CN103632157B - Individual plant wheat fringe portion kernal number method of counting - Google Patents
Individual plant wheat fringe portion kernal number method of counting Download PDFInfo
- Publication number
- CN103632157B CN103632157B CN201210304177.1A CN201210304177A CN103632157B CN 103632157 B CN103632157 B CN 103632157B CN 201210304177 A CN201210304177 A CN 201210304177A CN 103632157 B CN103632157 B CN 103632157B
- Authority
- CN
- China
- Prior art keywords
- wheat
- fringe portion
- kernal number
- area
- image
- 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
Landscapes
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of wheat fringe portion kernal number method of counting, the method comprising the steps of：S1. using the front view of CCD camera collection wheat fringe portion, and the front view to collecting pre-processes；S2. mathematical morphology is analyzed：The small part of structural element is reasonably removed using opening operation in morphology processing；Operation is filled to each cavity, using area-method by the object whole impurity elimination less than area P；S3. wheat fringe portion characteristic parameter extraction：To establish the dependency relation between wheat fringe portion relevant feature parameters and kernal number, the area features and cob length characteristic of wheat fringe portion are extracted；S4. founding mathematical models：Mathematical modeling between wheat fringe portion area and cob length characteristic parameter and its kernal number is established using least square method；S5. mathematical modeling is examined：Dependency relation between wheat fringe portion area features and major axis feature and its kernal number is verified, it is ensured that the practicality of both models.The method of the present invention can be by extracting wheat fringe portion image area feature and long axis length feature, establish mathematical modeling between fringe portion image area feature and long axis length feature and its kernal number, so as to relatively accurately objectively count out wheat fringe portion kernal number rapidly, the efficiency of fringe portion seed counting is greatly improved compared with artificial.
Description
Technical field
The present invention relates to a kind of method that individual plant wheat fringe portion kernal number is counted out using digital image processing techniques.
Background technology
The quantity of all plump seeds of fringe portion when wheat fringe portion kernal number is exactly wheat maturation.
Ultimate output during Wheat Production is the thing that agricultural producer most pays close attention to all the time, is actually to be easy to wheat
Postpartum management processing, it is that the producer and the urgent of agricultural management department are essential all the time accurately to know that wheat yield is assessed as early as possible
Will.Therefore wheat field Granule weight need it is accurate when knowing wheat seed maturation on the wheat head specific seed number.Agronomy
When breeding and cultivation expert carry out laboratory species test, it is also desirable to it is quick to know wheat fringe portion kernal number amount, accelerate wheat breeding work
Make.On the basis of by counting the kernal number on the wheat head, according to the mass of 1000 kernel of this kind and other relevant parameters, so as to very fast
Crop field wheat yield is speculated exactly.
It is as follows to survey production method for agricultural at this stage：Sampling point area and investigation content.When theory surveys production, every takes 1 square metre, adjusts
Look into a mu spike number；Take 20 fringes at random in each sampling point, investigate grain number per spike；Mass of 1000 kernel is calculated by the long-term mass of 1000 kernel of the kind.
Theoretical yield calculates according to below equation：Per mu yield (kg/acre)=mu spike number × grain number per spike × mass of 1000 kernel ×
0.85。
It follows that be that theoretical yield measures one of important basic step to wheat fringe portion kernal number, wheatear at this stage
Portion's kernal number is obtained by artificial counting method, and so as to obtain the yield of wheat, this method counting is time-consuming, arduously,
Certain error is equally existed in counting process, different survey production technical staff may obtain different results, and subjectivity influences
It is larger, the shortcomings of accuracy difference.
Therefore developing a kind of fast and accurately machine counting wheat fringe portion kernal number has realistic meaning, and to reach machine
Count wheat fringe portion kernal number, it is necessary first to the method that wheat fringe portion kernal number is counted automatically is determined, it is small accurately obtaining
On the basis of wheat head portion kernal number, the per mu yield of wheat is obtained according to the mass of 1000 kernel of wheat seed and mu spike number.Wheat fringe portion seed
The characteristic parameters such as grain number and wheat head properties and characteristicses cob length and fringe portion area are relevant, therefore the present invention passes through digital image processing
Counting acquisition wheat fringe portion kernal number has theoretical foundation.
