and outside parameter that (3) are found out is an approximate solution, being carried out by Maximum-likelihood estimation non-
Linear optimization.Meanwhile putting plane target drone in corn growth plane, the image of plane target drone is obtained by camera.By aforementioned mark
Determine principle, that is, can determine and singly answer relational matrix between camera image plane and corn growth plane, it is thick for calculating plant height strain
Etc. growth planes information.
2, optic plane equations parameter calibration
(1) internal reference of above method calibration for cameras is utilized；
(2) calibration target is put into different location in space, so that projection light of the structure light in target plane
Item can be in camera image plane blur-free imaging；
(3) in the sub-pixel laser stripe for obtaining laser stripe image using the grey scale centre of gravity method based on striation skeleton
The heart
(4) according to structured light vision sensor measurement model and camera calibration matrix, determine optical plane in camera coordinates system
Under normal vector, and finally determine expressed intact equation of the optical plane under camera coordinates system；
(5) optical plane each in laser sensor is repeated the above process, that is, can determine that all light are flat in laser sensor
The expression equation of face in the camera.So far, being changed to optic plane equations parameter calibration terminates.
3, global calibration
Due between single camera coordinate system independently of one another, it is therefore desirable to using global calibration method by each camera
Under local coordinate system unification to global coordinate system, that is, each camera coordinates system is determined to the spin matrix of global coordinate system and is put down
Move matrix.
It is mounted with 3 groups of scanning heads in the implementation case, utilizes camera calibration method and line-structured light laser stripe image
Rebuild the plant three-dimensional point cloud under each camera visual angle.In order to which the three-dimensional point cloud that will acquire organic can be stitched together, this reality
Case is applied using No. 1 camera coordinates system as the frame of reference i.e. global coordinate system, then calibrates other camera coordinates system overall situations
To the frame of reference.Overall situation calibration is exactly the rotational translation matrix for calculating other camera coordinates systems to the frame of reference.
The transformational relation of each camera coordinates system are as follows:
Wherein: RnFor 3 × 3 spin matrixs between two cameras；TnFor 3 × 1 translation vectors between two cameras.
Using equation (1) by calculating the rotational translation matrix between camera two-by-two, determine that all camera coordinates systems arrive
The three dimensional point cloud obtained under each camera can then be carried out unification by the rotational translation matrix between global coordinate system.
After device starting, when step motor control rack and scanning head are at the uniform velocity scanned from top to bottom, linear laser
Device projects laser stripe to plant surface to be measured, and 3 cameras realize picture synchronization collection, acquisition gained by signal generator
Image data is sent into computer by network interface card and carries out real-time three-dimensional reconstruction, completes when probe to be scanned is from upper scanning to bottom
The three-dimensional reconstruction of entire plant is separated with organ point cloud.Detailed process is as follows:
Step 1: cable architecture Light stripes center extraction
Plant to be reconstructed is placed into test platform center, scanning head starts to scan from top to bottom, linear laser
Device projects laser striation to tested plant surface, and camera acquires the laser stripe image Jing Guo plant surface modulation.Due to each
Camera lens added the narrow band filter that central wavelength is 532nm respectively, so most of background quilt of camera acquisition image
It filters out, main target region is laser stripe.The implementation case line-structured light fringe center extracting method is mainly using based on light
The grey scale centre of gravity method of skeleton extracts optical losses, and main flow is as follows: carrying out figure first with threshold filter, gaussian filtering
As denoising and image binaryzation processing, the normal side of every, skeleton secondly is determined to bianry image skeletal extraction using morphology
To finally finding out grey scale centre of gravity in the normal direction, and the grey scale centre of gravity in normal direction everywhere is connected, just obtain
The center line of striation.
Step 2: plant three-dimensional reconstruction
1, plant three-dimensional point cloud obtains
Laser stripe central point is extracted using above-mentioned laser stripe extracting method, and utilizes each camera calibration parameter, then
The three-dimensional point cloud under each camera perspective can be obtained, then utilizes camera coordinates system global calibration method by all three-dimensional point clouds
It adjusts to unified coordinate system, completes the splicing of three-dimensional point cloud, obtain initial plant three-dimensional point cloud.
