CN102749290A - Method for detecting growth state of branches of crown canopy of cherry tree - Google Patents
Method for detecting growth state of branches of crown canopy of cherry tree Download PDFInfo
- Publication number
- CN102749290A CN102749290A CN2012102245485A CN201210224548A CN102749290A CN 102749290 A CN102749290 A CN 102749290A CN 2012102245485 A CN2012102245485 A CN 2012102245485A CN 201210224548 A CN201210224548 A CN 201210224548A CN 102749290 A CN102749290 A CN 102749290A
- Authority
- CN
- China
- Prior art keywords
- cherry tree
- growth conditions
- frequency spectrum
- tree canopy
- canopy branch
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a method for detecting growth state of branches of crown canopy of a cherry tree. The method comprises the following steps: (1) respectively collecting near infrared images of crown canopy branch samples from a plurality of cherry trees with different growth states; (2) carrying out fast Fourier transform on the near infrared images to obtain a two-dimensional spatial frequency spectrum of the near infrared images, dividing the two-dimensional spatial frequency spectrum into 10-25 rings with equal area, and calculating out a frequency spectrum value of each ring; (3) traversing all crown canopy branch samples of the cherry trees to obtain frequency spectrum values respectively corresponding to each crown canopy branch sample of the cherry trees, and establishing a model by using the frequency spectrum values as input and growth states of corresponding crown canopy branch samples of the cherry trees as output; and (4) acquiring a frequency spectrum value corresponding to crown canopy branches of a cherry tree to be tested, and substituting the frequency spectrum value into the model to obtain the growth state of the crown canopy branches of the cherry tree to be tested. The invention can realize rapid and accurate detection on the growth state of crown canopy branches of cherry trees.
Description
Technical field
The invention belongs to fruit tree growth status detection method field, relate in particular to a kind of detection method of cherry tree canopy branch growth conditions.
Background technology
The growth of crop be unable to do without photosynthetic influence, and wherein the structure influence of crop canopies intercepting and capturing and light the distribution situation in orchard of crop to light, therefore directly affects the output height of fruit.The output of fruit per unit area has been proved to be with crop the intercepting and capturing of light has been existed positive correlation, and light is intercepted and captured and is photodistributed inhomogeneous, all can reduce fruit size, reduce the output and the quality of fruit.The method that improves light intercepting and capturing amount has high-density planting, keeps the tree crown structure of suitable size and shape, and improves leaf area index etc., wherein the reasonably optimizing of crop canopies structure has been become an important technology index of orchard management.The crop canopies structure mainly receives the influence of technology of prunning branches, therefore efficiently accurately the growth conditions of crop is detected, and carries out rational beta pruning, is the precondition to the management of crop canopies structure rationalization.
At present at home; The means that crop growthing state is detected are backward relatively, and the overwhelming majority leans on peasant's experience to judge that this subjective assessment method receives condition effect such as personal experience, environment; Its objectivity, accuracy are relatively poor, need consume great amount of manpower and material resources simultaneously.
Have report to adopt spectral reflectance information, transmission information etc. that the canopy structure of crop is carried out inverting both at home and abroad, but because the limitation of spectral technique, the result receives the influence of different phase and canopy difference of different cultivars, growth of crop bigger easily.Employing spectral radiometers such as Yang Changming are observed the canopy spectra radiation energy of the different plant type rice varieties that are in same growth conditions; The result shows that the canopy spectra reflectivity of different plant type rice varieties exists the evident difference (can be referring to Yang Changming; Yang Linzhang; Wei Chaoling, the ultra dirt .2002. of fourth " comparative studies of different cultivars Rice Population canopy spectra characteristic " Chinese Journal of Applied Ecology 13,689-692.).Wu Chunxia etc. study the reflectance signature of cotton canopy; The result shows that the different cotton variety canopy spectra reflectivity that are in same growth conditions there are differences, and different breeding time difference obviously (can appoint the hilllock referring to Wu Chunxia; Zhao Junrong; Bai Li, Bai Shujun .2008. " based on the cotton canopy reflectance signature research of high spectral technique " agricultural and technology 4,56-60.).
Summary of the invention
The object of the present invention is to provide a kind of detection method of the cherry tree canopy branch growth conditions based on multi-optical spectrum imaging technology; It is big to have solved in the testing process of cherry tree canopy branch growth conditions error, receives blade, canopy shape and kind to influence big problem.
A kind of cherry tree canopy branch growth conditions detection method may further comprise the steps:
(1) gathers the near-infrared image of the cherry tree canopy branch sample of several different growth conditions respectively;
(2) near-infrared image is carried out Fast Fourier Transform (FFT), obtain the two-dimensional space frequency spectrum of near-infrared image, this two-dimensional space frequency spectrum is divided into the annulus of 10~25 homalographics, and try to achieve the frequency spectrum value of each annulus;
(3) travel through all cherry tree canopy branch samples, obtain respectively and the corresponding frequency spectrum value of each cherry tree canopy branch sample;
Said frequency spectrum value is input, is output with the growth conditions of the cherry tree canopy branch sample of correspondence, sets up model;
(4) obtain and the corresponding frequency spectrum value of cherry tree canopy branch to be measured according to step (1) and (2), the said model of substitution obtains the growth conditions of cherry tree canopy branch to be measured.
Said near-infrared image is to be the image that the near-infrared band of 780~1100nm obtains at wavelength, and said cherry tree is the cherry tree that is in the growth animated period, and the age of tree surpasses the cherry tree in 1 year and is its growth animated period in annual 3~September.
In the said step (1); The growth conditions of cherry tree canopy branch can be divided into three types: inabundant beta pruning (under-pruning) state; Suitable beta pruning (premium-pruning) state, excessive beta pruning (over-pruning) state, wherein; Fully the beta pruning state is not meant that cherry tree canopy branch is long or overstocked, is unfavorable for the state that light is intercepted and captured; Suitable beta pruning state is meant that cherry tree canopy branch length, density suit, and help the state that light is intercepted and captured; Excessively the beta pruning state is meant that cherry tree canopy branch is too short or too sparse, is unfavorable for the state that light is intercepted and captured.The principle of demarcating cherry tree canopy branch growth conditions is following: the elongated shoot of master, side shoot, and when increment generally reached 40~50 centimetres, the suitable scope that deducts was 1/4~1/3; Simultaneously, the blade coverage rate should remain on 60~75%, the elongated shoot of the skeleton branch that growing way is more weak; Should suitably grow and stay, to increase its total increment, to all the other branches on the strong prosperous branch; Particularly the vertical branch of back can only stay 3~5cm to weigh cutting back.Be that at most it is abundant beta pruning state at least that excessive beta pruning state, pruning rate are crossed if pruning rate is crossed, pruning rate is moderate then to be suitable beta pruning state.
In the said step (1); Said cherry tree canopy branch sample is answered the cherry tree canopy dendritic structure of containing three kinds of different growth conditions as much as possible; Just can have good prediction effect to cherry tree canopy branch growth conditions to be measured, but sample size too much can be for the collection of sample, the correction of model bring bigger workload, the quantity of said cherry tree canopy branch sample is preferably 60~90; And the cherry tree canopy branch sample size of three kinds of growth conditions is consistent; With this understanding, the precision of forecast model not only can be guaranteed, and human cost can be saved.
In the said step (1), the image of collection is the near-infrared image of cherry tree canopy branch sample, with respect to the visible spectrum image, gathers near-infrared image and carries out Fast Fourier Transform (FFT) described in the step (2), and effect is more excellent.
In the said step (1);, accurately near-infrared image more clear in order to obtain; And then the more accurate near-infrared image textural characteristics information of extraction; The resolution of said near-infrared image is preferably 1920 * 1080, and the collection distance of near-infrared image is preferably 2~3m, and promptly the distance between the probe of images acquired and the cherry tree canopy branch sample is 2~3m.
In the said step (2), the shape of image and texture have certain occurrence frequency in image space, and therefore can carry out spectrum analysis to image extracts image texture features information; The present invention utilizes Fast Fourier Transform (FFT); To frequency domain, obtain the two-dimensional space frequency spectrum of near-infrared image, and said two-dimensional space frequency spectrum is divided into 10~25 annulus to spatial domain signal transformation; The frequency spectrum value of trying to achieve each annulus (can be referring to Fang Ruming; The flourish .1999 of Cai Jian, computer vision technique and the application in agricultural engineering thereof. the 45-47. of publishing house of Tsing-Hua University), the frequency spectrum value of each annulus can reflect image texture features information.