The content of the invention
(1) technical problems to be solved
The technical problem to be solved in the present invention is：One kind is provided and counts out wheat fringe portion seed using digital image processing techniques
The method of grain number, the area and cob length of wheat head seed region are calculated by the method for Digital Image Processing, established
Dependency relation between fringe portion seed region area and cob length and its fringe portion seed number.
(2) technical scheme
To achieve the above object, the invention provides a kind of method of counting of wheat fringe portion kernal number, this method to include step
Suddenly：
S1. using the front view of CCD camera collection wheat fringe portion, and the front view to collecting pre-processes；
S1 specific steps include：Image obtains, gray processing processing, binary conversion treatment, wherein pretreatment includes gray processing
Processing and two steps of binary conversion treatment.
S1.1 images are obtained mainly using the front view of CCD camera collection wheat fringe portion, gather wheat fringe portion image
Device have, CCD camera, objective table, computer etc., illumination condition is natural lighting.Wherein objective table table top is white bottom
Plate, CCD camera are suspended on above objective table by frame, so as to which frame links into an integrated entity objective table and CCD camera.
Under natural lighting, individual plant wheatear is placed on objective table according to front, objective table table top is color uniformly white negative,
Take same angle CCD camera to be taken pictures from top, gather image when take same light intensity, same angle (vertical), same focal length, together
One height.
The processing of S1.2 gray processings：Gray processing processing is carried out to coloured image, the gray level image of the wheat head is obtained, by gray level image
Obtain the histogram of image.
S1.3 binary conversion treatments：Background area and the threshold value that wheat head area grayscale is distributed are determined by histogram, used
Otsu method choices maximize the threshold value of inter-class variance, so as to which wheat head region and background area are separated, obtain the wheat head and the back of the body
The distinct image of scene area.
S2. mathematical morphology is analyzed：Morphology processing is carried out to pretreated front view picture, using mathematics shape
Opening operation processing reasonably removes the small part of structural element in state credit analysis；Operation is filled to each cavity, utilizes face
Area method is by the object whole impurity elimination less than area P；
S2 specific steps include：Expansion and the combinatorial operation of corrosion, image completion and impurity elimination processing
S2.1 expands and the combinatorial operation of corrosion：Opening operation is employed in image processing process, and (computing is first to image
Corroded, then the image after to corrosion expands, and operates with same structural element twice) handle image.
The awn of wheat or miscellaneous point after morphology processing in the image of (opening operation) have obvious reduction, smaller than structural element
Part reasonably remove.
S2.2 image completions and impurity elimination processing：Operation is filled to bianry image.For bianry image, it will connect
The value of background pixel 0 be changed to foreground pixel 1 and be worth, the border until reaching object.One important application of padding is pair
Image is carried out filling out hole operation, and operation is filled to each hole.Unnecessary object is removed using area-method simultaneously, area will be less than
P object whole impurity elimination.It is this time that the bianry image after opening operation is filled, goes miscellaneous operation in image processing process.
It is clear, clear in image after processing.Only remaining wheat head part, the i.e. awn of wheat, miscellaneous point etc. rationally remove, are follow-up
The extraction of characteristic parameter provides advantage.
S3. wheat fringe portion characteristic parameter extraction：Wheat fringe portion image features are wheat fringe portion image fundamental characteristics
Basic description, it is the theoretical foundation for the mathematical modeling established between wheat fringe portion relevant feature parameters and kernal number, the present invention is only
Area features and cob length characteristic only to wheat fringe portion extract.
The specific step of S3 wheat fringe portion characteristic parameter extractions includes：Wheat fringe portion area features parameter extraction and wheatear
Portion's cob length characteristic extraction
S3.1 area features parameter extractions：It is pixel value 1 for only surplus wheat head part to be filled through with the bianry image of impurity elimination
Region, the area that the number of pixels that pixel value is 1 part is wheat head part is calculated using image processing techniques.
S3.2 cobs length characteristic is extracted：Wheat head image cob length can be obtained by connecting two points on ellipse
Nose section.
S4. founding mathematical models：Mathematical modulo between wheat fringe portion characteristic parameter and its kernal number is established using least square method
Type.
S4, which establishes mathematical modeling between wheat fringe portion characteristic parameter and its kernal number, includes step：Wheat fringe portion area features are joined
Several mathematical modelings between kernal number on the actual wheat head, wheat fringe portion cob length characteristic parameter and kernal number on the actual wheat head
Between mathematical modeling.