2, point cloud denoising
Image Acquisition is carried out based on polyphaser line-structured light integrated approach, completes point cloud using global calibration method,
Will cause splicing part to a certain extent has a cloud data redundancy problem.For this purpose, the application utilizes existing preferable solution party
Method carries out the removal of internal high frequency point cloud to the three-dimensional point cloud for having spliced completion.On the one hand it avoids going out during subsequent triangle network forming
Existing mistake, improves the precision of three-dimensional reconstruction；On the other hand the data volume for reducing three-dimensional point cloud, improves operation time.
Step 3: plant phenotypic parameter extracts
The plant three-dimensional point cloud that will acquire is projected to the OXY plane of global coordinate system, forms plant two-dimensional projection point
Image, the stem region of manual interaction two-dimensional projection point image obtain stem point using the projection relation of three-dimensional point and subpoint
Cloud approximate region；Finally, realizing each organ point cloud segmentation of plant using K mean cluster algorithm.Manual interaction is exactly to utilize
Left mouse button is clicked along stem zone boundary outline line, obtains stem contour area.
Three-dimensional point using two-dimensional projection's point is generated on camera matrix projection to two dimensional image, establish projection two-dimensional points with
The corresponding relationship of three-dimensional point, two-dimensional projection's point triangle network forming and lighting process generate two-dimension projection, pass through manual interaction two dimension
Stem edge in perspective view obtains stem contour area, and the two dimension being enclosed in profile is calculated according to stem contour area and is thrown
Shadow point has corresponding relationship according to two-dimensional projection's point and three-dimensional point, has been known that in three-dimensional point cloud which point cloud genera is in stem area
Domain；
According to obtained three-dimensional stem point cloud sector domain, three-dimensional point cloud can be divided into stem point cloud sector domain and blade point
Cloud sector domain, since blade three-dimensional point cloud is segmented into different leaves part (based on neighbor point space topological knot according to Euclidean distance
The dividing method of structure), each separated region, which is then based on, using K-means cluster segmentation obtains final each organ pair
The point cloud sector domain answered.
Step 4: calculating phenotypic parameter
1, plant height measures
It finds whole plant stem point cloud minimum point and Euclidean distance is sought in highest point.Choose a point P (x at the top of planti,yi,
zi) and one point Q (x of bottomj,yj,zj), then plant plant height are as follows:
2, stem thickness measures
Stem region point cloud is obtained after each organ point cloud segmentation of plant, is taken x, the y-coordinate of point cloud data, will be put cloud
Data projection obtains fitting result, elliptic equation to OXZ plane, using ellipse fitting method are as follows:
Ax2+Bxz+Cz2+ Dx+Ez+F=0 (3)
Stem thickness parameter extraction: being fitted the ellipse of acquisition, and long axis and short axle parameter can be equal to the stem thickness of corn, benefit
Elliptical center X can be acquired with equation group (4)c、Yc, equation group (5) can acquire elliptical major semiaxis L and semi-minor axis S.
3, blade area
Using triangulation to blade point cloud data network forming, the area of tri patch is calculated using Heron's formula, adds up three
The area of edged surface piece obtains spoon of blade area.Blade area calculation formula is as follows:
Formula (7) is Heron's formula, and wherein a, b, c are respectively Atria side length, and p is triangle semi-perimeter, SiIt is one
The area of a triangle, S is cumulative for all triangle areas in formula (8).
4, blade length and width measure
Blade point cloud is projected on the face XOY, carry out triangle network forming using triangulation methodology and adds lighting process.Benefit
Blade profile edge is extracted with image outline extraction algorithm and seeks rectangle minimum circumscribed rectangle, the length and width of minimum circumscribed rectangle
The as length and width of blade, as shown in figure 4, Fig. 4 is angles of corn plant leaves piece length and width measurement figure in present invention specific implementation case,
Wherein the part a is plant blade point cloud chart, and the part b is plant blade perspective view.In figure, W and H are respectively blade
Length and width.
