CN102612892B - Identification method for sprouting conditions of wheat ears - Google Patents
Identification method for sprouting conditions of wheat ears Download PDFInfo
- Publication number
- CN102612892B CN102612892B CN201210053700.8A CN201210053700A CN102612892B CN 102612892 B CN102612892 B CN 102612892B CN 201210053700 A CN201210053700 A CN 201210053700A CN 102612892 B CN102612892 B CN 102612892B
- Authority
- CN
- China
- Prior art keywords
- wheat
- germination
- wheat head
- 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
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses an identification method for sprouting conditions of wheat ears, which comprises the following steps: S1 collecting hyperspectral images of the wheat ears; S2 performing pretreatment to the hyperspectral images of the wheat ears; S3 synthesizing the hyperspectral images of the wheat ears after the pretreatment to obtain red green blue (RGB) images; S4 analyzing areas of the RGB images which operators are interested in, and obtaining average spectra of the areas which the operators are interested in; S5 judging whether the wheat ears sprout under a characteristic wave band; S6 calculating the proportion of sprouting areas in the whole wheat ear area; S7 extracting spectral reflectivity of the sprouting areas, and judging the sprouting degree of the wheat ears; and S8 calculating the sprouting grade according to the proportion of the sprouting areas in the whole wheat ear area and the spectral reflectivity. The identification method adopts the hyperspectral image technology to perform wheat ear sprouting detection on the wheat ears normally harvested in the fields, and rapid screening of wheat ear sprouting can be achieved.
Description
Technical field
The present invention relates to crop breeding and image recognition technology field, relate in particular to a grow wheat wheat head germination recognition methods.
Background technology
When wheat ear germinating refers to and runs into rainy weather before results, seed is in the phenomenon of head sprouting, and fringe germinates and not only affects output, and has a strong impact on quality (especially processing quality) and plant by value.There is the general underproduction of wheatland 10% left and right that visible fringe germinates, when serious, can have no harvest.Wheat ear germinating not only makes the remarkable underproduction of wheat, and its processing, nutritional quality and kind are all affected with being worth, and causes serious economic loss, and therefore, the identification of wheat ear germinating is identified significant for wheat breeding.At some Mai Qu, before harvesting wheat, often run into continuous rainy weather, cause fringe to germinate and occur.In recent years, along with the adjustment of agricultural planting structure, the cultivated area of white skin wheat is expanding gradually, but due to the ear germinating resistance of white skin wheat generally a little less than, so wheat ear germinating problem is day by day serious.
The authentication method of wheat ear germinating and index have multiple, and existing test differentiates that the technology of wheat ear germinating has: artificial visually examine's method, mainly relies on the state of people's micro-judgment wheat ear germinating; And the method such as alpha-amylase activity mensuration.
Wherein, artificial visually examine identifies wheat ear germinating method, is subject in early days Subjective Factors larger, and recognition result standardization is poor; Adopt that biochemical method is destructive to be detected, but these methods too complexity or identification of indicator reliability poor, alpha-amylase activity is measured process more complicated, efficiency is low; Culture dish seed sprouting needs larger lab space, and will carry out water management and Continuous Observation monitoring, lacks technological means harmless, that can monitor continuously.The needed authentication method of wheat breeding man must have the feature of simple and effective, emphasizes its applicability.
Summary of the invention
(1) technical problem that will solve
The technical problem to be solved in the present invention is: the recognition methods of a kind of wheat head of wheat fast and effectively germination is provided.
(2) technical scheme
For addressing the above problem, the invention provides a grow wheat wheat head germination recognition methods, comprise the following steps:
S1: gather wheat head high spectrum image;
S2: described wheat head high spectrum image is carried out to pretreatment;
S3: by the synthetic RGB image that obtains of pretreated described wheat head high spectrum image;
S4: select interested region to analyze in described RGB image, obtain the averaged spectrum of described area-of-interest;
S5: whether germinate at the characteristic wave bands wheat head that judges;
S6: calculate the ratio that germination region accounts for whole wheat head area;
S7: extract the spectral reflectivity in germination region, the germination degree of the judgement wheat head;
S8: account for ratio and the spectral reflectivity of whole wheat head area according to described germination region, calculate grade of germination.
