CN100487362C - In-situ measuring method of plant root system geometry - Google Patents

In-situ measuring method of plant root system geometry Download PDF

Info

Publication number
CN100487362C
CN100487362C CNB2005101264972A CN200510126497A CN100487362C CN 100487362 C CN100487362 C CN 100487362C CN B2005101264972 A CNB2005101264972 A CN B2005101264972A CN 200510126497 A CN200510126497 A CN 200510126497A CN 100487362 C CN100487362 C CN 100487362C
Authority
CN
China
Prior art keywords
root system
plant
image
root
dimensional
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.)
Active
Application number
CNB2005101264972A
Other languages
Chinese (zh)
Other versions
CN1776357A (en
Inventor
罗锡文
周学成
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China Agricultural University
Original Assignee
South China Agricultural University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China Agricultural University filed Critical South China Agricultural University
Priority to CNB2005101264972A priority Critical patent/CN100487362C/en
Publication of CN1776357A publication Critical patent/CN1776357A/en
Application granted granted Critical
Publication of CN100487362C publication Critical patent/CN100487362C/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The method includes steps: using imaging device of biomedicine obtains layered sequence images in situ of root system of plant grown at different opaque medium. Processing and analyzing the sequence images obtains spatial geometrical characteristic parameters in situ of root system of plant. Based on the said parameters, model of solid geometry is built. 3D visualized reconstruction is carried out for model of solid geometry so as to build 3D display model of root system. Features are: obtaining images in situ of root system of plant precisely and in non-destructive, sampling procedure is simplified so as to save labor to separate soil, greatly raised accuracy.