The implementation case can be realized it is following the utility model has the advantages that
1, a kind of plant phenotypic parameter measuring system integrated based on multi-line structured light by more cameras and is swashed
The cooperation of optical transmitting set carries out comprehensive scanning to plant from three angles, blind area occurs when avoiding data splicing, the system
Carry out that 3-D scanning is more efficient, strong applicability, at low cost to plant, and point cloud data is more complete, rebuilds effect also more
It is good.
Since green blade has absorption peak in feux rouges and blue wave band, so we select green (light) laser to carry out laser
Scanning, and in order to enhance the contrast of laser stripe, to obtain higher precision, we add on the camera lens of each camera
Central wavelength is the narrow band filter of 532nm, extracts and improve cable architecture striation which simplify line-structured light strip area
Center extraction precision；
3, in order to complete plant three dimensional structure separation, three-dimensional point is projected and generates two-dimensional projection's point and based on two-dimensional points
It carries out triangle network forming and lighting process generates two-dimensional projection image, stem region is partitioned into based on two-dimensional projection image, then be based on
The corresponding relationship of two-dimensional projection point and three-dimensional point, it is automatic to obtain stem point cloud sector domain and blade point cloud sector domain in three-dimensional point cloud, by
Different leaves point cloud sector domain is segmented into according to Euclidean distance in blade three-dimensional point cloud, each separated region is then based on and adopts
Each of final cloud sector domain is obtained with K-means cluster segmentation.Three-dimensional computations are converted to two dimension and are calculated by whole process, drop
It is low direct to divide the complexity for carrying out organ separation using three-dimensional point cloud.
4, three-dimensional point cloud segmentation is on the basis of obtaining based on stem three-dimensional point cloud region, to carry out different leaves point cloud sector domain
Segmentation, then K-means cluster segmentation is carried out based on each point cloud sector domain divided.Entire three-dimensional point cloud segmentation, which is compared, not to be had
The point cloud segmentation of priori knowledge, segmentation is more efficient, divides more acurrate.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with its
The difference of his embodiment, the same or similar parts in each embodiment may refer to each other.For being disclosed in embodiment
For system, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method portion
It defends oneself bright.
Used herein a specific example illustrates the principle and implementation of the invention, above embodiments
Illustrate to be merely used to help understand method and its core concept of the invention；At the same time, for those skilled in the art,
According to the thought of the present invention, there will be changes in the specific implementation manner and application range.In conclusion this specification
Content should not be construed as limiting the invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910135784.1A CN109708578B (en) | 2019-02-25 | 2019-02-25 | Device, method and system for measuring plant phenotype parameters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910135784.1A CN109708578B (en) | 2019-02-25 | 2019-02-25 | Device, method and system for measuring plant phenotype parameters |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109708578A true CN109708578A (en) | 2019-05-03 |
| CN109708578B CN109708578B (en) | 2020-07-24 |
Family