In the said step (2), for the integrality and the accuracy of the image texture characteristic information extraction of further guaranteeing the cherry tree beta pruning, save the operation time of model simultaneously, the number of annulus is preferably 20.
In the said step (3), the growth conditions of cherry tree canopy branch sample is confirmed through the principle of demarcating cherry tree canopy branch growth conditions.
In the said step (3),, be preferably based on the multiple regression algorithm and set up said calibration model in order to improve the accuracy of model prediction; Said multiple regression algorithm is partial least-squares regression method algorithm, principal component regression algorithm, progressively regression algorithm, artificial neural network algorithm or algorithm of support vector machine;
Artificial neural network algorithm more preferably; Owing to the relative complex that concerns between input value in this model and the output valve; Sample size is big, and artificial neural network has self-learning capability, can improve analysis precision through study; Big at sample size, concern that complicated model more has superiority aspect setting up, that the while artificial neural network algorithm has is anti-interference, noise resisting ability and the strong advantage of non-linear conversion ability; Most preferably be three layers of BP neural network algorithm.
In the said step (3); Said multiple regression algorithm is an artificial neural network algorithm, and when said artificial neural network algorithm was three layers of BP neural network algorithm, the hidden layer nodal point number of said three layers of BP neural network was preferably 9~15; The choice relation of hidden layer nodal point number is to the artificial nerve network model prediction accuracy; If the neuron of hidden layer knot very little, the reliability of model will reduce, and the network node number is too much; Possibly make network training innumerable, also can cause network can not discern the new samples pattern.
In the said step (4), when cherry tree near-infrared image to be measured is gathered, must be consistent with obtaining the method that cherry tree beta pruning sample near-infrared image adopted, promptly the resolution of IMAQ, collection distance should be consistent.
The present invention adopts multi-optical spectrum imaging technology to obtain cherry tree canopy branch chart picture; And utilize Fast Fourier Transform (FFT) that image is handled; The image texture information that obtains does not receive the influence of blade shape etc., can realize cherry tree canopy branch growth conditions is carried out real-time, quick, accurate, nondestructive detection, for the automation mechanized operation that realizes cherry tree and other fruit beta prunings precondition is provided simultaneously; Can reduce labor cost, have extremely strong applicability, economy and using value.
Description of drawings
The two-dimensional space frequency spectrogram that Fig. 1 obtains after Fast Fourier Transform (FFT) for near-infrared image in the embodiment of the invention;
Fig. 2 is the synoptic diagram of the total area in near-infrared image of 20 homalographic annulus in the embodiment of the invention.
Embodiment
The system that the present invention is used to detect cherry tree canopy branch growth conditions comprises visible and near infrared multispectral imaging appearance and computing machine; It is thus clear that and transmit data through image collection card between near infrared multispectral imaging appearance and the computing machine; Image pick-up card is connected on visible and the near infrared multispectral imaging appearance; It is thus clear that and the near infrared multispectral imaging appearance is connected with computing machine through RS-232 Serial Port Line and image acquisition data line card; Said computing machine is provided with image processing software, and wherein, visible and near infrared multispectral imaging appearance is the MS3100Duncan Camera of U.S. Redlake company; The bottom be provided with adjustable-angle, highly, the tripod of movable base; Camera lens and ground level images acquired information, image collection card are the PCI1424 or 1428 data collecting cards of American National Instrument Instrument company, and the visible and used light source of near infrared multispectral imaging appearance collection image is a natural light.
Choose 100 cherry trees from the field, wherein the canopy branch of 75 cherry trees is used to proofread and correct forecast model as cherry tree canopy branch sample, and the canopy branch of all the other 25 cherry trees is as cherry tree canopy branch to be measured; Utilize the visible near-infrared image that the near infrared multispectral imaging appearance obtains the canopy branch of said 100 cherry trees that reaches; After near-infrared image being carried out image pre-service such as background removal; It is carried out Fast Fourier Transform (FFT); Obtain the two-dimensional space frequency spectrum (referring to Fig. 1) of near-infrared image, through digitized processing picture frequency is composed the annulus (total area of 20 annulus is referring to Fig. 2) that is divided into 20 homalographics equably then, try to achieve the frequency spectrum value of each annulus; Obtain the corresponding actual measurement growth conditions of 75 cherry tree canopy branch sample near-infrared images according to the demarcation principle of canopy branch growth conditions simultaneously.
As input, is output with the actual measurement growth conditions of the corresponding cherry tree canopy branch sample of near-infrared image with the frequency spectrum value of 20 annulus of 75 cherry tree canopy branch sample near-infrared images, sets up the BP neural network computing model that contains input layer, hidden layer and output layer three-decker; Wherein, the input layer nodal point number is 20 (being respectively the frequency spectrum value of 20 annulus), and the output layer nodal point number is 3; The hidden layer nodal point number is 12; The minimum training speed of setting network is 0.1, and dynamic parameter is 0.9, and the data-switching mode is the standardization conversion; Maximum iteration time is 1000 times; Obtain the mapping relations like table 1, present embodiment is chosen typical 15 samples and is listed in table 1 as space is limited.
Table 1 is used for the partial database of modelling
NO | X 1 | X 2 | X 3 | X 4 | X 5 | X 6 | X 7 | X 8 | … | Y 1 | Y 2 | Y 3 |
1 | 2.20 | 0.75 | 0.53 | 0.42 | 0.34 | 0.30 | 0.26 | 0.24 | … | 1 | 0 | 0 |
2 | 2.20 | 0.79 | 0.55 | 0.43 | 0.37 | 0.32 | 0.28 | 0.25 | … | 1 | 0 | 0 |
3 | 1.69 | 0.57 | 0.38 | 0.31 | 0.25 | 0.22 | 0.19 | 0.18 | … | 1 | 0 | 0 |
4 | 2.16 | 0.73 | 0.50 | 0.39 | 0.32 | 0.27 | 0.24 | 0.22 | … | 1 | 0 | 0 |
5 | 2.05 | 0.71 | 0.49 | 0.38 | 0.31 | 0.27 | 0.24 | 0.21 | … | 1 | 0 | 0 |
6 | 1.78 | 0.59 | 0.40 | 0.31 | 0.26 | 0.22 | 0.20 | 0.18 | … | 1 | 0 | 0 |
7 | 2.15 | 0.72 | 0.48 | 0.38 | 0.31 | 0.26 | 0.23 | 0.21 | … | 1 | 0 | 0 |
8 | 1.91 | 0.62 | 0.42 | 0.33 | 0.27 | 0.23 | 0.20 | 0.18 | … | 0 | 1 | 0 |
9 | 2.42 | 0.81 | 0.54 | 0.42 | 0.35 | 0.29 | 0.26 | 0.23 | … | 1 | 0 | 0 |
10 | 1.62 | 0.52 | 0.35 | 0.27 | 0.23 | 0.20 | 0.17 | 0.15 | … | 0 | 1 | 0 |
11 | 1.08 | 0.32 | 0.22 | 0.17 | 0.14 | 0.12 | 0.10 | 0.09 | … | 0 | 0 | 1 |
12 | 2.59 | 0.80 | 0.55 | 0.43 | 0.36 | 0.30 | 0.27 | 0.24 | … | 1 | 0 | 0 |
13 | 2.29 | 0.70 | 0.49 | 0.38 | 0.31 | 0.26 | 0.23 | 0.21 | … | 1 | 0 | 0 |
14 | 1.61 | 0.48 | 0.32 | 0.25 | 0.21 | 0.18 | 0.16 | 0.14 | … | 0 | 1 | 0 |
15 | 2.54 | 0.84 | 0.56 | 0.43 | 0.36 | 0.30 | 0.26 | 0.23 | … | 1 | 0 | 0 |
Annotate: in the output row, the insufficient beta pruning state of 100 representatives; The suitable beta pruning state of 010 representative; The excessive beta pruning state of 001 representative.
After accomplishing the domestication of model, good neural network model is set up in the frequency spectrum value input of 20 annulus of 25 cherry tree canopy branch near-infrared images to be measured, drawn output data Y
1, Y
2And Y
3, as shown in table 2, through the output data of model, confirm the model prediction growth conditions of cherry tree canopy branch to be measured according to following principle: if Y
1Predict the outcome between 0.5~1.5 Y
2And Y
3Data then are judged to be not fully beta pruning state between-0.5~0.5; If Y
2Predict the outcome between 0.5~1.5 Y
1And Y
3Data then are judged to be suitable beta pruning state between-0.5~0.5; If Y
3Predict the outcome between 0.5~1.5 Y
1And Y
2Data then are judged to be excessive beta pruning state between-0.5~0.5; Analyze to know, except the 11st group, the 12nd group, the 14th group and the 20th group model prediction growth conditions mistake, all the other each group models prediction growth conditions all with survey growth conditions and conform to model prediction rate of accuracy reached to 84%.
Table 2 cherry tree canopy to be measured branch prediction growth conditions and actual measurement growth conditions
Annotate: 0 representative is unpredictable; The insufficient beta pruning of 1 representative; The suitable beta pruning of 2 representatives; The excessive beta pruning of 3 representatives.
Claims (8)
1. the detection method of a cherry tree canopy branch growth conditions is characterized in that, may further comprise the steps:
(1) gathers the near-infrared image of the cherry tree canopy branch sample of several different growth conditions respectively;
(2) near-infrared image is carried out Fast Fourier Transform (FFT), obtain the two-dimensional space frequency spectrum of near-infrared image, this two-dimensional space frequency spectrum is divided into 10~25 homalographic annulus equably, and try to achieve the frequency spectrum value of each annulus;
(3) travel through all cherry tree canopy branch samples, obtain respectively and the corresponding frequency spectrum value of each cherry tree canopy branch sample;
With said frequency spectrum value is input, is output with the growth conditions of the cherry tree canopy branch sample of correspondence, sets up model;
(4) obtain and the corresponding frequency spectrum value of cherry tree canopy branch to be measured according to step (1) and (2), the said model of substitution obtains the growth conditions of cherry tree canopy branch to be measured.
2. the detection method of cherry tree canopy branch growth conditions as claimed in claim 1 is characterized in that, in the said step (2), said annulus is 20.
3. the detection method of cherry tree canopy branch growth conditions as claimed in claim 1 is characterized in that, in the said step (3), sets up said calibration model based on the multiple regression algorithm; Said multiple regression algorithm is partial least-squares regression method algorithm, principal component regression algorithm, progressively regression algorithm, artificial neural network algorithm or algorithm of support vector machine.
4. the detection method of cherry tree canopy branch growth conditions as claimed in claim 3 is characterized in that said multiple regression algorithm is an artificial neural network algorithm.
5. the detection method of cherry tree canopy branch growth conditions as claimed in claim 4 is characterized in that, said artificial neural network algorithm is three layers of BP neural network algorithm.
6. the detection method of cherry tree canopy branch growth conditions as claimed in claim 5 is characterized in that, said artificial neural network algorithm is three layers of BP neural network algorithm, and the hidden layer nodal point number of said three layers of BP neural network is 9~15.
7. the detection method of cherry tree canopy branch growth conditions as claimed in claim 1 is characterized in that the quantity of said cherry tree beta pruning sample is 60~90.
8. the detection method of cherry tree canopy branch growth conditions as claimed in claim 1 is characterized in that the resolution of said near-infrared image is 1920 * 1080.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210224548.5A CN102749290B (en) | 2012-07-02 | 2012-07-02 | Method for detecting growth state of branches of crown canopy of cherry tree |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210224548.5A CN102749290B (en) | 2012-07-02 | 2012-07-02 | Method for detecting growth state of branches of crown canopy of cherry tree |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102749290A true CN102749290A (en) | 2012-10-24 |
CN102749290B CN102749290B (en) | 2014-12-03 |
Family
ID=47029655
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210224548.5A Active CN102749290B (en) | 2012-07-02 | 2012-07-02 | Method for detecting growth state of branches of crown canopy of cherry tree |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102749290B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104914053A (en) * | 2015-05-15 | 2015-09-16 | 中国农业大学 | Tree canopy photosynthetic character prediction model construction and photosynthetic character detection |
CN109961035A (en) * | 2019-03-19 | 2019-07-02 | 广东省农业科学院水稻研究所 | A kind of rice plant of tillering stage growing way wisdom monitoring adjusting method |
WO2019179270A1 (en) * | 2018-03-23 | 2019-09-26 | 广州极飞科技有限公司 | Plant planting data measuring method, working route planning method, device and system |
CN111108368A (en) * | 2017-09-19 | 2020-05-05 | 富士胶片株式会社 | Teaching data production method and device and defect inspection method and device |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000065750A (en) * | 1998-06-10 | 2000-03-03 | Kansai Paint Co Ltd | Method for quantitatively evaluating brilliant feeling of coating film |
US20020165684A1 (en) * | 2001-05-04 | 2002-11-07 | Electronics For Imaging, Inc. | Methods and apparatus for correcting spectral color measurements |
KR20050071350A (en) * | 2003-12-30 | 2005-07-07 | 롬 앤드 하스 캄파니 | Method for detecting contaminants |
US7006680B2 (en) * | 2002-05-03 | 2006-02-28 | Vialogy Corp. | System and method for characterizing microarray output data |
CN1975378A (en) * | 2006-11-24 | 2007-06-06 | 浙江大学 | Gray mold early diagnosing system based on visable/near infrared multispectral imaging |
CN101059427A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method for quickly non-destructive measurement for nitrogen content of tea using multiple spectrum imaging technology |
CN101059425A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method and device for identifying different variety green tea based on multiple spectrum image texture analysis |
CN101135652A (en) * | 2007-10-15 | 2008-03-05 | 清华大学 | Weld joint recognition method based on texture partition |
CN101210875A (en) * | 2007-12-25 | 2008-07-02 | 浙江大学 | Damage-free measurement method for soil nutrient content based on near infrared spectra technology |
CN101210876A (en) * | 2007-12-25 | 2008-07-02 | 浙江大学 | Rice nutrient information measuring method based on visible/near infrared multispectral imaging |
CN101520412A (en) * | 2009-03-23 | 2009-09-02 | 中国计量学院 | Near infrared spectrum analyzing method based on isolated component analysis and genetic neural network |
US20090258684A1 (en) * | 2004-07-28 | 2009-10-15 | Cnh America Llc | Apparatus and method for analysing the composition of crop in a crop-conveying machine |
CN101881726A (en) * | 2010-06-18 | 2010-11-10 | 北京农业智能装备技术研究中心 | Nondestructive detection method for comprehensive character living bodies of plant seedlings |
CN101957313A (en) * | 2010-09-21 | 2011-01-26 | 吉林大学 | Method and device for computer visual inspection classification of quality of fresh corn ears |
-
2012
- 2012-07-02 CN CN201210224548.5A patent/CN102749290B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000065750A (en) * | 1998-06-10 | 2000-03-03 | Kansai Paint Co Ltd | Method for quantitatively evaluating brilliant feeling of coating film |
US20020165684A1 (en) * | 2001-05-04 | 2002-11-07 | Electronics For Imaging, Inc. | Methods and apparatus for correcting spectral color measurements |
US7006680B2 (en) * | 2002-05-03 | 2006-02-28 | Vialogy Corp. | System and method for characterizing microarray output data |
KR20050071350A (en) * | 2003-12-30 | 2005-07-07 | 롬 앤드 하스 캄파니 | Method for detecting contaminants |
US20090258684A1 (en) * | 2004-07-28 | 2009-10-15 | Cnh America Llc | Apparatus and method for analysing the composition of crop in a crop-conveying machine |
CN1975378A (en) * | 2006-11-24 | 2007-06-06 | 浙江大学 | Gray mold early diagnosing system based on visable/near infrared multispectral imaging |
CN101059425A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method and device for identifying different variety green tea based on multiple spectrum image texture analysis |
CN101059427A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method for quickly non-destructive measurement for nitrogen content of tea using multiple spectrum imaging technology |
CN101135652A (en) * | 2007-10-15 | 2008-03-05 | 清华大学 | Weld joint recognition method based on texture partition |
CN101210875A (en) * | 2007-12-25 | 2008-07-02 | 浙江大学 | Damage-free measurement method for soil nutrient content based on near infrared spectra technology |
CN101210876A (en) * | 2007-12-25 | 2008-07-02 | 浙江大学 | Rice nutrient information measuring method based on visible/near infrared multispectral imaging |
CN101520412A (en) * | 2009-03-23 | 2009-09-02 | 中国计量学院 | Near infrared spectrum analyzing method based on isolated component analysis and genetic neural network |
CN101881726A (en) * | 2010-06-18 | 2010-11-10 | 北京农业智能装备技术研究中心 | Nondestructive detection method for comprehensive character living bodies of plant seedlings |
CN101957313A (en) * | 2010-09-21 | 2011-01-26 | 吉林大学 | Method and device for computer visual inspection classification of quality of fresh corn ears |
Non-Patent Citations (2)
Title |
---|
徐贵力,毛罕平: "利用傅里叶变换提取图像纹理特征新方法", 《光电工程》, vol. 31, no. 11, 30 November 2004 (2004-11-30) * |
邵咏妮,曹芳, 何勇,: "基于独立组分分析和BP神经网络的可见/近红外光谱稻谷年份的鉴别", 《红外与毫米波学报》, vol. 26, no. 6, 31 December 2007 (2007-12-31) * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104914053A (en) * | 2015-05-15 | 2015-09-16 | 中国农业大学 | Tree canopy photosynthetic character prediction model construction and photosynthetic character detection |
CN104914053B (en) * | 2015-05-15 | 2017-07-28 | 中国农业大学 | Build trees canopy photosynthesis trait predictive model and detection Photosynthetic Characters |
CN111108368A (en) * | 2017-09-19 | 2020-05-05 | 富士胶片株式会社 | Teaching data production method and device and defect inspection method and device |
CN111108368B (en) * | 2017-09-19 | 2022-06-28 | 富士胶片株式会社 | Teaching data production method and device and defect inspection method and device |
US11386542B2 (en) | 2017-09-19 | 2022-07-12 | Fujifilm Corporation | Training data creation method and device, and defect inspection method and device |
WO2019179270A1 (en) * | 2018-03-23 | 2019-09-26 | 广州极飞科技有限公司 | Plant planting data measuring method, working route planning method, device and system |
US11321942B2 (en) | 2018-03-23 | 2022-05-03 | Guangzhou Xaircraft Technology Co., Ltd. | Method for measuring plant planting data, device and system |
CN109961035A (en) * | 2019-03-19 | 2019-07-02 | 广东省农业科学院水稻研究所 | A kind of rice plant of tillering stage growing way wisdom monitoring adjusting method |
Also Published As
Publication number | Publication date |
---|---|
CN102749290B (en) | 2014-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108921885B (en) | Method for jointly inverting forest aboveground biomass by integrating three types of data sources | |
Tilly et al. | Multitemporal crop surface models: accurate plant height measurement and biomass estimation with terrestrial laser scanning in paddy rice | |
Liu et al. | A vision-based robust grape berry counting algorithm for fast calibration-free bunch weight estimation in the field | |
CN103065149A (en) | Netted melon fruit phenotype extraction and quantization method | |
Estornell et al. | Tree extraction and estimation of walnut structure parameters using airborne LiDAR data | |
CN107767364B (en) | Method for accurately extracting temperature of tree canopy based on infrared thermal image | |
Bi et al. | Estimating vertical chlorophyll concentrations in maize in different health states using hyperspectral LiDAR | |
CN110189043B (en) | Usable land resource analysis system based on high-score satellite remote sensing data | |
CA2981473A1 (en) | Forecasting national crop yield during the growing season | |
CN113252583B (en) | Method for calculating alpine hay coverage based on hay vegetation index | |
CN102749290B (en) | Method for detecting growth state of branches of crown canopy of cherry tree | |
CN113269825B (en) | Forest breast diameter value extraction method based on foundation laser radar technology | |
dos Santos et al. | Use of allometric models to estimate leaf area in Hymenaea courbaril L. | |
CN115170341A (en) | Forestry resource carbon sink accurate measurement and calculation method for correcting small samples by utilizing laser radar | |
CN103940748A (en) | Citrus canopy nitrogen content prediction and visualization method based on hyperspectrum technology | |
CN102592118B (en) | Automatic detection method for time emergence of seedling of corns | |
CN115700370A (en) | Carbon reserve calculation method, carbon reserve calculation device, electronic device, and storage medium | |
CN109752487A (en) | Wheat Leavess nitrogen content predictor method and device | |
CN104236486B (en) | A kind of cotton leaf area index quick nondestructive assay method | |
Carrasco-Benavides et al. | Assessment of an automated digital method to estimate leaf area index (LAI) in cherry trees | |
Wei et al. | Dynamic simulation of leaf area index for the soybean canopy based on 3D reconstruction | |
CN113343808A (en) | Tropical forest resource measuring method based on satellite remote sensing technology | |
Sukojo et al. | Landsat 8 satellite imagery analysis for rice production estimates (Case study: Bojonegoro regencys) | |
Meng et al. | Predicting optimal soybean harvesting dates with satellite data | |
Li et al. | Use of terrestrial laser scanning to obtain the stem diameters of Larix olgensis and construct compatible taper-volume equations |
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 |