Description

A kind of in-situ measuring method of root system of plant geometric configuration
Technical field
The invention belongs to the Automatic Measurement Technique field, particularly a kind of original position of root system of plant geometric configuration, non-destructive measuring method.
Background technology
Root system is plant obtains nutrient and moisture from media environments such as soil vitals.Plant is drawn the size of nutrient, moisture ability, depends on its root system situation that form distributes in media environment to a great extent.Therefore, just begun systematic study work as far back as 18th century people to root system of plant.Media such as soil are that plant relies the underground environment according to existence, and they also give the home position observation of root system and directly measure to be provided with the barrier that is difficult to go beyond when providing nutrient, moisture to root system of plant.Growing environment (opaque, invisible) that root system of plant is special and complicated morphosis thereof (interlaced, block) make people be far longer than the plant aerial part its measurement, research difficulty.In order to obtain the form distribution situation of root system of plant in soil space, traditional research method was divided into for two steps usually and carries out: sampling and measuring.The problem that at first needs to solve be exactly how complete, carry out in-situ sampling exactly.From the needs of Root morphology research, not only the root system obtained of requirement is complete, and the main root system that relates to research purpose at least must not breakthrough; And, also should keep its original state in media environment indeformable, be not shifted, require sampling work fast, accurately.In the root system research another one basic, critical work is exactly that the related parameter that has of root system is measured accurately.The parameter of root growth and distribution is described, because of different research purposes, different root configuration require differently, commonly used have root length, radical, root surface area, root volume, root footpath, root approximate number, root system intensity, root weight, root density, root angle, a lateral root spacing etc.Because root system complexity, special morphosis, carrying out accurately to it, parametric measurement is the work that difficulty is very big equally.The root structure branch is various, and thickness differs, and blocks mutually, intersects, soft movability, and easily dehydration distortion behind soil.Therefore, carry out accurately the height of big, the numerous and diverse degree of its difficulty of in-site detecting to root system well imagines.
Existing root system observation procedure can be divided into two big classes: disruptive method and non-destructive method substantially.Disruptive method is planted method, mesh bag method etc. as nail-plate method, container method, air culture method, pipe, often needs to carry out the destructiveness sampling, and root system and soil are separated; Carry out manual measurement then, or item by item relevant morphological parameters (majority is a panel data) is carried out artificial subsidiary by means of instrumentation, whole sampling and measuring process not only time and effort consuming, but also exist than mistake, be difficult to realize the in-situ sampling of root system and accurately measure (displacement is arranged, be completely cured and problem such as moisture ease mistake).Existing non-destructive method, comprise the research of tracer method, subterraneous root chamber and little chamber etc., although do not need destructive sampling, avoided sample procedure meticulous, time and effort consuming, but can only obtain some limited in-situ observation data, be difficult to obtain complete, three-dimensional configuration data accurately.In addition, underground root system research department is expensive very big, and observation process still needs the long time and a large amount of artificial, and also there are bigger problems such as irradiation hazard in isotope method.In fact, shortage accurately, the roots in situ observation procedure has constituted the technical bottleneck in root system research field easily, makes the research of Root morphology structure lag behind the plant shoot branch greatly.
In recent years, along with the fast development of modern science and technology and pressing for of root system scrutiny, new and high technology means are penetrated into root system observational study field gradually, and have obtained some impressive progresses.
For speed and precision, minimizing manually-operated and the personal error that improves the root system parameter measurement, some automatic Measurement and analysis methods based on computer image processing technology have appearred in recent years.(multiple collecting devices such as X-RAY) and sweep optical square obtain the two-dimensional image data of root system at first to adopt charge coup (CCD), X line machine, utilize image processing techniques that the root structure form is measured and analyzed then on computers, be mainly used in the root system plane configuration.Applicant of the present invention once presided over " application of digital image processing techniques in Root morphology is analyzed " problem of finishing and just belonged to this technology category, this project has been set up Root morphology parametric measurement and conformational analysis system based on image processing techniques, realized that the Root morphology parameter measures fast and accurately and analyze, for the research of root system two dimensional surface configuration provides a kind of efficient, instrument reliably.The correlation technique achievement drops into practical application gradually at present, as the DT-SCAN image analysis system of Australian ICT Co., CI-400/600 image analysis system of U.S. CID company and the WinRHIZO image analysis system of France etc.
The introducing of computer image processing technology, although improved speed and the precision that root system is measured effectively, and alleviated greatly manual measurement workload, saved Measuring Time, but still exist some shortcomings.On the one hand, the sample before soil sampling and measurement of root system still need to be prepared a large amount of artificial, and what is more important, this technology can only record the two dimensional configurations data of root system.In order directly to obtain the 3-d modelling parameter of sampling back root system, experts and scholars both domestic and external have made multiple trial, can be divided into contact measurement technology and noncontact measurement two class situations substantially.
Directly the contact measurement technology is based on the root system of plant node of contact-type 3 D digitized instruments such as mechanical type, electromagnetic type after to sampling and carries out the point-to-point measurement location, and records each node corresponding construction parameter (as diameter etc.); Then each node coordinate and parameter manually or are automatically imported computing machine, node data is carried out a series of calculating, processing, and analyze the 3-d modelling parameter that provides root system by computing machine.F.Danjon as France, H.Sinoquet, people such as F.Colin andM.Drexhage adopt the 3D digitizer of a kind of 3SPACE Fastrak by name and a kind of AMAPmod plant architecture special software by name respectively the pine tree (Pinus Pinaster) of the Oak Tree (Quercus Petraea) of 3 lifes in 20~28 years and 30 lifes in 5 years to be carried out the Measurement and analysis of root configuration.3SPACE Fastrak can carry out contact (click) to the three dimensional space coordinate of root system and measure.The AMAPmod special software then is used for the feature of analyzing evaluation root configuration, comprises locus, root length, volume, branch's order, branch's pattern etc., and can carry out the reconstruction of root system 3D rendering with the check measurement error according to the data that record.The research work at first utilizes large excavator that tree root is pulled up fast, this obviously can cause being completely cured and the displacement problem, mainly the structure root more than the 2mm (structural root) is carried out the pointwise determination and analysis then, the node number that average every tree needs contact measurement is up to 757, and maximum Oak Tree node numbers have surpassed 1475.Obviously the measuring process of contact-type 3 D measuring technique still be unable to do without a large amount of artificial participations.
Contactless 3D automatic measurement technology is to be core with the computing machine, based on principles such as acoustics, optics, magnetics exploitations, can gather the object space coordinate quickly and accurately, measuring process does not need artificial auxiliary.This technology has more application in the research that plant shoot divides, use then seldom report in the root structure morphological Study.Agricultural University Of South China root system biological study center attempts by the machine vision technique based on CCD the root system of plant of rotation being taken continuously, and utilize computing machine to carry out three-dimensional reconstruction the two dimensional image of gathering, obtained bigger progress, be expected to the three-dimensional configuration parameter of sparse root system is realized non-cpntact measurement.Yet also there are two weakness that are difficult to overcome in this technology, and has limited it and has measured and the application of research at root system.The one, when adopting this technology, root system must be taken a sample from soil, thereby this destructive sampling mode must be difficult to keep the precision of the original position state of the root system-later stage is measured to lose practical significance; The 2nd, for dense, tiny root system, crop root especially, interlock, block between its root system phenomenon very serious-this has proposed a fatal difficult problem just for stereoscopic vision measurement technology.
Summary of the invention
(1) technical matters that will solve
The objective of the invention is to overcome above-mentioned the deficiencies in the prior art, a kind of in-situ measuring method of root system of plant geometric configuration is provided.The growth distribution state of root system of plant in opaque medium not only can accurately, nondestructively be obtained and show to this method, and can carry out quantitative measurement and analysis to the original position geometric parameter and the form distribution situation of root system of plant.
(2) technical scheme
For achieving the above object, the present invention adopts following technical scheme:
The in-situ measuring method of root system of plant geometric configuration of the present invention comprises the steps:
(1) adopt biomedical imaging equipment to obtain the original position tomography sequence image that is grown in root system of plant in the opaque medium;
(2) the original position space geometry characteristic parameter that draws root system of plant is handled and analyzed to the sequence image of gained in the step (1), wherein, the treatment step that sequence image is carried out data processing mainly comprises image noise reduction, figure image intensifying, image segmentation and feature extraction, and, described image segmentation adopts the three-dimensional segmentation method of man-machine interactive, and makes up solid geometric pattern based on described characteristic parameter;
(3) solid geometric pattern according to step (2) carries out the three-dimensional visualization reconstruction, make up the 3-D display model of root system, and, described 3-D display model, adopt stepping cube method in the voxel level method for reconstructing at first to make up the three-dimensional contour surface model of root system of plant, generate in conjunction with illumination model and the visual angle factor then.
(4) Visualization Model of utilizing step (3) to make up is a Man Machine Interface, and the structure parameters of root system of plant assigned address or the degree of depth and root system integral body is implemented to measure and analyze.
Wherein said biomedical imaging equipment is X computer on line tomographic apparatus or MR imaging apparatus.
The advantage and the good effect of the in-situ measuring method of root system of plant geometric configuration of the present invention are:
1. method of the present invention need not to compare with traditional sampling technique method from the soil sampling, has not only saved the hand digging of wasting time and energy, cautious native detachment process, and can guarantee the integrality and the original position of root system.Here so-called integrality is meant that there is not damage in the root system of plant that adopts technical method gained of the present invention and destructive problem such as fracture.So-called original position then is meant and adopts technical method of the present invention can avoid phenomenons such as the displacement that occurs in the root system sampling process and distortion, thereby the space distribution state of root system in medium can not change before and after guaranteeing to take a sample.
2. compare with existing destructive sampling method, method of the present invention utilizes biomedical imaging technological means such as XCT, MRI to obtain the original position faultage image of root system of plant, belongs to noncontact, harmless sampling (collection) technology.Therefore, the present invention not only can not destroy the integrality of root system, and even can not influence the normal growth growth course of root system.Like this, adopt technical method provided by the invention not only can obtain the more static geometric configuration of in-situ root system, and can observe the dynamic process that root system of plant grows, the Three-Dimensional Dynamic geometric configuration that obtains root system in medium, this is that all art methods all are difficult to realize at all at present.
3. compare with art methods, adopt method of the present invention that the measurement of root system structure parameters is not only automatically, and be three-dimensional.Traditional root system measuring method measuring process is fully manually carried out, time and effort consuming not only, and can bring personal error for measurement result inevitably, influence the accuracy of measurement data.And just plane (two dimension) structure parameters or part (top layer) geometric parameter that adopt flat scanning and machine vision technique to record.In addition,, there is bigger limitation, is difficult to obtain complete, 3-d modelling parameter (as root footpath size, surface area etc.) accurately, and exists bigger radiation risk although isotope method can obtain certain 3-d modelling parameter.By contrast, that technical method provided by the invention then can record automatically is complete, in-situ three-dimensional structure parameters accurately, and this is that any existing technical method all is difficult to do at present.
Description of drawings
The process flow diagram of the in-situ measuring method of Fig. 1 a kind of root system of plant geometric configuration of the present invention;
The original position tomography sequence image that Fig. 2 for the banyan shoot root that adopts 4 row's spiral CT equipment to collect is;
The original position stereopsis that the banyan shoot root that Fig. 3 generates for three-dimensional reconstruction is.
Embodiment
Following examples are used to illustrate the present invention, but are not used for limiting protection scope of the present invention.
Realize that overall plan of the present invention is: at first adopt biomedical imaging technological means such as XCT/MRI to obtain the original position faultage image of root system of plant in the media such as being grown in soil, then utilize computer image processing technology that original position tomography sequence image is carried out suitable noise reduction, cuts apart the supervisor processing, and therefrom extract root system of plant space geometry characteristic parameter; Then, utilize of Geometric Modeling, visual reconstruction and the stereo display of the geometrical characteristic parameter of gained by computer graphics disposal technology realization root system of plant; Display model with root system of plant is man-machine interface platform at last, root system is carried out virtual, mutual measurement and analysis, and the geometric parameter of outputting plant root system appointed part, the distribution of specific level and the geometric parameter and the form distribution situation of root system integral body.Overall plan of the present invention relates to three joint sport technique segments, is respectively original position faultage image acquisition technique, the sequence image processing of original position tomography and the analytical technology of root system of plant and based on the three-dimensional visualization reconstruction technique of root system space geometry feature.
The root system of plant original position faultage image acquisition technique that the present invention relates to is meant that modern biomedical imaging techniques such as adopting XCT/MRI and equipment obtain the technological means of the tomography sequence image that is grown in root system of plant in the opaque medium.The equipment of this means indication had both comprised the medical imaging devices that employing is existing and developing listing, as 64 layers or more high-precision spiral CT equipment, 3T or stronger magnetic resonance imaging (MRI) equipment, also comprise the specialized equipment of development such as the biomedical imaging know-why that adopts XCT or MRI.The technology of this means indication mainly is meant the tomography technology scheme that is grown in specific root system in the particular medium that is applicable to, as imaging technique parameter of MSCT etc.
The original position tomography sequence image that the present invention relates to is handled and analytical technology, is meant the Flame Image Process and analytical algorithm and the program that are applicable to roots in situ tomography sequence image.This algorithm and program can be present existing suitable algorithm and program, also can be according to the special development of the space geometry feature of the characteristics of root system tomography sequence image and root system of plant, particularly including analysis extraction algorithm and program based on the image segmentation algorithm and the root system of plant space geometry characteristic parameter of root system space geometry feature.Here the basic task of the Flame Image Process of indication and analytical technology is to provide the original position space geometry characteristic parameter of root system of plant, comprises the important parameter such as characteristic dimension, branch directions angle of basic point position, branch node coordinate, the node region of root system.
The three-dimensional visualization reconstruction technique of the root system of plant that the present invention relates to is meant generating algorithm and program based on the solid geometric pattern and the three-dimensional visualization model of root system of plant space geometry characteristic parameter.This algorithm and program can be at present existing suitable algorithm and program, also can be to rebuild its geometric model and display model and special development according to the space geometry characteristic parameter of root system of plant.Here the basic task of the three-dimensional visualization reconstruction technique of indication is to generate the model space geometric and the 3-D display model of root system of plant, for further man-machine interaction Measurement and analysis is built interface platform.
In site measurement of root system of plant geometric configuration and the quantified system analysis that the present invention relates to are the ways of realization of the technology of the present invention method, are made of hardware system and software systems two parts.Hardware system comprises ingredients such as original position faultage image collecting device, computing machine, display, printer and view data special purpose memory devices.Software systems are made up of image capture module (supporting with the special image collecting device), image processing module, three-dimensional reconstruction module, display module, Measurement and analysis output module and the system integration (framework) module supervisor module.Software systems are to utilize above-mentioned theory of algorithm achievement in research, adopt the exploitation of C++ or Matlab language system to form.Total system is a platform with the three-dimensional in-situ root system that computer reconstruction generates, need in conjunction with root morphology, biological research, adopt man-machine interaction mode, extract and analyze required root system three-dimensional configuration parameter, and the geometric parameter of outputting plant root system appointed part, the distribution of specific level and the geometric parameter and the form distribution situation of root system integral body, and generate the Measurement and analysis result automatically.
Embodiment 1
Referring to Fig. 1.The in-situ measuring method of root system of plant geometric configuration of the present invention specifically comprises the steps:
With the banyan seedling is that example describes.
(1) selected potted plant root system sample was planted more than 1 week, carried out suitable encapsulation process, guaranteed scanning process safety, cleaning, accurate;
(2) carry out preparative experiment, determine sweeping scheme.According to concrete equipment and root system sample characteristic, selected sweep parameter.Technical parameters such as tube voltage, tube current, pitch and bed thickness as MSCT.
(3) carry out root system sample scanning, the root system sample is vertically put into the scanning plane bed, behind fixing, the calibrating position, through the prescan spatial dimension, implement scanning by selected sweep parameter, obtain original position tomography sequence image, and it is standby to deposit this image sequence in supporting hard disk;
As shown in Figure 2, promptly be the original position tomography sequence image that is grown in banyan shoot root system in the organic matter that utilizes medical 4 row's spiral CT equipment to collect, resolution is 512 * 512, and gray shade scale is 12, and whole root system sample collects 208 tomographic images altogether.The important technological parameters of used sweeping scheme is that tube voltage 120kV, tube current 90mA, layer are apart from 1mm, bed thickness 1mm.
(4) original position tomography sequence image is carried out data processing, treatment step mainly comprises sport technique segments such as image noise reduction, figure image intensifying, image segmentation and feature extraction, and purpose is the geometric properties information improving picture quality, obtain root system of plant.Wherein steps such as image noise reduction, figure image intensifying, feature extraction can adopt existing technology realization in this area, image segmentation is the key technique of Flame Image Process and analysis, and the height of segmentation precision directly has influence on further three-dimensional reconstruction quality and total system is measured and the accuracy of analysis.
The task of image segmentation is exactly the zone at root system of plant place to be separated from original tomography sequence image among the present invention, for further feature extraction is prepared.Original image involved in the present invention is a kind of by the faultage image of computing machine through the root system of plant of the high resolving power of backprojection reconstruction generation, high gray shade scale.The basic task of the analysis extraction algorithm of root system of plant space geometry characteristic parameter is to utilize the tomography sequence image after cutting apart to calculate, extract the original position space geometry characteristic parameter of root system of plant, comprises the important parameters such as characteristic dimension, branch directions angle of basic point position, branch node coordinate, the node region of root system.
Highlight image segmentation algorithm below.
In order to improve the efficient of cutting apart of root system CT image, the present invention fully utilizes the gray feature of image and the three-dimensional segmentation method that the spatial shape feature proposes a kind of man-machine interactive.Its basic thought is: at first, according in the root system CT image layer and the consistance or the similarity feature of interlayer gray scale, carry out three-dimensional Threshold Segmentation, the medium background is removed in initial gross separation; Then, adopt the appropriate mathematical morphological method, carry out the zone finishing, further improve segmentation precision; Then, utilize position and section shape change between root system CT image layer continuity, globality feature, remove the impurity pixel close with root density; Adopt the seed filling method at last, extract the root system zone, finish 3D and cut apart.
Its algorithm flow is described below:
1. adopt the generic threshold value method of discrimination, analyze the threshold range of determining root system zone in the MSCT sequence image.
Threshold value method of discrimination commonly used has between histogram analysis method, p-parametric method, homogeneity measure class between ultimate range method, maximum entropy method (MEM) and maximum kind/and the class internal variance is than method etc.Here be example with the maximum entropy method (MEM), provide the algorithm flow that threshold value is differentiated:
1. calculate the intensity profile probability of all pixels in the MSCT sequence image
p i = N i N
Wherein, N iFor gray-scale value is the pixel count of i, i=0,1,2 ..., L-1; L is whole gray shade scales of image, to 12 bit images, L=4096; N aBe the sum of all pixels of sequence image, i.e. N a=∑ N i
2. a given initial threshold t is divided into C with image 1And C 2Two classes;
Definition class C1 be used for describing 0,1 ..., the intensity profile of t}, class C2 be used for describing t+1, t+2 ..., the intensity profile of L-1}, then their probability distribution is:
C 1 : p 0 P t , p 1 P t , · · · , p t P t
C 2 : p t + 1 1 - P t , p t + 2 1 - P t , · · · , p L - 1 1 - P t
In the formula, P t = Σ i = 0 t p i
3. calculate the average relative entropy of two classes respectively
H ( C 1 ) = - Σ i = 0 t p i P t ln p i P t = ln P t + H t P t
H ( C 2 ) = - Σ i = 0 t p i 1 - P t ln p i 1 - P t = ln ( 1 - P t ) + H L - 1 - H t 1 - P t
In the formula, H t = - Σ i = 0 t ( p i ln p i ) , H L - 1 = - Σ i = 0 L - 1 ( p i ln p i )
4. define criterion function ψ (t)
ψ ( t ) = H ( C 1 ) + H ( C 2 ) = ln P t ( 1 - P t ) + H t P t + H L - 1 - H t 1 - P t
5. find the solution optimal threshold τ, make criterion function ψ (t) obtain maximal value.That is,
&tau; = max 0 < t < L - 1 - 1 { &psi; ( t ) }
2. three-dimensional Threshold Segmentation is carried out binary conversion treatment to the MSCT sequence image.
F ( x , y , z ) = 0 f ( x , y , z ) < &tau; 1 f ( x , y , z ) > &tau;
3. utilize principles of mathematical morphology,, carry out opening operation and handle the binary sequence image after the three-dimensional Threshold Segmentation.
If F is the root system CT sequence image after the Threshold Segmentation, E is a structural element; Then F is with respect to the opening operation of E:
X?o?E={x∈X,for?t∈X-E,x∈Et?and?Et?X}
Or X o E=(X Θ E) E
Wherein, o, Θ, be unlatching, corrosion and dilation operation symbol respectively, Et represents that one of structural element translation is apart from t.
4. start with by basic root faultage image, set up the recursion basis.
1. the manual root zone position of determining L-1 layer CT image.
2. calculate the regional location and the region area of basic root section in the m tomographic images such as top L-1, L-2, L-3;
3. calculate the interlayer rate of change of each radical center coordinate and root district area, and find out wherein maximal value and mean value.
4. remove the Fei Jigen zone of above each layer, and meter last one deck wherein is that the k layer is k=N-m+1.
5. based on basic root faultage image data, successively infer the root system zone downwards, reject the impurity pixel.