ID=66263758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910135784.1A Active CN109708578B (en) | 2019-02-25 | 2019-02-25 | Device, method and system for measuring plant phenotype parameters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109708578B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110470241A (en) * | 2019-08-19 | 2019-11-19 | 天津大学 | A kind of refractory brick curvature detection system and method based on structure light vision |
| CN110849262A (en) * | 2019-10-17 | 2020-02-28 | 中国科学院遥感与数字地球研究所 | A kind of vegetation phenotype structure parameter measurement method, device and system |
| CN111487646A (en) * | 2020-03-31 | 2020-08-04 | 安徽农业大学 | A method for online detection of corn plant morphology |
| CN111578837A (en) * | 2020-04-30 | 2020-08-25 | 北京农业智能装备技术研究中心 | Plant shape visual tracking measurement method for agricultural robot operation |
| CN111696122A (en) * | 2020-06-12 | 2020-09-22 | 北京数字绿土科技有限公司 | Crop phenotype parameter extraction method and device |
| CN111768413A (en) * | 2020-05-29 | 2020-10-13 | 北京农业信息技术研究中心 | A method and system for 3D point cloud segmentation of plants |
| CN112036407A (en) * | 2020-07-30 | 2020-12-04 | 北京农业信息技术研究中心 | Plant population three-dimensional phenotype in-situ acquisition method and device |
| CN112212801A (en) * | 2020-09-30 | 2021-01-12 | 水利部牧区水利科学研究所 | A sand-fixing shrub configuration data processing system |
| CN112595243A (en) * | 2020-12-02 | 2021-04-02 | 中国科学院空天信息创新研究院 | Automatic vegetation plant height measuring method and system suitable for field continuous observation |
| CN112930926A (en) * | 2021-01-28 | 2021-06-11 | 电凯(常州)新能源技术研究院有限公司 | Plant production parameter optimization method based on deep learning |
| WO2021212844A1 (en) * | 2020-04-21 | 2021-10-28 | 广东博智林机器人有限公司 | Point cloud stitching method and apparatus, and device and storage device |
| CN113822896A (en) * | 2021-08-31 | 2021-12-21 | 北京市农林科学院信息技术研究中心 | Plant population three-dimensional phenotype data acquisition device and method |
| CN113920106A (en) * | 2021-10-29 | 2022-01-11 | 吉林农业大学 | Corn growth three-dimensional reconstruction and stem thickness measurement method based on RGB-D camera |
| CN114792333A (en) * | 2022-04-12 | 2022-07-26 | 深圳市格灵精睿视觉有限公司 | Light bar matching method, device and system and storage medium |
| CN114812418A (en) * | 2022-04-25 | 2022-07-29 | 安徽农业大学 | Portable plant density and plant spacing measurement system |
| CN115266490A (en) * | 2022-07-13 | 2022-11-01 | 交通运输部公路科学研究所 | Machine-made sand particle three-dimensional shape testing method and system based on multi-view scanning |
| CN119295564A (en) * | 2024-12-16 | 2025-01-10 | 浙江托普云农科技股份有限公司 | Plant overall measurement calibration method, system and device based on computer vision |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004095071A3 (en) * | 2003-04-17 | 2006-08-03 | Kenneth Sinclair | Object detection system |
| CN101509764A (en) * | 2009-02-27 | 2009-08-19 | 东南大学 | Method for rapidly acquiring object three-dimensional form |
| CN103106684A (en) * | 2013-01-15 | 2013-05-15 | 北京农业信息技术研究中心 | Method and system of three-dimensional reconstruction of shape structure of trees in leaved state |
| CN103940369A (en) * | 2014-04-09 | 2014-07-23 | 大连理工大学 | Quick morphology vision measuring method in multi-laser synergic scanning mode |
| EP2848121A1 (en) * | 2013-09-13 | 2015-03-18 | Palo Alto Research Center Incorporated | Unwanted laser plant removal system having a stabilization system |
| CN104484873A (en) * | 2014-11-28 | 2015-04-01 | 浙江大学 | Method for separating leaves and stem in three-dimensional point cloud of plant |
| CN105675549A (en) * | 2016-01-11 | 2016-06-15 | 武汉大学 | Portable crop parameter measurement and growth vigor intelligent analysis device and method |
| CN105844244A (en) * | 2016-03-28 | 2016-08-10 | 北京林业大学 | Fruit tree ratoon identifying and positioning method |
| US20170053168A1 (en) * | 2015-08-21 | 2017-02-23 | Murgyver Consulting Ltd. | Method and system for the optimization of plant growth |
| CN107869962A (en) * | 2017-10-31 | 2018-04-03 | 南京农业大学 | A high-throughput facility crop 3D shape information measurement system based on space sculpting technology |