The dependency relation that S4.1 is established between area features parameter and its kernal number：Established by least square method at image
Reason calculates the mathematical modeling between kernal number on obtained area features parameter and the actual wheat head.
The dependency relation that S4.2 is established between cob length parameter and its kernal number：Established by least square method at image
Reason calculates the mathematical modeling between kernal number on obtained cob length characteristic parameter and the actual wheat head.
S5. mathematical modeling is examined：To the mathematics between wheat fringe portion area features and cob length characteristic and its kernal number
Model is verified, it is ensured that the practicality of both models.
Related Mathematical Models verification step between S5 wheat fringe portion kernal numbers and its feature includes：Wheat fringe portion area is special
Mathematical modeling between sign and its kernal number verifies that the mathematical modeling between wheat fringe portion cob length characteristic and its kernal number is tested
Card.
Model between S5.1 wheats fringe portion area features parameter and kernal number is tested：With the kernal number of sample survey
Measure as standard to weigh the seed quantity of the wheatear obtained by mathematical modeling.With the checking wheat samples collection actually obtained
Seed quantity as foundation between actual value and the wheat fringe portion kernal number amount predicted value obtained by mathematical modeling it is related close
It is curve, and compares both difference degrees.
Model between S5.2 wheatears cob length characteristic parameter and kernal number is tested：Using identical with S5.1
Method the model between wheatear cob length and kernal number is tested.
(3) beneficial effect
The present invention by wheat fringe portion front view as gray processing and binary conversion treatment, with Mathematical Morphology Method handle
The awn of wheat is split with the wheat head, so as to obtain only wheat fringe portion principal part image, so as to continue with image processing method extraction
Fringe portion area features and cob length characteristic.Established by the wheat fringe portion area features and cob length characteristic parameter that have extracted
Related Mathematical Models between its kernal number obtained with artificial counting, further as model verification method prove obtained by mathematics
Model accuracy is higher, can effectively substitute artificial counting.
Brief description of the drawings
Fig. 1 wheatear kernal number number system courses of work
The wheat fringe portion front view picture of Fig. 2 collections
The gray processing processing of Fig. 3 coloured images
Fig. 4 grey level histograms
Fig. 5 gray-scale maps are to binary picture
Binary map after the combinatorial operation of Fig. 6 expansions and corrosion
Fig. 7 is filled and the image after impurity elimination
Embodiment
Wheat fringe portion seed method of counting proposed by the present invention, describe in detail in conjunction with the accompanying drawings and embodiments as follows.
For quick obtaining wheat fringe portion area features parameter and cob length characteristic parameter, area features parameter and fringe are established
Dependency relation between shaft length characteristic parameter and its kernal number, therefore key point has two, one is quick obtaining wheatear
Portion's area features and cob length characteristic parameter, another is that the two characteristic parameters and its fringe portion kernal number are established mathematical modulo
Type, so as to establish basis for quick obtaining wheat fringe portion kernal number.
According to Fig. 1, the method counted according to a kind of quick obtaining wheat fringe portion kernal number of the present invention includes step：
S1. using the front view of CCD camera collection wheat fringe portion, and the front view to collecting pre-processes；
S1 specific steps include：Image obtains, gray processing processing, binary conversion treatment
S1.1 images obtain：Image is obtained mainly using the front view of CCD camera collection wheat fringe portion, gathers wheat
The device of fringe portion image has, CCD camera, objective table, and computer etc., illumination condition is natural lighting.Wherein objective table table top
For white bottom plate, CCD camera is suspended on above objective table by frame, so as to which frame connects objective table with CCD camera
Integrally.Under natural lighting, individual plant wheatear is placed on objective table according to front, objective table table top is that color is uniform
White negative, take same angle CCD camera to be taken pictures from top, gather image when take same light intensity, same angle (vertical), together
One focal length, sustained height.Obtained according to the above method and be illustrated in figure 2 wheat fringe portion front view picture.
The pretreatment of wheat fringe portion front view picture includes gray processing processing and two steps of binary conversion treatment.