Preferably, the pretreatment of described step S2 comprises that the wheat head high spectrum image to gathering splices, and forms the step of the image of BSQ form.
Preferably, the step that the pretreatment of described step S2 comprises that wheat head high spectrum image to gathering is proofreaied and correct, filtering and enhancing are processed.
Preferably, step S3 specifically comprises: pretreated described wheat head high spectrum image is extracted to spectrum dimension characteristic of correspondence wavelength and be respectively the spectrum picture of 680nm, 550nm and 450nm and carry out Integral Transformation, synthesize and obtain described RGB image.
Preferably, in step S4, interested region is analyzed specifically and comprised: described area-of-interest is amplified, extract the spectrum of each pixel in described area-of-interest, then calculating mean value.
Preferably, the wave-length coverage of the high spectrum image in described step S1 is 400-1000nm.
Preferably, the characteristic wave bands described in step S5 is in 450~900nm spectral wavelength scope.
Preferably, the characteristic wave bands described in step S5 is 675nm spectral wavelength.
Preferably, step S6 specifically comprises: respectively described area-of-interest is carried out to Threshold segmentation, extract the pixel value in wheatear portion germination region and the pixel value of the whole wheat head, according to the number of pixel value, calculate the area that calculates respectively germination region and the whole wheat head, the gross area of obtaining again germination region accounts for the percentage of whole wheat head image area, thereby the area that obtains germinateing accounts for the percentage of the whole wheat head.
(3) beneficial effect
The high light spectrum image-forming of the present invention by combining spectral technique and image technique carries out total score after to the withdrawing spectral information of the morphological feature of wheat ear germinating and wheat ear germinating area-of-interest analyses, thereby the fringe germination that realizes the whole wheat head of wheat is identified fast.Relatively traditional non-imaging spectral analysis, the present invention can intuitively find out the position of germination; With respect to machine vision imaging technology, the present invention can, by the reflectivity judgement grade of germination of spectrum, can judge wheat ear germinating situation in early days more accurately.
Accompanying drawing explanation
Fig. 1 is the steps flow chart schematic diagram according to embodiment of the present invention recognition methods.
Fig. 2 is according to the germination of embodiment of the present invention recognition methods extraction and the average light spectrogram at the position of not germinateing.
Embodiment
Below in conjunction with drawings and Examples, that the present invention is described in detail is as follows.
As shown in Figure 1, the present embodiment has been recorded a grow wheat wheat head germination recognition methods.
In the present embodiment, laboratory sample is divided into four groups, is respectively: water every day and water three water, every days the dry wheat head that water, a whole day are soaked, do not watered for 24 hours, each 30 strains, gather its high spectrum image information, observation wheat ear germinating situation.Before doing high spectrum image experiment, the wheat head sample through watering is all dried to the effect consistent with original wheatear, prevent that in experimentation, wheat further germinates, and then gather its high spectrum image, reduce the impact of moisture difference on spectrum simultaneously.
The recognition methods of the present embodiment comprises the following steps:
S1: gather wheat head high spectrum image;
In the present embodiment, adopt the push-broom type hyperspectral imager of 400-1000nm wavelength band, the wheat wheat head is scanned.The wheat head is kept flat, gather wheatear portion positive and negative two sides collection of illustrative plates respectively once, guarantee the omnibearing observation to the wheat head.
Wherein, the hyperspectral imager that the present embodiment adopts is the PIS112 bis-generations hyperspectral imager of Chinese University of Science and Technology's development, spectrometer has adopted the CCD of 1400 (space dimension) * 1024 (spectrum dimension) to carry out array push-scanning image, spectral range: 400~1000nm; Spectral resolution: 2nm, sampling interval 0.7nm; Sample frequency: 8-30 width/second; The angle of visual field: 16 °.
S2: described wheat head high spectrum image is carried out to pretreatment;
The step that in the present embodiment, described pretreatment comprises that the wheat head high spectrum image to gathering is proofreaied and correct, filtering and enhancing are processed, the noise causing to eliminate the reasons such as the inhomogeneous and quantum efficiency imbalance of non-linear sensitivity, spectral response due to CCD.Described correction is processed and is comprised blank and dark current correction.
The pretreatment of described step S2 comprises that the wheat head high spectrum image to gathering splices, and forms the step of the image of BSQ form (wave band order format).Because the initial data of imaging spectrometer collection is the picture of BMP form, first need to be spliced into the image of BSQ form.Processing procedure concrete in the present embodiment is: first by Matlab software programming program, the picture of BMP form is spliced into the view picture image of BIL form; Then with IDL programming, carry out the extraction of image reflectivity, comprise that the reflectivity based on experience linear approach extracts, five steps are the smoothing processing of the method for average progressively, finally saves as the image of BSQ form.
S3: by the synthetic RGB image that obtains of pretreated described wheat head high spectrum image;
High spectrum image is integrated traditional images dimension and spectrum dimension information fusion, when obtaining wheatear portion spatial image, obtain the continuous spectrum information of each fringe portion pixel, be that imaging spectrum is an image cube, it is comprised of three parts: spatial image dimension, spectrum dimension, characteristic spectrum dimension.The present embodiment step S3 extracts spectrum dimension characteristic of correspondence wavelength to pretreated described wheat head high spectrum image and is respectively the spectrum picture of 680nm, 550nm and 450nm (three wavelength are corresponding with R component, G component and the B component of color of image respectively) and carries out Integral Transformation, synthesizes and obtains described RGB image.
S4: select interested region to analyze in described RGB image, obtain the averaged spectrum of described area-of-interest;
In the present embodiment, step 4 is specially: according to synthetic RGB image, select interested region to amplify analysis, extract the spectrum of each pixel in described area-of-interest, then calculating mean value, obtain the averaged spectrum of area-of-interest, for follow-up, the spectral informations such as spectral absorption characteristics parameter (absorbing wavelength position, the absorption degree of depth, absorption width) are excavated and laid the foundation.
S5: whether germinate at the characteristic wave bands wheat head that judges;
Described characteristic wave bands is in 450~900nm spectral wavelength scope; Preferably, described characteristic wave bands is 675nm spectral wavelength.
For the wheat head different parts averaged spectrum of extracting, draw following features: in 450~900nm wave-length coverage, the spectral reflectivity of germinated wheat fringe is reflection-absorption paddy at 675nm place, and 714nm place is reflection-absorption peak, and this is substantially similar to typical vegetation curve of spectrum variation characteristic; And germinated wheat fringe does not occur without spectral absorption paddy at 675nm place, so the present embodiment is using the characteristic absorption of 675nm as judgement wheat ear germinating whether foundation.As shown in Figure 2, the present embodiment can tentatively judge by extracting the single image of wheatear portion spectrum picture at 675nm place whether identification wheatear portion germinates.
S6: calculate the ratio that germination region accounts for whole wheat head area;
Step S6 specifically comprises: respectively described area-of-interest is carried out to Threshold segmentation, the algorithm of use mainly contains: inverse, maximum variance between clusters binaryzation, medium filtering; Extract the pixel value in wheatear portion germination region and the pixel value of the whole wheat head, according to the number of pixel value, calculate the area that calculates respectively germination region and the whole wheat head, the gross area of obtaining again germination region accounts for the percentage of whole wheat head image area, thereby the area that obtains germinateing accounts for the percentage of the whole wheat head.
S7: extract the spectral reflectivity in germination region, the germination degree of the judgement wheat head;
Owing to germinateing, part contains chlorophyll composition information, the curve map trend that its spectral curve presents is substantially similar to green crop chlorophyll spectral reflectivity curve map, there is following characteristics: the chlorophyll characteristic wave bands of wheat ear germinating part is in 675nm position, there is obvious absorption trough, according to the light and heavy degree of the large I judgement identification wheat ear germinating of this wave band place spectral reflectivity.Germinate serious, chlorophyll content is high, and the corresponding spectral reflectivity therefore extracting can be higher.
S8: account for ratio and the spectral reflectivity of whole wheat head area according to described germination region, the light and heavy degree grade of judgement identification wheat ear germinating, with this comprehensive situation of passing judgment on the germination of the wheat wheat head.The grade that fringe germinates comprises without germination, slight germination, severe germination Three Estate.
The present invention, by gathering unknown wheat ear germinating sample high spectrum image, extracts spectral information, by image, is processed and can be carried out the monitoring of wheat ear germinating.The present invention can adopt high light spectrum image-forming technology to carry out fringe germination and detect for the wheat head of field normal harvest, can realize the rapid screening that fringe germinates.
Above embodiment is only for illustrating the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; without departing from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (6)
1. a grow wheat wheat head germination recognition methods, is characterized in that, comprises the following steps:
S1: gather wheat head high spectrum image;
S2: described wheat head high spectrum image is carried out to pretreatment, and wherein, described pretreatment comprises: the wheat head high spectrum image gathering is spliced, form the step of the image of BSQ form; Or, to the step that the wheat head high spectrum image gathering is proofreaied and correct, filtering and enhancing are processed;
S3: by the synthetic RGB image that obtains of pretreated described wheat head high spectrum image;
S4: select interested region to analyze in described RGB image, obtain the averaged spectrum of described area-of-interest;
S5: whether germinate at the characteristic wave bands wheat head that judges;
S6: calculate the ratio that germination region accounts for whole wheat head area, specifically comprise:
Respectively described area-of-interest is carried out to Threshold segmentation, extract the pixel value in wheatear portion germination region and the pixel value of the whole wheat head, according to the number of pixel value, calculate respectively the area of germination region and the whole wheat head, the gross area of obtaining again germination region accounts for the percentage of whole wheat head image area, thereby the region area that obtains germinateing accounts for the percentage of whole wheat head area;
S7: extract the spectral reflectivity in germination region, the germination degree of the judgement wheat head;
S8: account for ratio and the spectral reflectivity of whole wheat head area according to described germination region, calculate grade of germination.
2. wheat wheat head germination as claimed in claim 1 recognition methods, it is characterized in that, step S3 specifically comprises: pretreated described wheat head high spectrum image is extracted to spectrum dimension characteristic of correspondence wavelength and be respectively the spectrum picture of 680nm, 550nm and 450nm and carry out Integral Transformation, synthesize and obtain described RGB image.
3. wheat wheat head germination as claimed in claim 1 recognition methods, it is characterized in that, in step S4, interested region is analyzed specifically and comprised: described area-of-interest is amplified, extract the spectrum of each pixel in described area-of-interest, then calculating mean value.
4. wheat wheat head germination as claimed in claim 1 recognition methods, is characterized in that, the wave-length coverage of the high spectrum image in described step S1 is 400-1000nm.
5. wheat wheat head germination as claimed in claim 1 recognition methods, is characterized in that, the characteristic wave bands described in step S5 is in 450~900nm spectral wavelength scope.
6. wheat wheat head germination as claimed in claim 5 recognition methods, is characterized in that, the characteristic wave bands described in step S5 is 675nm spectral wavelength.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210053700.8A CN102612892B (en) | 2012-03-02 | 2012-03-02 | Identification method for sprouting conditions of wheat ears |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210053700.8A CN102612892B (en) | 2012-03-02 | 2012-03-02 | Identification method for sprouting conditions of wheat ears |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102612892A CN102612892A (en) | 2012-08-01 |
CN102612892B true CN102612892B (en) | 2014-05-07 |
Family
ID=46553376
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210053700.8A Active CN102612892B (en) | 2012-03-02 | 2012-03-02 | Identification method for sprouting conditions of wheat ears |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102612892B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102855485B (en) * | 2012-08-07 | 2015-10-28 | 华中科技大学 | The automatic testing method of one grow wheat heading |
CN102948282B (en) * | 2012-10-31 | 2013-12-25 | 北京农业信息技术研究中心 | Wheatear germination degree detection method |
CN104849287B (en) * | 2015-06-10 | 2018-05-01 | 国家电网公司 | A kind of composite insulator pollution level non-contact detection method |
CN105300895B (en) * | 2015-11-05 | 2017-12-26 | 浙江大学 | A kind of method using characteristic point tangent line angle early warning potato sprouting defect |
CN105424622B (en) * | 2015-11-05 | 2018-01-30 | 浙江大学 | A kind of method using feature triangle area early warning potato sprouting |
CN105424623B (en) * | 2015-11-05 | 2017-12-12 | 浙江大学 | A kind of method using feature triangle height early warning potato sprouting |
US11010913B2 (en) * | 2016-10-19 | 2021-05-18 | Basf Agro Trademarks Gmbh | Determining the grain weight of an ear |
CN109164014B (en) * | 2018-06-28 | 2020-11-06 | 浙江理工大学 | Hyperspectral image processing-based method for identifying wetting area of multicolor fabric |
WO2020082264A1 (en) * | 2018-10-24 | 2020-04-30 | 合刃科技(深圳)有限公司 | Coating region positioning method and apparatus based on hyperspectral optical sensor, and adhesive removal system |
CN110132862B (en) * | 2019-05-30 | 2021-07-16 | 安徽大学 | Construction method and application of disease index special for wheat scab detection |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1900695A (en) * | 2005-07-21 | 2007-01-24 | 李少昆 | Field quick monitoring method for wheat nitrogen content and seed protein quality based on high light spectrum |
-
2012
- 2012-03-02 CN CN201210053700.8A patent/CN102612892B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN102612892A (en) | 2012-08-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102612892B (en) | Identification method for sprouting conditions of wheat ears | |
US10127451B1 (en) | Method of detecting and quantifying sun-drying crops using satellite derived spectral signals | |
CN109948596B (en) | Method for identifying rice and extracting planting area based on vegetation index model | |
CN110163138B (en) | Method for measuring and calculating wheat tillering density based on multispectral remote sensing image of unmanned aerial vehicle | |
CN102948282B (en) | Wheatear germination degree detection method | |
CN103278503B (en) | Multi-sensor technology-based grape water stress diagnosis method and system therefor | |
CN111612777B (en) | Soybean mapping method based on leaf aging and water loss index | |
Graeff et al. | Evaluation of Image Analysis to Determine the N-Fertilizer Demand of Broccoli Plants (Brassica oleracea convar. botrytis var. italica). | |
CN105300895B (en) | A kind of method using characteristic point tangent line angle early warning potato sprouting defect | |
CN115953690B (en) | Lodging crop identification method for unmanned harvester travel calibration | |
CN108037123A (en) | A kind of hybrid paddy rice disc type sows performance parameter accurate detecting method | |
CN114049564A (en) | Pine wood nematode disease grade prediction model construction method based on hyperspectral remote sensing image | |
Payne et al. | Machine vision in estimation of fruit crop yield | |
Tavakoli et al. | Evaluation of different sensing approaches concerning to nondestructive estimation of leaf area index (LAI) for winter wheat | |
CN105426585B (en) | A kind of potato sprouting method for early warning based on SIN function fitting process | |
CN106683069A (en) | Method for recognizing inline crops and weeds in seedling stage of farmland | |
CN105424622B (en) | A kind of method using feature triangle area early warning potato sprouting | |
CN101911877A (en) | Seed vitality authentication device and method based on laser light diffuse reflection image technology | |
CN107169940B (en) | Single pear tree yield obtaining method based on electronic identification | |
Di et al. | The research on the feature extraction of sunflower leaf rust characteristics based on color and texture feature | |
CN114782843A (en) | Crop yield prediction method and system based on unmanned aerial vehicle multispectral image fusion | |
KR20220094512A (en) | A method for monitoring crop growth by analyzing spectroscopic images through remote sensing | |
CN108982386B (en) | Multispectral image determination method and system for total iron content of sandalwood leaves | |
Shi et al. | Method for crop classification based on multi-source remote sensing data | |
CN106872368A (en) | A kind of tea tree tender leaf recognition methods based on EC concentration differences |
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 |