| CN108981569A (en) * | 2018-07-09 | 2018-12-11 | 南京农业大学 | A kind of high-throughput hothouse plants phenotype measuring system based on the fusion of multispectral cloud |
-
2019
- 2019-02-25 CN CN201910135784.1A patent/CN109708578B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004095071A3 (en) * | 2003-04-17 | 2006-08-03 | Kenneth Sinclair | Object detection system |
| CN101509764A (en) * | 2009-02-27 | 2009-08-19 | 东南大学 | Method for rapidly acquiring object three-dimensional form |
| CN103106684A (en) * | 2013-01-15 | 2013-05-15 | 北京农业信息技术研究中心 | Method and system of three-dimensional reconstruction of shape structure of trees in leaved state |
| EP2848121A1 (en) * | 2013-09-13 | 2015-03-18 | Palo Alto Research Center Incorporated | Unwanted laser plant removal system having a stabilization system |
| CN103940369A (en) * | 2014-04-09 | 2014-07-23 | 大连理工大学 | Quick morphology vision measuring method in multi-laser synergic scanning mode |
| CN104484873A (en) * | 2014-11-28 | 2015-04-01 | 浙江大学 | Method for separating leaves and stem in three-dimensional point cloud of plant |
| US20170053168A1 (en) * | 2015-08-21 | 2017-02-23 | Murgyver Consulting Ltd. | Method and system for the optimization of plant growth |
| CN105675549A (en) * | 2016-01-11 | 2016-06-15 | 武汉大学 | Portable crop parameter measurement and growth vigor intelligent analysis device and method |
| CN105844244A (en) * | 2016-03-28 | 2016-08-10 | 北京林业大学 | Fruit tree ratoon identifying and positioning method |
| CN107869962A (en) * | 2017-10-31 | 2018-04-03 | 南京农业大学 | A high-throughput facility crop 3D shape information measurement system based on space sculpting technology |
| CN108981569A (en) * | 2018-07-09 | 2018-12-11 | 南京农业大学 | A kind of high-throughput hothouse plants phenotype measuring system based on the fusion of multispectral cloud |
Non-Patent Citations (3)
| Title |
|---|
| MANUEL VÁZQUEZ-ARELLANO等: "Determination of stem position and height of reconstructed maize plants using a time-of-flight camera", 《COMPUTERS AND ELECTRONICS IN AGRICULTURE》 * |
| 杜建军等: "基于穗粒分布图的玉米果穗表型性状参数计算方法", 《农业工程学报》 * |
| 高宇等: "植物表型监测技术研究进展及发展对策", 《江苏农业科学》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110470241A (en) * | 2019-08-19 | 2019-11-19 | 天津大学 | A kind of refractory brick curvature detection system and method based on structure light vision |
| CN110849262A (en) * | 2019-10-17 | 2020-02-28 | 中国科学院遥感与数字地球研究所 | A kind of vegetation phenotype structure parameter measurement method, device and system |
| CN111487646A (en) * | 2020-03-31 | 2020-08-04 | 安徽农业大学 | A method for online detection of corn plant morphology |
| WO2021212844A1 (en) * | 2020-04-21 | 2021-10-28 | 广东博智林机器人有限公司 | Point cloud stitching method and apparatus, and device and storage device |
| CN111578837A (en) * | 2020-04-30 | 2020-08-25 | 北京农业智能装备技术研究中心 | Plant shape visual tracking measurement method for agricultural robot operation |
| CN111768413A (en) * | 2020-05-29 | 2020-10-13 | 北京农业信息技术研究中心 | A method and system for 3D point cloud segmentation of plants |
| CN111768413B (en) * | 2020-05-29 | 2023-12-05 | 北京农业信息技术研究中心 | A plant three-dimensional point cloud segmentation method and system |
| CN111696122A (en) * | 2020-06-12 | 2020-09-22 | 北京数字绿土科技有限公司 | Crop phenotype parameter extraction method and device |
| CN112036407B (en) * | 2020-07-30 | 2024-02-06 | 北京农业信息技术研究中心 | Plant population three-dimensional phenotype in-situ acquisition method and device |
| CN112036407A (en) * | 2020-07-30 | 2020-12-04 | 北京农业信息技术研究中心 | Plant population three-dimensional phenotype in-situ acquisition method and device |
| CN112212801A (en) * | 2020-09-30 | 2021-01-12 | 水利部牧区水利科学研究所 | A sand-fixing shrub configuration data processing system |
| CN112595243A (en) * | 2020-12-02 | 2021-04-02 | 中国科学院空天信息创新研究院 | Automatic vegetation plant height measuring method and system suitable for field continuous observation |
| CN112595243B (en) * | 2020-12-02 | 2022-05-17 | 中国科学院空天信息创新研究院 | A method and system for automatic determination of plant height of vegetation suitable for continuous observation in the field |
| CN112930926A (en) * | 2021-01-28 | 2021-06-11 | 电凯(常州)新能源技术研究院有限公司 | Plant production parameter optimization method based on deep learning |
| CN113822896A (en) * | 2021-08-31 | 2021-12-21 | 北京市农林科学院信息技术研究中心 | Plant population three-dimensional phenotype data acquisition device and method |
| CN113822896B (en) * | 2021-08-31 | 2024-04-23 | 北京市农林科学院信息技术研究中心 | A plant population three-dimensional phenotypic data acquisition device and method |
| CN113920106A (en) * | 2021-10-29 | 2022-01-11 | 吉林农业大学 | Corn growth three-dimensional reconstruction and stem thickness measurement method based on RGB-D camera |
| CN113920106B (en) * | 2021-10-29 | 2024-05-10 | 吉林农业大学 | Corn growth vigor three-dimensional reconstruction and stem thickness measurement method based on RGB-D camera |
| CN114792333A (en) * | 2022-04-12 | 2022-07-26 | 深圳市格灵精睿视觉有限公司 | Light bar matching method, device and system and storage medium |
| CN114812418B (en) * | 2022-04-25 | 2023-10-27 | 安徽农业大学 | A portable plant density and spacing measurement system |
| CN114812418A (en) * | 2022-04-25 | 2022-07-29 | 安徽农业大学 | Portable plant density and plant spacing measurement system |
| CN115266490A (en) * | 2022-07-13 | 2022-11-01 | 交通运输部公路科学研究所 | Machine-made sand particle three-dimensional shape testing method and system based on multi-view scanning |
| CN119295564A (en) * | 2024-12-16 | 2025-01-10 | 浙江托普云农科技股份有限公司 | Plant overall measurement calibration method, system and device based on computer vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109708578B (en) | 2020-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109708578A (en) | Device, method and system for measuring plant phenotype parameters | |
| CN110288642B (en) | Three-dimensional object rapid reconstruction method based on camera array | |
| CN108269279B (en) | Three-dimensional reconstruction method and device based on monocular 3 D scanning system | |
| CN106091984B (en) | A method for acquiring 3D point cloud data based on line laser | |
| CN114998499B (en) | Binocular three-dimensional reconstruction method and system based on line laser galvanometer scanning | |
| US6781618B2 (en) | Hand-held 3D vision system | |
| CN107063129B (en) | A kind of array parallel laser projection three-dimensional scan method | |
| CN109887020B (en) | Plant organ separation method and system | |
| EP2313737B1 (en) | System for adaptive three-dimensional scanning of surface characteristics | |
| CN110243307B (en) | An automated three-dimensional color imaging and measurement system | |
| CN110230979B (en) | A three-dimensional target and a three-dimensional color digital system calibration method thereof | |
| CN113446957B (en) | Three-dimensional contour measuring method and device based on neural network calibration and speckle tracking | |
| CN106802138A (en) | A kind of 3 D scanning system and its scan method | |
| CN108151671A (en) | A kind of 3 D digital imaging sensor, 3 D scanning system and its scan method | |
| US20100328308A1 (en) | Three Dimensional Mesh Modeling | |
| CN110246186A (en) | A kind of automatized three-dimensional colour imaging and measurement method | |
| CN111060006B (en) | A viewpoint planning method based on three-dimensional model | |
| WO2022078442A1 (en) | Method for 3d information acquisition based on fusion of optical scanning and smart vision | |
| CN113566733B (en) | A line laser vision three-dimensional scanning device and method | |
| CN112257537B (en) | Intelligent multi-point three-dimensional information acquisition equipment | |
| CN104331924A (en) | Three-dimensional reconstruction method based on single camera SFS algorithm | |
| CN107633518A (en) | A kind of product design detection method based on Kinect | |
| CN113012238B (en) | Method for quick calibration and data fusion of multi-depth camera | |
| CN110425983A (en) | A kind of monocular vision three-dimensional reconstruction distance measuring method based on polarization multi-spectrum | |
| CN112254638B (en) | Intelligent visual 3D information acquisition equipment that every single move was adjusted |
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 |