The processing of S1.2 gray processings：Gray processing processing is carried out to coloured image, obtains the gray level image of the wheat head, wheat fringe portion is just
View image handles to obtain image as shown in Figure 3 through gray processing, and as can be seen from the figure gray level image effectively saves the wheat head
The details in region, edge clear.The histogram of gray level image is obtained by gray level image.The histogram of gray level image 3 such as Fig. 4 institutes
Show.The gray distribution features of wheat head region and background area are distinguished obvious in histogram, are easily separated.
S1.3 binary conversion treatments：Background area and the threshold value that wheat head area grayscale is distributed are determined by histogram, used
Otsu method choices maximize the threshold value of inter-class variance, so as to which wheat head region and background area are separated, obtain the wheat head and the back of the body
The distinct image of scene area.It is as shown in Figure 5 through binary image obtained by binary conversion treatment.
S2. mathematical morphology is analyzed：Morphology processing is carried out to pretreated front view picture, using mathematics shape
Opening operation processing reasonably removes the small part of structural element in state credit analysis；Operation is filled to each cavity, utilizes face
Area method is by the object whole impurity elimination less than area P；Edge extraction processing based on mathematical morphology is better than transporting based on differential
The Boundary extracting algorithm of calculation, edge smoother low to noise susceptibility, while extracting.Form filter based on mathematical morphology
Ripple device can effective filter out noise by means of the geometric properties Information Pull morphological operator of priori, can retain again in image
Original information.
S2 specific steps include：Expansion and the combinatorial operation of corrosion, image completion and impurity elimination
S2.1 expands and the combinatorial operation of corrosion
Opening operation is employed in image processing process, and (computing is corroded to image first, then after to corrosion
Image is expanded, operate with same structural element twice) handle image.This time using circle in processing procedure
Structural element (disk), radius R take 7, and result is as shown in Figure 6.
The awn of wheat or miscellaneous point after morphology processing in the image of (opening operation) have obvious reduction, smaller than structural element
Part reasonably remove, while wheat fringe portion fringe stalk also removed by this computing.
S2.2 image completions and impurity elimination
Operation is filled to bianry image.For bianry image, the value of background pixel 0 of connection is changed to prospect picture
Element 1 is worth, the border until reaching object.One important application of padding is that image is carried out to fill out hole operation, to each
Hole is filled operation.Unnecessary object is removed using area-method simultaneously, by the object whole impurity elimination less than area P.This time scheme
As being the bianry image after opening operation is filled, to remove miscellaneous operation, result as shown in Figure 7 in processing procedure
It is clear, clear in image after processing.Only remaining wheat head part, the i.e. awn of wheat, miscellaneous point etc. rationally remove, are follow-up
The extraction of characteristic parameter provides advantage.
S3. wheat fringe portion characteristic parameter extraction：Wheat fringe portion image features are wheat fringe portion image fundamental characteristics
Basic description, it is the theoretical foundation for the mathematical modeling established between wheat fringe portion relevant feature parameters and kernal number, the present invention is only
Area features and cob length characteristic only to wheat fringe portion extract.
The specific step of S3 wheat fringe portion characteristic parameter extractions includes：Wheat fringe portion area features amount calculates and wheat fringe portion
Cob length characteristic is extracted
S3.1 area features parameter extractions：It is pixel value 1 for only surplus wheat head part to be filled through with the bianry image of impurity elimination
Region, the area that the number of pixels that pixel value is 1 part is wheat head part is calculated using image processing techniques.Calculate in border
Area of portion's pixel count as its image Grain parts.
S3.2 cob length characteristic parameter extractions：Wheat head image cob length is by connecting two point institute energy on ellipse
The nose section of acquisition.
S4. founding mathematical models：Founding mathematical models are between wheat fringe portion image characteristic parameters and its time kernal number
Dependency relation, establish mathematical modeling between wheat fringe portion characteristic parameter and its kernal number using least square method.
The dependency relation that S4.1 is established between area features parameter and its kernal number：Established by least square method at image
Reason calculates the mathematical modeling between kernal number on the area features parameter obtained and the actual wheat head.
In order to count the seed number on the wheat head, this method is built using the size for identifying same kind wheat head image
Vertical relation between wheat head seed number and its image area, using sample of 100 plants of wheatear heads as founding mathematical models.
The dependent linearity model established by the method between wheat head image area characteristic parameter and wheatear kernal number is：Y=-
1.7871+0.0007x R=0.9583.