1. make i=k-1, j=1.
2. utilize polynomial interpolation to calculate according to preceding m layer radical center position coordinates and determine radical center desired location P in the i tomographic image j, and judge whether its pairing pixel value is 0.
3. calculate in the i tomographic image and comprise P jThe position coordinates and the region area in zone calculate its position rate and area change rate simultaneously.
4. screening, the root district in the mark i tomographic image.
5. remove the non-root system zone of i layer.
If 6. i〉0, then change 3.; Otherwise change 6.
6. utilize se ed filling algorithm to extract the root system zone, finish sequence image and cut apart.
After above-mentioned processing, obtain root system space surface coordinate, thereby construct solid geometric pattern, generate virtual root system.
(5) virtual root system is carried out three-dimensional visualization and rebuild, construct the 3-D display model of root system.
The three-dimensional visualization of root system of plant is rebuild, and is the important technical links that realizes in-situ observation and its geometric configuration of quantitative test.The method of 3 d objects reconstruction is a lot of from three-dimensional data, and the rank difference of the fundamental element of handling by process of reconstruction can be divided into two big classes: voxel level method for reconstructing and slice-level method for reconstructing.When the resolution of original image was higher, the voxel level was rebuild reliable more, effective.Rebuild for the three-dimensional visualization of realizing root system of plant, the present invention proposes two kinds of models: geometric model and display model.3-D geometric model is to rebuild according to the space geometry characteristic parameter of root system of plant to generate, and is mainly used in the quantitative measurment and the analysis of root system of plant, helps to significantly improve the speed and the precision of Measurement and analysis.The 3-D display model, (Marching Cubes, MC) technology at first makes up the three-dimensional contour surface model of root system of plant, generates in conjunction with illumination model and the visual angle factor then can to adopt in the voxel level method for reconstructing stepping cube commonly used.Like this, help to reduce three-dimensional picture rebuilds operand, improve three-dimensional reconstruction and speed of displaying and quality.
The MC basic idea is the cube (voxel) in the deal with data field one by one, sorts out and cube that contour surface intersects, adopts interpolation calculation to go out the intersection point on contour surface and cube limit.According to the relative position of each summit of cube and contour surface, the intersection point on contour surface and cube limit is connected by certain way generates contour surface, approach expression as contour surface in this cube.Thereby, in the MC algorithm in each unit iso-surface extracting all need carry out: the contour surface that approached by triangular plate in the voxel calculates and each vertex scheme vector calculating of triangular plate.
The basic procedure that MC extracts iso-surface algorithms is:
1. the layering of 3 d-dem regular data field is read in;
2. scan two layer data, construct voxel one by one, 8 angle points in each voxel are taken from adjacent two-layer;
3. the functional value of each angle point of voxel and given equivalent face amount are compared,, construct the concordance list of this voxel according to comparative result;
4. draw and with contour surface the boundary voxel of intersection point to be arranged according to concordance list;
5. calculate the intersection point of voxel seamed edge and contour surface by linear interpolation method;
6. utilize the central difference method, obtain the normal vector at each angle point place of voxel,, obtain the normal direction at each place, summit of tri patch again by linear interpolation method;
7. draw the contour surface image according to the coordinate and the normal vector on each each summit of tri patch.
Fig. 3 adopts the original position stereopsis of the banyan shoot root system of this three-dimensional rebuilding method generation to show among the present invention.
(6) Measurement and analysis
3-D geometric model is to rebuild according to the space geometry characteristic parameter of root system of plant to generate, and is mainly used in the quantitative measurment and the analysis of root system of plant, helps to significantly improve the speed and the precision of Measurement and analysis.Adopt existing measuring and analysis system that the virtual root system of banyan is measured and analyzed, and output is measured and analysis result, if this result is reasonable, then measure and the analytic process end, if this result is unreasonable, then carry out data correction and handle, and the return measurement analytic system, until rationally, procedure structure then.
The in-situ measuring method of root system of plant geometric configuration of the present invention at first adopts modern biomedical image technologies such as XCT, MRI to obtain the original tomography sequence image that is grown in root system of plant in soil or other opaque medium; Then utilize the computer graphic image treatment technology, original position tomography sequence image is carried out suitable processing, analysis and three-dimensional reconstruction, the in-situ three-dimensional image of root system of plant be can clearly observe on computers, and the accurate measurement and the quantitative test of root system of plant geometric configuration realized.
Method of the present invention is from the requirement of in-situ sampling, can obtain to entirely accurate the in-situ image of root system of plant, need not to have saved the actual sampling process of time and effort consuming from the soil sampling, certainly just can not destroy the original position state of root system yet, even not influence the normal growth of root system.This not only helps accurately obtaining static state (a certain moment) three-dimensional geometry of root system, also provides technical support for the research of further carrying out based on the Three-Dimensional Dynamic configuration of root system live body.Requirement from root system observation, its measuring process is carried out on computers, not only can realize the measurement accurately and fast of root system 3-d modelling, and can reappear a certain root system of plant live body (non-analogy model) growing state of (static state) or continuous (dynamically) at a time on computers intuitively with the form of three-dimensional picture at any time.Therefore, technical scheme of the present invention be can be accurately and fast, original position, nondestructively realize the root system live body three-dimensional static and the dynamic non-destructive scheme of configuration observation.
More than be preferred forms of the present invention, according to content disclosed by the invention, some identical, replacement schemes that those of ordinary skill in the art can expect apparently all should fall into the scope of protection of the invention.

Claims (2)

1. the in-situ measuring method of a root system of plant geometric configuration is characterized in that comprising the steps:
(1) adopt biomedical imaging equipment to obtain the original position tomography sequence image that is grown in root system of plant in the opaque medium;
(2) the original position space geometry characteristic parameter that draws root system of plant is handled and analyzed to the sequence image of gained in the step (1), wherein, the treatment step that sequence image is carried out data processing mainly comprises image noise reduction, figure image intensifying, image segmentation and feature extraction, and, described image segmentation adopts the three-dimensional segmentation method of man-machine interactive, and based on described characteristic parameter structure solid geometric pattern, wherein, described three-dimensional segmentation method comprises:
(2-1) adopt the generic threshold value method of discrimination;
(2-2) implement three-dimensional Threshold Segmentation;
(2-3) utilize principles of mathematical morphology, the binary sequence image after the three-dimensional Threshold Segmentation is carried out opening operation handle;
(2-4) start with, set up the recursion basis by basic root faultage image;
(2-5) based on basic root faultage image data, successively infer the root system zone downwards, deletion impurity pixel;
(2-6) utilize se ed filling algorithm to extract the root system zone, finish sequence image and cut apart;
(3) solid geometric pattern according to step (2) carries out the three-dimensional visualization reconstruction, make up the 3-D display model of root system, and, described 3-D display model, adopt stepping cube method in the voxel level method for reconstructing at first to make up the three-dimensional contour surface model of root system of plant, generate in conjunction with illumination model and the visual angle factor then;
(4) Visualization Model of utilizing step (3) to make up is a Man Machine Interface, and the structure parameters of root system of plant assigned address or the degree of depth and root system integral body is implemented to measure and analyze.
2. in-situ measuring method as claimed in claim 1 is characterized in that described biomedical imaging equipment is X computer on line tomographic apparatus or MR imaging apparatus.
CNB2005101264972A 2005-12-14 2005-12-14 In-situ measuring method of plant root system geometry Active CN100487362C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2005101264972A CN100487362C (en) 2005-12-14 2005-12-14 In-situ measuring method of plant root system geometry

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2005101264972A CN100487362C (en) 2005-12-14 2005-12-14 In-situ measuring method of plant root system geometry

Publications (2)

Publication Number Publication Date
CN1776357A CN1776357A (en) 2006-05-24
CN100487362C true CN100487362C (en) 2009-05-13

Family

ID=36765997

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2005101264972A Active CN100487362C (en) 2005-12-14 2005-12-14 In-situ measuring method of plant root system geometry

Country Status (1)

Country Link
CN (1) CN100487362C (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101209015B (en) * 2006-12-29 2012-03-28 中国科学院东北地理与农业生态研究所 Corn roots space-time structure measuring device
CN101266690B (en) * 2007-03-15 2012-02-29 华南农业大学 Plant root species form 3-D image reconstruction system and method
CN101216470B (en) * 2008-01-08 2011-05-04 上海大学 Mucky beach plant root system growth in situ monitoring method
CN101658107B (en) * 2009-09-18 2011-06-22 华南农业大学 Method for in-situ dynamic observation and measurement of root system of plant
CN102680509A (en) * 2012-05-17 2012-09-19 西北农林科技大学 Plant online nuclear magnetic resonance imaging detection analyzing device under artificial climate
CN102967617B (en) * 2012-11-23 2016-02-03 浙江大学 A kind of measurement mechanism of crop root nuclear magnetic resonance research and method
CN103728418A (en) * 2013-12-31 2014-04-16 马玮 Analyzing method for researching parameters of architecture and spatial distribution of crop roots
CN105830847B (en) * 2014-10-24 2018-08-14 三峡大学 A kind of root system rock breaking mechanism method for visually monitoring
CN106248700B (en) * 2016-08-22 2019-11-05 江苏大学 A kind of seedling alms bowl root structure Morphology observation method based on CT
CN106556614B (en) * 2016-10-12 2019-06-14 北京师范大学 A method of root system spacial framework is quantified based on CT scan technology
CN110398503A (en) * 2019-02-27 2019-11-01 广西壮族自治区农业科学院 A kind of plant pest method of inspection based on geometric shape transmission measurement
CN110866972B (en) * 2019-11-12 2023-10-13 广西大学 In-situ observation device for sugarcane root system configuration and analysis method thereof
CN112913554A (en) * 2021-01-18 2021-06-08 田举鹏 Century ancient tree transplanting method
CN113575397B (en) * 2021-08-24 2022-08-02 中国科学院南京地理与湖泊研究所 Method for constructing three-dimensional geometric configuration of aquatic plant root system
CN115183105B (en) * 2022-07-29 2023-12-15 江苏大学 Device and method for identifying healthy seedlings of plug tray

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
基于XCT技术的植物根系原位形态可视化研究. 罗锡文,周学成,严小龙,罗良平,向子云.农业机械学报,第35卷第2期. 2004
基于XCT技术的植物根系原位形态可视化研究. 罗锡文,周学成,严小龙,罗良平,向子云.农业机械学报,第35卷第2期. 2004 *
多层螺旋CT原位观测作物根系的研究. 向子云,周学成,罗锡文,罗良平,赵相胜.中国CT和MRI杂志,第2卷第1期. 2004
多层螺旋CT原位观测作物根系的研究. 向子云,周学成,罗锡文,罗良平,赵相胜.中国CT和MRI杂志,第2卷第1期. 2004 *
植物根系生长的三维可视化模拟. 钟南,罗锡文,严小龙,廖红.华中农业大学学报,第24卷第5期. 2005
植物根系生长的三维可视化模拟. 钟南,罗锡文,严小龙,廖红.华中农业大学学报,第24卷第5期. 2005 *

Also Published As

Publication number Publication date
CN1776357A (en) 2006-05-24

Similar Documents

Publication Publication Date Title
CN100487362C (en) In-situ measuring method of plant root system geometry
EP2960869B1 (en) Ct system for security check and method thereof
Mairhofer et al. RooTrak: automated recovery of three-dimensional plant root architecture in soil from X-ray microcomputed tomography images using visual tracking
WO2003088085A1 (en) Three-dimensional digital library system
Gallay et al. Geomorphometric analysis of cave ceiling channels mapped with 3-D terrestrial laser scanning
US20080225044A1 (en) Method and Apparatus for Editing Three-Dimensional Images
Shin et al. Analysis of the vesicular structure of basalts
CN101673413B (en) System and method for modeling three-dimensional configuration of plant branches based on surface data
Lai et al. Surface roughness of rock faces through the curvature of triangulated meshes
CN109191510B (en) 3D reconstruction method and device for pathological section
CN104268940B (en) MEMS structure reconstruction and detection method based on CT scanned images
CN109410139A (en) Inside a kind of historical relic and surface disease digital assay appraisal procedure
Gavryushkina The potential and problems of volumetric 3D modeling in archaeological stratigraphic analysis: A case study from Chlorakas-Palloures, Cyprus
CN101996415B (en) Three-dimensional modeling method for eyeball
CN115018986A (en) Strong-structure active region rock microstructure information interpretation and three-dimensional modeling method
Monnet et al. Three-dimensional morphometric ontogeny of mollusc shells by micro-computed tomography and geometric analysis
Pérez-Ramos et al. Toward an “ancient” virtual world: improvement methods on X-ray CT data processing and virtual reconstruction of fossil skulls
Sahebdivani et al. Deep learning based classification of color point cloud for 3D reconstruction of interior elements of buildings
Tian 3D modeling and digital preservation of ancient architectures based on autoCAD and 3Dmax
Zhou et al. A method of 3D nondestructive detection for plant root in situ based on CBCT imaging
CN109147061B (en) Method for carrying out human-computer interaction editing on segmentation result of volume data
Xiao Detecting changes in trees using multi-temporal airborne LIDAR point clouds
Qing et al. Classified denoising method for laser point cloud data of stored grain bulk surface based on discrete wavelet threshold
Zhang Comparative Study of Pore Structure Characterization for Different Rock Samples Using Digital Images
Corkum Three-dimensional triangulated boundary element meshing of underground excavations and visualization of analysis data.

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant