CN110274556B - Plant phenotype information extraction method - Google Patents
Plant phenotype information extraction method Download PDFInfo
- Publication number
- CN110274556B CN110274556B CN201910514370.XA CN201910514370A CN110274556B CN 110274556 B CN110274556 B CN 110274556B CN 201910514370 A CN201910514370 A CN 201910514370A CN 110274556 B CN110274556 B CN 110274556B
- Authority
- CN
- China
- Prior art keywords
- plant
- visible light
- light camera
- phenotype information
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/28—Measuring arrangements characterised by the use of optical techniques for measuring areas
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
- G01B11/0608—Height gauges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/08—Measuring arrangements characterised by the use of optical techniques for measuring diameters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a plant phenotype information extraction method, and belongs to the field of plant phenotype information acquisition equipment. According to the plant phenotype information extraction method, the acquisition device is used for acquiring the plant phenotype information, the images of multiple angles in the whole plant space are extracted by controlling the shooting time of the visible light camera, the average value of the images is calculated, the influence of illumination inconsistency on different images is minimized, and errors caused by the occlusion of branches and leaves or the spatial stretching and curling of the leaves on the measurement of phenotype parameters such as leaf area, plant height and base diameter are reduced.
Description
Technical Field
The invention relates to the technical field of plant phenotype information acquisition equipment, in particular to a plant phenotype information extraction method.
Background
Plant phenotype refers to a complex plant trait that is genetically and environmentally determined or influenced, including morphological structures and physiological and biochemical parameters exhibited by growth, development, tolerance, resistance, physiology, structure, yield, and the like. The measurable morphological phenotypic parameters include plant height, basal diameter, leaf characteristics, shoot characteristics, fruit characteristics, and the like. Phenotype is the systematic measurement and characterization of the morphological, structural, physical, chemical and biological characteristics of a plant, from macroscopic to microscopic.
Most typical parameters capable of measuring morphological structure phenotype are plant height, basal diameter and leaf area. The plant height and the base diameter represent the growth vigor, the growth speed and the like of the plant. The leaf is the main organ for photosynthesis of plant to produce oxygen and synthesize nutrient, and the leaf is the main path for transpiration of plant to provide the power for root to absorb and transport water from outside.
The plant phenotype has the characteristics of complexity, variability influenced by environment, dynamic change in the whole process and the like, the traditional phenotype information is measured by a manual measuring mode, if the plant height is measured by a tape measure, the base diameter is measured by a vernier caliper, the area of the leaf is measured by a grid counting method after in-vitro picking, namely, the leaf is placed on a horizontal plane and covered by transparent grid paper, and the area of the leaf is calculated by a method for counting the number of grids in the leaf and the number of grids at the edge of the leaf. Therefore, the traditional method for manually measuring the phenotype information has the problems of small scale, low efficiency, poor precision, more errors, weak continuity, strong destructiveness and the like.
In the measurement of the morphological structure phenotype of the plant, the visible light camera is utilized to fixedly acquire morphological phenotype information from a single angle, so that the method is simple and convenient, but the problems of branch and leaf shielding, inaccurate single-angle projection value, large calculation error of information such as leaf area and the like exist. Therefore, morphological structure information of plants at different angles is generally acquired by adopting a mode that the plant movement is based on a conveyor belt or a rotating disc and a visible camera is fixed, and relevant phenotypic parameters are extracted through algorithm processing; however, the movement of the conveyor belt or the rotation of the rotary disc can cause the shaking of the plant, particularly the organs of the plant with tiny leaves and slender stem, and the quality of the acquisition of the morphological structure and phenotype information of the plant is influenced, so that the noise of the image is too large, and the acquisition and analysis of the phenotypic information of the plant are influenced.
Generally, a visible light camera can be controlled by a rotating device to rotate around a plant to realize multi-angle acquisition of plant phenotype information, for example, a Chinese patent document (application number is 2018213848722) named as a plant phenotype acquisition device is invented and created, a supporting base of the application is provided with a rotating mechanism, and the rotating mechanism comprises an installation cylinder fixed on the supporting base and a rotating cylinder sleeved on the outer side of the installation cylinder and capable of rotating; the rotary drum is provided with at least one side wing plate, the side face of the side wing plate is connected with an assembling support, and an acquisition device for acquiring plant phenotype information is arranged on the assembling support. When the scheme is used for analyzing the obtained image, a mode that the plant is placed in the middle for rotation and the sensor is fixed is adopted, and the phenotype information acquisition mode can cause the plant, particularly the organs of the plant with tiny leaves and slender stems to shake, so that image noise is brought, and the measurement effect is influenced.
Disclosure of Invention
1. Technical problem to be solved by the invention
The invention aims to provide a plant phenotype information extraction method, which is used for extracting plant phenotype information by using an acquisition system, has no destructiveness, no influence on plant growth, high speed and efficiency, high precision and better acquisition effect.
2. Technical scheme
In order to achieve the purpose, the technical scheme provided by the invention is as follows:
the invention relates to a plant phenotype information extraction method, which utilizes a plant phenotype information acquisition system to acquire the phenotype information of plants, wherein the acquisition system comprises,
the bearing platform is used for placing plants;
the information acquisition device comprises a visible light camera and an adjusting mechanism, wherein the visible light camera is arranged at the upper end of the adjusting mechanism, faces the bearing table and is used for acquiring the phenotype information of the plants;
the rotating device comprises a rotating ring, a supporting seat and a driving vehicle, the rotating ring is arranged on the bearing platform and can rotate relative to the bearing platform, one end of the supporting seat is connected with the rotating ring, the driving vehicle is arranged at the other end of the supporting seat, and the information acquisition device is installed on the supporting seat.
The extraction method comprises the following steps:
placing a container filled with plants on a bearing table, and adjusting the illumination intensity and the illumination direction of ambient light;
adjusting the position of the visible light camera through an adjusting mechanism;
thirdly, the rotating device drives the information acquisition device to rotate for 1 circle around the bearing platform, and the rotating time is recorded; meanwhile, calibrating and correcting the acquired image through 6 calibration plates on the circumferential direction of the curtain, and calculating a projection matrix of the visible light camera to obtain a geometric model of the visible light camera;
step four, setting the shooting time of the visible light camera according to the rotation time in the step three;
and step five, the rotating device drives the information acquisition device to rotate for 1 circle around the bearing table, the visible light camera shoots the plants for 10 times according to the shooting time in the step four, and the average value of the phenotype information of the plants is calculated.
In the fifth step, the specific step of calculating the mean value of the plant phenotype information is,
s1, segmenting a plant region from the RGB image;
s2, converting the RGB image of the plant region in the S1 into a single-waveband image by using an index 2G/(R + B);
s3, converting the single-waveband image obtained in the S2 into a binary image, and screening and denoising the binary image;
and S4, respectively calculating plant phenotype information parameters of the plants in the 10 binary images, including plant height, basal diameter and leaf area, and calculating the average value of the plant phenotype information parameters.
Further, before the first step, the plummer is arranged on the curtain, and 6 calibration plates are equidistantly arranged in the circumferential direction of the curtain.
Further, when the plant height is calculated in S4, setting a polygon limit framing range, using a global image threshold, scanning line by line, and setting the distance from the highest point where pixels at the top of the plant are continuous to the upper edge of the pot to the plant height of the plant;
when the base diameter is calculated, scanning line by line from top to bottom, identifying the container, taking the part which is close to the upper edge of the container and has continuous pixels as the stem, and calculating the width of the stem part as the base diameter of the plant;
when calculating the leaf area, a structural unit is arranged, isolated noise is eliminated by using a form opening technology, and the number of white pixels is calculated to be the projection area of the plant leaf.
3. Advantageous effects
Compared with the prior art, the technical scheme provided by the invention has the following beneficial effects:
(1) according to the plant phenotype information acquisition system, the information acquisition device comprises the visible light camera and the adjusting mechanism for adjusting the position of the visible light camera, and the information acquisition device is driven to rotate around the bearing table through the rotating device, so that the plant phenotype information acquisition system can adapt to the phenotype information acquisition of plants in different growth periods, and can acquire the phenotype information of the plants from multiple angles, therefore, the plant phenotype information acquisition system can perform nondestructive measurement on morphological structure phenotype three-dimensional parameters in the whole process of plant growth, and the acquisition effect is good.
(2) The plant phenotype information acquisition system is also provided with a curtain and 6 chessboard marking plates, wherein the bearing table is arranged on the curtain, the curtain can improve the ambient light intensity acquired by plants, so that the acquisition effect is improved, the 6 marking plates are arranged in the circumferential direction of the curtain at equal intervals, the acquired images are corrected through the 6 marking plates, the deformation of a single marking plate after the information acquisition device rotates is avoided, the calibration is carried out through views at different angles, the projection matrix of the visible light camera is calculated, the geometric model of the visible light camera is obtained, and the high-precision acquired images are obtained.
(3) The adjusting mechanism comprises at least 3 supporting legs, and the visible light camera can keep better stability when the information acquisition device rotates through the matching of the 3 supporting legs, so that a shot image is clearer; the supporting leg passes through on the slider sets up the supporting seat, and the position of conveniently adjusting visible light camera is matched with the slide rail on the supporting seat to the gyro wheel of slider.
(4) According to the plant phenotype information extraction method, the shooting time of the visible light camera is controlled, the multi-angle images in the whole plant space are collected, the average value of the images is calculated, the influence of illumination inconsistency on different images is minimized, and errors caused by the fact that branches and leaves are shielded or the space of the leaves stretches and curls and the like are used for measuring phenotype parameters such as leaf area, plant height, basic diameter and the like are reduced.
Drawings
FIG. 1 is a schematic diagram of a plant phenotype information acquisition system of the present invention;
FIG. 2 is a schematic view of the distribution of calibration plates in the present invention;
FIG. 3 is a schematic view of a slider according to the present invention;
FIG. 4 is a schematic diagram of a plant binary image from multiple angles;
FIG. 5 is a schematic flow chart of the extraction method.
The reference numerals in the schematic drawings illustrate: 100. a bearing table; 200. an information acquisition device; 210. a visible light camera; 220. an adjustment mechanism; 221. supporting legs; 222. a slider; 223. a support; 224. a rotating sleeve; 225. a roller; 226. locking the bolt; 227. fixing the bolt; 300. a rotating device; 310. a rotating ring; 320. a supporting seat; 321. a slide rail; 330. driving the vehicle; 410. a curtain; 420. calibrating the plate; 500. a plant.
Detailed Description
For a further understanding of the invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings and examples.
The structure, proportion, size and the like shown in the drawings are only used for matching with the content disclosed in the specification, so that the person skilled in the art can understand and read the description, and the description is not used for limiting the limit condition of the implementation of the invention, so the method has no technical essence, and any structural modification, proportion relation change or size adjustment still falls within the scope of the technical content disclosed by the invention without affecting the effect and the achievable purpose of the invention. In addition, the terms "upper", "lower", "left", "right" and "middle" used in the present specification are for clarity of description, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the relative positions may be changed or adjusted without substantial technical changes.
The embodiments provided in the embodiments of the present specification may be combined with each other as necessary, and when a plurality of embodiments are present in a single embodiment, the embodiments may be combined with each other as necessary. Unless otherwise indicated, the combined examples and embodiments are not repeated in this specification because they still have the unexpected effect and are within the scope of the present invention.
In the measurement of the morphological structure phenotype of the plant, the visible light camera is utilized to fixedly acquire morphological phenotype information from a single angle, so that the method is simple and convenient, but the problems of branch and leaf shielding, inaccurate single-angle projection value, large calculation error of information such as leaf area and the like exist. Therefore, morphological structure information of plants at different angles is generally acquired by adopting a mode that the plant movement is based on a conveyor belt or a rotating disc and a visible camera is fixed, and relevant phenotypic parameters are extracted through algorithm processing; however, the movement of the conveyor belt or the rotation of the rotary disc can cause the shaking of the plant, particularly the organs of the plant with tiny leaves and slender stem, and the quality of the acquisition of the morphological structure and phenotype information of the plant is influenced, so that the noise of the image is too large, and the acquisition and analysis of the phenotypic information of the plant are influenced.
In order to overcome the above problem, referring to fig. 1, the plant phenotype information acquisition system of the present embodiment includes a carrier 100, an information acquisition device 200, and a rotation device 300, wherein the rotation device 300 drives the information acquisition device 200 to rotate relative to the carrier 100, and the information acquisition device 200 is used for acquiring the phenotype information of a plant 500 on the carrier 100.
Specifically, plants 500 with collected phenotypic data may be placed on the carrier 100. The information collecting device 200 comprises a visible light camera 210 and an adjusting mechanism 220, the visible light camera 210 is disposed toward the plummer 100 to shoot the image of the plant 500, the visible light camera 210 can adopt a common shooting device, such as a single-lens reflex visible light camera, and the specific structure of the visible light camera is not limited as long as the shooting can be performed; the adjusting mechanism 220 is provided with the visible light camera 210, and the adjusting mechanism 220 is used for adjusting the position of the visible light camera 210 so as to meet the plant phenotype information acquisition requirements in different growth periods. The rotating device 300 comprises a rotating ring 310, a supporting seat 320 and a driving trolley 330, wherein the rotating ring 310 is arranged on the bearing platform 100 and can rotate relative to the bearing platform 100, one end of the supporting seat 320 is connected with the rotating ring 310, the other end of the supporting seat 320 is provided with the driving trolley 330, and the information acquisition device 200 is arranged on the supporting seat 320.
When the plant phenotype information acquisition system of the present embodiment is used to acquire plant phenotype information, the plant 500 is placed on the carrier 100, the position of the visible light camera 210 is adjusted by the adjusting mechanism 220, and the support base 320 is driven to rotate around the center of the carrier 100 by the driving vehicle 330, so as to drive the visible light camera of the information acquisition device 200 to rotate around the carrier 100, and the visible light camera 210 completes acquisition of the phenotype information of the plant 500.
In addition, the plant phenotype information acquisition system of the embodiment further includes a curtain 410 and a plurality of calibration plates 420, the plummer 100 is arranged on the curtain 410, the curtain 410 may be a black flannelette, and the curtain 410 may improve the ambient light intensity of plant acquisition, thereby improving the image acquisition effect; the plurality of calibration plates 420 are equidistantly arranged along the circumferential direction of the curtain 410, the calibration plates 420 may be black and white checkerboard calibration plates, and the number of the calibration plates 420 may be 6 or more than 6. The acquired images are corrected through the calibration plates 420, image deformation caused by rotation of the single calibration plate 420 in the information acquisition device is avoided, the calibration is performed through views at different angles, internal and external parameters of the camera are obtained for many times, a projection matrix of the visible light camera is calculated, a geometric model of the visible light camera is obtained, distortion correction and polar line correction are performed on the acquired images, the calibration effect is ensured, the real effect is restored as far as possible, and the high-precision acquired images are obtained.
As a further optimization of the present embodiment, referring to fig. 2, the rotating ring 310 is sleeved on the plummer 100, and when the rotating device 300 rotates, the rotating ring 310 can provide a guiding function; the outer side of the bearing table 100 is provided with a rotation groove, and a plurality of balls are arranged in the rotation groove, namely, the balls are arranged between the rotating ring 310 and the bearing table 100, so that the sliding friction between the rotating ring 310 and the bearing table 100 is converted into rolling friction force, and the rotating device 300 can rotate more easily. In addition, when the rotating ring 310 is sleeved on the bearing table 100, when the rotating device 300 rotates, the information acquisition device 200 is more stable, so that the shot image is clearer, and the plant phenotype information acquisition effect is better.
As a further optimization of the present embodiment, the adjusting mechanism 220 includes a support platform and at least 3 support legs 221 disposed on the support platform, a sliding member 222 is disposed on the support legs 221, a plurality of sliding rails 321 is correspondingly disposed on the support base 320, and the sliding member 222 is capable of sliding on the sliding rails 321 in cooperation with the sliding rails 321. When the number of the support legs 221 is 3, referring to fig. 2, 3 mutually parallel slide rails 321 may be disposed on the support base 320 and all disposed along the length direction of the support base 320; the 3 sliding rails 321 can also be led out from the center of the supporting seat 320, and included angles formed between the sliding rails are the same; it is also possible to provide 2 mutually perpendicular slide rails 321, and 1 support leg 221 of the adjusting mechanism 220 is provided on 1 slide rail 321, and the other 2 support legs 221 are provided on 1 slide rail 321. The position relationship between the sliding rails 321 of the supporting base 320 is not particularly required, as long as the supporting angle of the supporting legs 221 can be changed by adjusting the position of the sliding member 222 on the sliding rails 321, and the purpose of adjusting the position of the visible light camera can be achieved by the mutual matching of the supporting legs 221.
Further, the slider 222 includes a bracket 223, and a rotating sleeve 224 and a roller 225 disposed on the bracket 223. The bracket 223 is sleeved on the sliding rail 321, the roller 225 can be arranged above the sliding rail 321, and the sliding part 222 slides on the sliding rail 321 through the matching of the roller 225 and the sliding rail 321; the rotary sleeve 224 is disposed on the upper side of the bracket 223, the rotary sleeve 224 may be a U-shaped structure, and the support leg 221 can be clamped in the U-shaped groove of the rotary sleeve 224 and can rotate relative to the rotary sleeve 224. In addition, the rotating sleeve 224 and the bracket 223 may be hinged, the rotating sleeve 224 may rotate relative to the bracket 223, and when the sliding member 222 slides on the sliding rail 321, the supporting leg 221 can rotate relative to the sliding member 222, so as to conveniently adjust the supporting angle of the supporting leg 221, and thus adjust the position of the visible light camera 210 relative to the plummer 100.
As a further optimization of the present embodiment, 2 rollers 225 are disposed in the bracket 223, the 2 rollers 225 are respectively located at the upper side and the lower side of the sliding rail 321, and meanwhile, the 2 rollers 225 are mutually matched to be able to clamp the sliding rail 321, so as to provide a guiding function for the sliding piece 222 to slide on the sliding rail 321.
Further, a locking bolt 226 and a bottom bolt are provided on the rotary sleeve 224, and a fixing bolt 227 is provided on the bracket 223. After the position of the visible light camera 210 is adjusted, the supporting leg 221 in the rotary sleeve 224 can be locked by screwing the locking bolt 226, the rotary sleeve 224 and the support 223 can be fixed by screwing the bottom bolt, the sliding piece 222 and the sliding rail 321 can be fixed by screwing the fixing bolt 227, so that the supporting leg 221 of the adjusting mechanism 220 cannot slide or rotate relative to the sliding rail 321, when the rotating device 300 drives the information acquisition device 200 to rotate relative to the bearing table 100, the visible light camera 210 of the information acquisition device 200 has better stability, the visible light camera 210 is prevented from shaking, the definition of a shot image is improved, and the accuracy and precision of plant phenotype information acquisition are improved.
Referring to fig. 5, in the embodiment, a plant phenotype information extraction method is further provided, where a plummer 100 is required to be arranged on a curtain 410, 6 calibration plates 420 are equidistantly arranged in the circumferential direction of the curtain 410, and collected images are corrected by the 6 calibration plates, so that deformation of a single calibration plate after rotation of an information collection device is avoided, calibration is performed by views at different angles, and a projection matrix of a visible light camera is calculated to obtain a geometric model of the visible light camera, so as to obtain a high-precision collected image.
The embodiment adopts the mode of visible light camera rotation and plant fixation for collection, and avoids the organ shaking of the plant. Before shooting images, a checkerboard is usually used for calibrating and acquiring an internal reference matrix of a camera, the coordinate relation between the camera and a plant is calculated, if the camera is fixed all the time in the shooting process, the calibration parameters are not changed, and the accuracy of three-dimensional information is not influenced. However, during the rotation shooting process, the position of the camera changes, so the correlation between each characteristic point on the calibration board and the projection point on the imaging plane changes. In the 6-time calibration technology of the embodiment, the internal and external parameters of the camera are obtained for multiple times in the camera rotation process, and the camera calibration parameters are determined to perform distortion correction and epipolar line correction on the obtained image, so that the real effect is restored as much as possible, and the calibration effect is ensured. Meanwhile, the embodiment regularly collects 10 images of all-around space and takes the average value as the final phenotype information parameter value, so that the error caused by branch and leaf shielding or space extension curling of leaves and the like when the phenotype information is extracted at a single angle is reduced, and the accuracy of the phenotype information is improved.
Specifically, when the collection system is used for collecting the phenotype information of the plant, the method may include the following steps:
step one, the container with the plant 500 is placed on the bearing table 100, and the illumination intensity and the illumination direction of the ambient light are adjusted.
Step two, the position of the visible light camera 210 is adjusted through the adjusting mechanism 220, during the adjustment, the locking bolt 226, the fixing bolt 227 and the bottom bolt are firstly unscrewed, then the plurality of supporting legs 221 of the adjusting mechanism 220 are respectively slid, so that the visible light camera 210 can be aligned to the plant 500, the shooting angle of the visible light camera 210 is adjusted, and then the locking bolt 226, the fixing bolt 227 and the bottom bolt are screwed, so that the position of the visible light camera 210 relative to the supporting seat 320 is fixed, and the visible light camera 210 is prevented from shaking when rotating along with the rotating device 300.
And step three, the rotating device 300 drives the information acquisition device 200 to rotate for 1 circle around the bearing table 100, and the rotation time is recorded. Meanwhile, the geometric model of the visible light camera is obtained by calibrating and correcting the acquired images and calculating the projection matrix of the visible light camera through 6 calibration plates arranged on the periphery of the curtain.
And step four, setting the shooting time of the visible light camera 210 according to the rotation time in the step three.
Step five, the rotating device 300 drives the information acquisition device to rotate around the bearing table 100 for 1 circle, and the visible light camera 210 shoots the plant 500 according to the shooting time in step four. Specifically, the visible light camera 210 is driven by the rotating device 300 to move around the plummer 100 at a constant speed, and take pictures at regular time, and the shutter of the visible light camera 210 is controlled to take an image every time the rotating device 300 rotates by 36 degrees, so that the acquisition of 10-degree plant images in the whole space is completed. Calculating the projection area of the leaf area, and then calculating the average value obtained by ten angles to be used as the phenotypic parameter value of the leaf area; calculating the height of the top leaf of the plant naturally extending to the top of the pot to the upper edge of the pot, and then calculating the average value of ten angles of the top leaf of the plant as the phenotypic parameter value of the plant height; calculating the diameter of the stem near the soil surface, namely the upper edge of the pot, of the plant, and then calculating the average value of ten angles of the plant, wherein the average value is used as the parameter value of the basal diameter phenotype. The method for averaging the images obtained at the ten angles minimizes the influence of inconsistent illumination on different images, and reduces errors caused by the occlusion of branches and leaves or the spatial extension and curling of the leaves on the measurement of phenotypic parameters such as leaf area, plant height, base diameter and the like.
In the fifth step, the specific step of calculating the mean value of the plant phenotype information is as follows:
and S1, segmenting the regions of the plants from the RGB images.
S2, converting the RGB image of the plant region in the S1 into a single-waveband image by using the index 2G/(R + B).
And S3, converting the single-waveband image obtained in the S2 into a binary image, and screening and denoising the binary image. Specifically, a region restriction technique is used to specify a polygonal region of interest, remove small objects from the binary image, denoise and segment plant pixels from the background.
And S4, respectively calculating plant phenotype information parameters of the plants in the 10 binary images, including plant height, basal diameter and leaf area, and calculating the average value of the plant phenotype information parameters. Specifically, when the plant height is calculated, a polygon limiting frame selection range is set, a global image threshold is used, line-by-line scanning is carried out, and the distance from the continuous highest position of pixels at the top of the plant to the upper edge of the pot is the plant height of the plant; when the base diameter is calculated, scanning line by line from top to bottom, identifying the container, taking the part which is close to the upper edge of the container and has continuous pixels as the stem, and calculating the width of the stem part as the base diameter of the plant; when calculating the leaf area, a structural unit is arranged, isolated noise is eliminated by using a form opening technology, and the number of white pixels is calculated to be the projection area of the plant leaf.
Fig. 4 is a 10-degree binary image of the plant phenotype information acquired according to the plant phenotype information extraction method provided in this embodiment, and the visible light cameras have photographing angles of 0 °, 36 °, 72 °, 108 °, 144 °, 180 °, 216 °, 252 °, 288 °, and 324 °, respectively. And carrying out data analysis on the 10 binary images according to the fifth step to obtain plant phenotype information parameters including average values of plant height, base diameter and leaf area, wherein the following table shows results of the plant phenotype information parameters.
As can be seen from the table, when the plant phenotype information is acquired at a single angle in the prior art, projection information obtained due to the shielding of branches and leaves or the spatial stretching and curling of leaves is one-sided, inaccurate and large in error, and each leaf cannot be picked for destructive measurement in the automatic rapid measurement process, so that the 10-angle averaging method of the embodiment ensures the acquisition efficiency and non-destructive measurement, and improves the acquisition and extraction accuracy of the phenotype information.
The present invention and its embodiments have been described above schematically, without limitation, and what is shown in the drawings is only one of the embodiments of the present invention, and the actual structure is not limited thereto. Therefore, if the person skilled in the art receives the teaching, without departing from the spirit of the invention, the person skilled in the art shall not inventively design the similar structural modes and embodiments to the technical solution, but shall fall within the scope of the invention.
Claims (3)
1. A plant phenotype information extraction method is characterized in that: acquiring phenotypic information of a plant (500) with a plant phenotypic information acquisition system comprising:
a carrying platform (100) for placing plants (500);
the information acquisition device (200) comprises a visible light camera (210) and an adjusting mechanism (220), wherein the visible light camera (210) is arranged at the upper end of the adjusting mechanism (220), and the visible light camera (210) faces the bearing table (100) and is used for acquiring the phenotype information of the plant (500);
the rotating device (300) comprises a rotating ring (310), a supporting seat (320) and a driving trolley (330), the rotating ring (310) is arranged on the bearing platform (100) and can rotate relative to the bearing platform (100), one end of the supporting seat (320) is connected with the rotating ring (310), the driving trolley (330) is arranged at the other end of the supporting seat (320), and the information acquisition device (200) is installed on the supporting seat (320);
the extraction method comprises the following steps:
firstly, placing a container filled with plants (500) on a bearing table (100), and adjusting the illumination intensity and the illumination direction of ambient light;
secondly, adjusting the position of the visible light camera (210) through an adjusting mechanism (220);
step three, the rotating device (300) drives the information acquisition device (200) to rotate for 1 circle around the bearing table (100), and the rotating time is recorded; meanwhile, calibrating and correcting the acquired image through 6 calibration plates (420) on the circumferential direction of the curtain (410), and calculating a projection matrix of the visible light camera to obtain a geometric model of the visible light camera;
step four, setting the shooting time of the visible light camera (210) according to the rotation time in the step three;
step five, the rotating device (300) drives the information acquisition device (200) to rotate for 1 circle around the bearing table (100), the visible light camera (210) shoots the plant for 10 times according to the shooting time in the step four, and the average value of the plant phenotype information is calculated;
in the fifth step, the specific step of calculating the mean value of the plant phenotype information is,
s1, segmenting a region of the plant (500) from the RGB image;
s2, converting the RGB image of the plant (500) region in the S1 into a single-waveband image by using the index 2G/(R + B);
s3, converting the single-waveband image obtained in the S2 into a binary image, and screening and denoising the binary image;
s4, respectively calculating plant phenotype information parameters of the plant (500) in the 10 binary images, including plant height, basal diameter and leaf area, and calculating the average value of the plant phenotype information parameters.
2. The method for extracting phenotypic information of plant as claimed in claim 1, wherein: before the first step, the bearing table (100) is arranged on the curtain (410), and 6 calibration plates (420) are equidistantly arranged on the circumferential direction of the curtain (410).
3. The method for extracting phenotypic information of plant as claimed in claim 1, wherein:
when the plant height is calculated in the S4, setting a polygon limit frame selection range, using a global image threshold value, scanning line by line, wherein the distance from the continuous highest position of pixels at the top of the plant to the upper edge of the pot is the plant height of the plant (500);
when the base diameter is calculated, scanning line by line from top to bottom, identifying the container, taking the part which is close to the upper edge of the container and has continuous pixels as the stem, and calculating the width of the stem part as the base diameter of the plant (500);
when calculating the leaf area, a structural unit is arranged, isolated noise is eliminated by using a form opening technology, and the number of white pixels is calculated to be the projection area of the plant (500) leaves.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011447395.1A CN112964202B (en) | 2019-06-14 | 2019-06-14 | Plant phenotype information acquisition system and extraction method |
| CN201910514370.XA CN110274556B (en) | 2019-06-14 | 2019-06-14 | Plant phenotype information extraction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910514370.XA CN110274556B (en) | 2019-06-14 | 2019-06-14 | Plant phenotype information extraction method |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011447395.1A Division CN112964202B (en) | 2019-06-14 | 2019-06-14 | Plant phenotype information acquisition system and extraction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110274556A CN110274556A (en) | 2019-09-24 |
| CN110274556B true CN110274556B (en) | 2020-12-22 |
Family
ID=67960833
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011447395.1A Active CN112964202B (en) | 2019-06-14 | 2019-06-14 | Plant phenotype information acquisition system and extraction method |
| CN201910514370.XA Active CN110274556B (en) | 2019-06-14 | 2019-06-14 | Plant phenotype information extraction method |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011447395.1A Active CN112964202B (en) | 2019-06-14 | 2019-06-14 | Plant phenotype information acquisition system and extraction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN112964202B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111288272B (en) * | 2020-03-17 | 2021-02-02 | 闫晓 | Garden vegetation information collection device based on database |
| CN111561873B (en) * | 2020-05-27 | 2021-07-30 | 南京林业大学 | Self-propelled seedling tree trunk phenotype information acquisition system and acquisition method thereof |
| CN111553943A (en) * | 2020-06-09 | 2020-08-18 | 山西农业大学 | Leaf area automatic measurement system based on multivariate normal distribution statistics and level set variational method |
| CN111623707A (en) * | 2020-06-11 | 2020-09-04 | 南京农业大学 | Plant overground part phenotype data acquisition and sample space coordinate measuring platform |
| CN111972159B (en) * | 2020-07-17 | 2022-07-05 | 武汉爱农云联科技有限公司 | Intelligent planter and intelligent illumination method thereof |
| CN112610828A (en) * | 2020-12-29 | 2021-04-06 | 河南科技大学 | Wheat image acquisition equipment in field |
| CN112710663A (en) * | 2021-01-17 | 2021-04-27 | 无锡职业技术学院 | High-throughput measurement system and measurement method for phenotype information of plant whole life cycle |
| CN113155192B (en) * | 2021-04-19 | 2022-09-16 | 河南省农业科学院农业经济与信息研究所 | Intelligent crop table type real-time acquisition device |
| CN113822896B (en) * | 2021-08-31 | 2024-04-23 | 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 |
| CN114339005B (en) * | 2022-01-06 | 2024-04-02 | 南京林业大学 | Movable plant phenotype information automatic acquisition platform based on multi-view images |
| CN114339004B (en) * | 2022-01-06 | 2024-03-29 | 南京林业大学 | Phenotype platform for comprehensively acquiring morphological structure information of multiple organs of plant |
| CN114757854B (en) * | 2022-06-15 | 2022-09-02 | 深圳市安星数字系统有限公司 | Night vision image quality improving method, device and equipment based on multispectral analysis |
| CN117388264B (en) * | 2023-12-13 | 2024-02-13 | 慧诺云谱(海南)科技有限公司 | High-flux plant root phenotype information acquisition and measurement system and method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103471523A (en) * | 2013-09-30 | 2013-12-25 | 北京林业大学 | Method for detecting profile phenotype of arabidopsis |
| CN204666144U (en) * | 2015-05-11 | 2015-09-23 | 中国科学院上海生命科学研究院 | A kind of plant phenotype scanister |
| CN207095515U (en) * | 2017-08-09 | 2018-03-13 | 南京林业大学 | A kind of phytomorph three-dimensional image acquisition apparatus |
| CN109870118A (en) * | 2018-11-07 | 2019-06-11 | 南京林业大学 | A kind of point cloud acquisition method of Oriented Green plant temporal model |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2719273B1 (en) * | 2012-10-12 | 2018-02-28 | Nederlandse Organisatie voor toegepast- natuurwetenschappelijk onderzoek TNO | Method and apparatus for cutting plants |
| WO2015006675A2 (en) * | 2013-07-11 | 2015-01-15 | Blue River Technology, Inc. | Method for automatic phenotype measurement and selection |
| CN106546568B (en) * | 2016-10-31 | 2019-04-16 | 浙江大学 | A kind of method and device obtaining plant three-dimensional chlorophyll fluorescence image information |
| CN107133976B (en) * | 2017-04-24 | 2020-04-14 | 浙江大学 | Method and device for acquiring three-dimensional hyperspectral information |
| CN109238166A (en) * | 2018-07-17 | 2019-01-18 | 北京农业信息技术研究中心 | A kind of plant phenotype acquisition device and its acquisition method |
-
2019
- 2019-06-14 CN CN202011447395.1A patent/CN112964202B/en active Active
- 2019-06-14 CN CN201910514370.XA patent/CN110274556B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103471523A (en) * | 2013-09-30 | 2013-12-25 | 北京林业大学 | Method for detecting profile phenotype of arabidopsis |
| CN204666144U (en) * | 2015-05-11 | 2015-09-23 | 中国科学院上海生命科学研究院 | A kind of plant phenotype scanister |
| CN207095515U (en) * | 2017-08-09 | 2018-03-13 | 南京林业大学 | A kind of phytomorph three-dimensional image acquisition apparatus |
| CN109870118A (en) * | 2018-11-07 | 2019-06-11 | 南京林业大学 | A kind of point cloud acquisition method of Oriented Green plant temporal model |
Non-Patent Citations (1)
| Title |
|---|
| 基于计算机视觉技术的油菜冠层营养信息监测;袁道军 等;《农业工程学报》;20091230;第25卷(第12期);174-179 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112964202B (en) | 2023-05-09 |
| CN112964202A (en) | 2021-06-15 |
| CN110274556A (en) | 2019-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110274556B (en) | Plant phenotype information extraction method | |
| US20210056685A1 (en) | Method and device for monitoring comprehensive growth of potted lettuce | |
| US11436824B2 (en) | Water stress detection method for tomatoes in seedling stage based on micro-CT and polarization-hyperspectral imaging multi-feature fusion | |
| CN109816680B (en) | High-throughput calculation method for crop plant height | |
| WO2021082655A1 (en) | Rhizo-box for acquiring crop phenotype | |
| CN102565061B (en) | Crop biomass nondestructive testing image acquisition and processing device and testing method | |
| CN103630091B (en) | Leaf area measurement method based on laser and image processing techniques | |
| WO2021082656A1 (en) | High-throughput photographing system for acquiring crop phenotypes | |
| CN112461135B (en) | Dendrobium growth parameter nondestructive online measuring device and measuring method thereof | |
| CN109191520B (en) | Plant leaf area measuring method and system based on color calibration | |
| CN110866975A (en) | Multi-vision-based rape image acquisition device and three-dimensional feature extraction method | |
| CN113607659B (en) | Conveyor belt type crop phenotype acquisition method, system and device | |
| CN116147490B (en) | High-flux plant phenotype information acquisition and measurement system and method | |
| CN109974854A (en) | A kind of radiation correction method of frame width formula FPI high spectrum image | |
| CN111693551A (en) | Nondestructive measurement device and method for three-dimensional characters of rice plants and roots | |
| Ahamed et al. | Tower remote-sensing system for monitoring energy crops; image acquisition and geometric corrections | |
| CN110736750B (en) | Wheat scab detection method based on multi-angle field high-definition imaging | |
| CN112036407A (en) | Plant population three-dimensional phenotype in-situ acquisition method and device | |
| CN113706512A (en) | Live pig weight measurement method based on deep learning and depth camera | |
| CN106023235A (en) | Crop effective grain number measuring method | |
| WO2021184667A1 (en) | Channel-based crop root phenotype acquisition system | |
| CN111896045A (en) | Greenhouse crop three-dimensional directional sensing and fine-grained automatic acquisition device and method | |
| CN113114766B (en) | Potted plant information detection method based on ZED camera | |
| CN210374998U (en) | High-flux photographing system for acquiring crop phenotype | |
| CN210298680U (en) | A cultivate and measuring device for high flux plant roots visual analysis |
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 |