The dependency relation that S4.2 is established between cob length parameter and its kernal number：Established by least square method at image
Reason calculates the mathematical modeling between kernal number on obtained cob length characteristic parameter and the actual wheat head.
In order to count the seed number on the wheat head, this method, which uses, identifies same kind wheat head image cob length scale,
The relation established between wheat head seed number and its image cob length, using 100 plants of wheatear heads as founding mathematical models
Sample.The relative coefficient R that wheatear cob length characteristic parameter and its seed quantity are established by the method is 0.9474,
Mathematical modeling expression formula is Y=0.1089X-24.958, then it is very notable to illustrate that wheatear cob length and its seed quantity have
Linear relationship.
S5. mathematical modeling is examined：To the mathematics between wheat fringe portion area features and cob length characteristic and its kernal number
Model is verified, it is ensured that the practicality of both models.
Related Mathematical Models verification step between S5.1 wheat fringe portion kernal numbers and its feature includes：Wheat fringe portion area
Mathematical modeling checking between feature and its kernal number, the mathematical modeling between wheat fringe portion cob length characteristic and its kernal number
Checking.40 verification samples (exceed test specimen 1/3rd) are employed in the invention to examine above-mentioned mathematical modeling
Test, 40 samples belong to same kind with test specimen, and utilize same image capturing system, and image processing method is with establishing
The method used during mathematical modeling is consistent.
S5.1 tests to the model between wheatear area and kernal number
Coefficient R between wheat fringe portion kernal number actual value and predicted value is 0.9781, and close to 1, then explanation should
Model can obtain the seed quantity of wheatear by Digital Image Processing.
Wheat samples of the error between 12%-16% only account for 2.50% (i.e. a sample) of total verification sample, 40
The error rate of verification sample is largely focused on 0-4%, 4%-8%, 8%-12%, and by Tu Ke get, error rate is below 12%
Wheat samples account for the 97.5% of total verification sample.
S4.2 tests to the model between wheatear cob length and kernal number
The model between wheatear cob length and kernal number is tested using identical method.
Coefficient R between wheat fringe portion kernal number actual value and predicted value is 0.9501, and close to 1, then explanation should
Model can obtain the seed quantity of wheatear by Digital Image Processing.
Wheat samples of the error between 12%-16% account for 12.50% (i.e. 5 samples) of total verification sample, and 40 are tested
The error rate of card sample is largely focused on 0-4%, 4%-8%, 8%-12%, and analysis can obtain, and error rate is below 12%
Wheat samples account for the 87.5% of total verification sample.The percentage gone out relative to a upper model testing is relatively low.
Find out from above two model testing result, the model testing result between wheat fringe portion area and kernal number is better than
Model testing result between wheatear shaft length and kernal number.
Above wheat fringe portion kernal number method of counting is to be directed to awned wheat fringe portion, and for awnless wheat fringe portion kernal number
Method of counting is equally applicable.
Claims (4)
1. a kind of wheat fringe portion kernal number method of counting, the method comprising the steps of：
S1. the front view picture of the wheat fringe portion to collecting extracts wheat fringe portion image seed region area after image procossing
Characteristic parameter and cob length characteristic parameter；
S2. the mathematical modeling established between wheat head image area and wheatear kernal number is：Y=-1.7871+0.0007x, the wheat head
Coefficient R=0.9583 of image area characteristic parameter and wheatear kernal number；Establish wheatear cob length characteristic parameter
Mathematical modeling with its seed quantity is y=0.1089x-24.958, wheatear cob length characteristic parameter and its seed quantity
Relative coefficient R=0.9474；
S3. the mathematical modeling between wheat fringe portion area features and cob length characteristic and its kernal number is verified, it is small
Coefficient R is 0.9781 between the wheat fringe portion kernal number actual value and predicted value of wheat head portion area features mathematical modeling, by mistake
Wheat samples of the rate below 12% account for the 97.5% of total verification sample；Wheat fringe portion cob length characteristic mathematical modeling it is small
Coefficient R is 0.9501 between wheat head portion kernal number actual value and predicted value, and wheat samples of the error rate below 12% account for
The 87.5% of total verification sample.
2. wheat fringe portion kernal number method of counting according to claim 1, it is characterised in that the fringe portion in step S1
Image area characteristic parameter and cob length characteristic parameter extraction include step：
S1.1 wheat fringe portion image area characteristic parameter extractions：To the front view picture of wheat fringe portion that collects through image procossing
The only bianry image of wheat head seed subregion and background area is obtained afterwards, and it is wheat to calculate wheat head seed area pixel number
The area features amount of fringe image；
S1.2 cob length characteristic parameter extractions：The front view picture of wheat fringe portion obtains only wheat head seed after image procossing
Subregion and the bianry image of background area, cob length are by two points on wheat head seed subregion connection ellipse
It can be obtained nose segment length.
3. wheat fringe portion kernal number method of counting according to claim 1, it is characterised in that the foundation in step S2
Mathematical modeling between wheat fringe portion characterisitic parameter and its kernal number includes step：
The Related Mathematical Models that S2.1 is established between wheat fringe portion image area characteristic parameter and kernal number：Built using least square method
Vertical image procossing calculates the mathematical modeling between the area features amount obtained and actual wheat fringe portion kernal number：Y=-1.7871+
0.0007x, the relative coefficient R=0.9583 of wheat head image area characteristic parameter and wheatear kernal number；
The Related Mathematical Models that S2.2 is established between wheat fringe portion image cob length characteristic parameter and kernal number：Utilize least square
The mathematical modeling that method is established between the cob length characteristic amount of image procossing calculating acquisition and actual wheat fringe portion kernal number：Y=
0.1089x-24.958, the relative coefficient R=0.9474 of wheatear cob length characteristic parameter and its seed quantity.
4. wheat fringe portion kernal number method of counting according to claim 1, it is characterised in that the checking in step S3
Mathematical modeling includes step：
S3.1 tests to the model between wheat fringe portion area and kernal number：With the actual seed for obtaining checking wheat samples
The wheat fringe portion kernal number amount that quantity obtains as actual value and by mathematical modeling establishes dependency relation song between predicted value
Line, and compare both difference degrees, the coefficient R between wheat fringe portion kernal number actual value and predicted value is 0.9781,
Wheat samples of the error rate below 12% account for the 97.5% of total verification sample；
S3.2 tests to the model between wheatear cob length and kernal number：To wheatear cob length and kernal number it
Between model test, the coefficient R between wheat fringe portion kernal number actual value and predicted value be 0.9501, error rate exists
Less than 12% wheat samples account for the 87.5% of total verification sample, and the practicality of both models is ensure that by checking.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210304177.1A CN103632157B (en) | 2012-08-24 | 2012-08-24 | Individual plant wheat fringe portion kernal number method of counting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210304177.1A CN103632157B (en) | 2012-08-24 | 2012-08-24 | Individual plant wheat fringe portion kernal number method of counting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103632157A CN103632157A (en) | 2014-03-12 |
| CN103632157B true CN103632157B (en) | 2018-01-05 |
Family
ID=50213185
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210304177.1A Active CN103632157B (en) | 2012-08-24 | 2012-08-24 | Individual plant wheat fringe portion kernal number method of counting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103632157B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021369A (en) * | 2014-04-30 | 2014-09-03 | 南京农业大学 | Grain counting method for spike of single rice based on digital image processing technology |
| CN105427275B (en) * | 2015-10-29 | 2017-11-24 | 中国农业大学 | Crop field environment wheat head method of counting and device |
| CN105574853B (en) * | 2015-12-07 | 2018-05-15 | 中国科学院合肥物质科学研究院 | The method and system that a kind of wheat head grain number based on image recognition calculates |
| EP3529771A1 (en) | 2016-10-19 | 2019-08-28 | BASF Agro Trademarks GmbH | Determining the grain weight of an ear |
| CN107064146A (en) * | 2017-05-15 | 2017-08-18 | 柳州环山科技有限公司 | A kind of computer based wheat spike number measurement apparatus |
| CN107590812B (en) * | 2017-09-01 | 2021-07-02 | 南京农业大学 | Wheat ear identification and counting method |
| CN109447945B (en) * | 2018-09-21 | 2021-11-30 | 河南农业大学 | Quick counting method for basic wheat seedlings based on machine vision and graphic processing |
| CN113011296B (en) * | 2021-03-08 | 2024-09-24 | 中国农业科学院农产品加工研究所 | Method and system for rapidly identifying purity of wheat varieties |
| CN113470104A (en) * | 2021-07-21 | 2021-10-01 | 浙江托普云农科技股份有限公司 | Wheat ear grain counting and measuring method and system based on gray value comparison |
| CN114067312A (en) * | 2021-09-22 | 2022-02-18 | 南京农业大学 | Rape grain number per pod image identification method |
| CN113936056A (en) * | 2021-10-20 | 2022-01-14 | 广东皓行科技有限公司 | Image recognition method, yield estimation method and electronic device |
| CN115375694A (en) * | 2022-10-27 | 2022-11-22 | 浙江托普云农科技股份有限公司 | Portable rice whole ear measuring method based on image recognition and application thereof |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0072237B1 (en) * | 1981-08-11 | 1987-04-29 | De La Rue Systems Limited | Apparatus for scanning a sheet |
| JP3478329B2 (en) * | 1999-10-01 | 2003-12-15 | セイコーエプソン株式会社 | Image processing apparatus and image processing method |
| CN101281112A (en) * | 2008-04-30 | 2008-10-08 | 浙江理工大学 | Image type automatic analysis method for mesh adhesion rice corn |
| EP2329243A2 (en) * | 2008-08-22 | 2011-06-08 | Pioneer Hi-Bred International, Inc. | Apparatus and systems for counting corn silks or other plural elongated strands and use of the count for characterizing the strands or their origin |
| CN101673414B (en) * | 2009-10-19 | 2012-04-11 | 北京农业信息技术研究中心 | Three-dimensional morphological virtual modeling method of corn silk |
| CN102072883B (en) * | 2010-07-07 | 2013-03-13 | 北京农业智能装备技术研究中心 | Device and method for detecting comprehensive quality of crop seeds |
| CN101944231B (en) * | 2010-08-19 | 2012-01-04 | 北京农业智能装备技术研究中心 | Method for extracting wheatear morphological parameters |
-
2012
- 2012-08-24 CN CN201210304177.1A patent/CN103632157B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103632157A (en) | 2014-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103632157B (en) | Individual plant wheat fringe portion kernal number method of counting | |
| CN104021369A (en) | Grain counting method for spike of single rice based on digital image processing technology | |
| Hoffmaster et al. | An automated system for vigor testing three-day-old soybean seedlings | |
| Huang et al. | Rice panicle length measuring system based on dual-camera imaging | |
| CN109827957B (en) | Rice leaf SPAD value estimation method and system based on computer vision | |
| CN103955938A (en) | Wheat growing status diagnosing method based on mobile internet mode and leaf color analysis | |
| CN103630091B (en) | Leaf area measurement method based on laser and image processing techniques | |
| CN102855485B (en) | The automatic testing method of one grow wheat heading | |
| CN106340000A (en) | Bone age assessment method | |
| CN113920106B (en) | Corn growth vigor three-dimensional reconstruction and stem thickness measurement method based on RGB-D camera | |
| CN102663396B (en) | Method for automatically detecting rice milky ripe stage | |
| CN111462058A (en) | Method for quickly detecting effective ears of rice | |
| CN111881864A (en) | Wisdom farming crops growth full-period full-dynamic monitoring management system | |
| CN114136920A (en) | Hyperspectrum-based single-grain hybrid rice seed variety identification method | |
| CN108037123A (en) | A kind of hybrid paddy rice disc type sows performance parameter accurate detecting method | |
| CN112580671A (en) | Automatic detection method and system for multiple development stages of rice ears based on deep learning | |
| CN106023235B (en) | A kind of method that the effective seed number of crops measures | |
| CN110619417B (en) | Theoretical yield measurement method based on image characteristics of rice canopy spike in field | |
| CN107818565A (en) | A kind of wheat one is to tri-leaf period wheat seeding Censes Methods | |
| CN106172082A (en) | For the method analyzing poultry growth performance automatic measurement system noisy data | |
| CN116824405A (en) | Green-peel walnut target detection method based on improved YOLOv8 under unmanned aerial vehicle aerial photographing visual angle | |
| CN116597309A (en) | Method for identifying and selecting plant resistant varieties by integrating physiological indexes of germination period growth and seedling period | |
| Aung et al. | Weight estimation of mango from single visible fruit surface using computer vision | |
| CN109115771A (en) | Sugarcane technical maturity automatic observation process | |
| Wang et al. | Determining rapeseed lodging angles and types for lodging phenotyping using morphological traits derived from UAV images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |