CN108734205A - A kind of simple grain for different cultivars wheat seed pinpoints identification technology - Google Patents
A kind of simple grain for different cultivars wheat seed pinpoints identification technology Download PDFInfo
- Publication number
- CN108734205A CN108734205A CN201810435100.5A CN201810435100A CN108734205A CN 108734205 A CN108734205 A CN 108734205A CN 201810435100 A CN201810435100 A CN 201810435100A CN 108734205 A CN108734205 A CN 108734205A
- Authority
- CN
- China
- Prior art keywords
- data
- seed
- wheat seed
- wheat
- sample
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of simple grains for different cultivars wheat seed to pinpoint identification technology, includes the following steps:The hyperspectral image data for obtaining wheat seed to be identified after being corrected to high spectrum image using black and white bearing calibration, carries out principal component transform;It selects Intact wheat seed as area-of-interest, extracts spectroscopic data of the averaged spectrum of relevant range as this sample;Spectroscopic data is smoothed using S-G exponential smoothings;Sample spectrum data are divided into training set and forecast set according to 3: 2 ratio using Kennard-Stone methods, are suitable for the optimal wave band that different Hybrid wheat seeds differentiate using successive projection algorithm picks;Build supporting vector machine model;In the SVM models that the input of optimal wave band is established, prediction result is indicated with different colours, completes the fixed point identification of seed mixture.The present invention realizes the on-line checking of the wheat seed of different Hybrids, and realizes the fixed point identification of simple grain.
Description
Technical field
The present invention relates to seeds to identify field, and in particular to a kind of simple grain for different cultivars wheat seed pinpoints identification
Technology.
Background technology
The purity of seed is to weigh the important indicator of seed quality.The purity of so-called wheat is exactly to test in seed batch to wrap
The ratio of the expection seed contained.With the extensive use of hybridization technique, wheat seed kind is on the increase, interracial similitude
It gradually increases, subjective perception is difficult to differentiate between.It is storing in transportational process, different wheat seeds may possibly be mixed in together, drop
The purity of low seed influences the character of filial generation.And illegal businessman mixes the seed of other kinds or underproof seed
Enter in qualified seed, causes crop failure, the serious interests for damaging grower.Traditional Seed purity test method mainly has
Morphological Identification, Seed Identification, physico-chemical analysis, protein electrophorese etc., but these methods need professional and professional equipment mostly,
It generally requires for a long time, there is destructiveness to seed specimen.Therefore, there is an urgent need to quickly and easily identify wheat product in the market
The method of kind.
In order to improve the rapidity of Seed inspection, since the 1990s, the technologies such as machine vision are examined in seed purity
Application in survey has obtained extensive research.Computer vision technique is a kind of new technology of substitution human vision sense organ, it can
By the mathematical model for establishing morphology characters of seeds (size, color, shape etc.), objective classification is carried out to seed variety.By
In the adequacy of collected characteristic information, reliability and validity directly affect the accuracy of classification.So data acquisition is
Most important part in whole experiment process.However, machine vision can only obtain the single traits of seed sample.High light spectrum image-forming
(HIS) system combines machine vision and spectroscopy, can obtain the spatially and spectrally information of seed to be analyzed.With it is single
Machine vision technique or spectral analysis technique are compared, and the information that HSI technologies provide includes the formalness feature of testee, interior
Portion's physical features and chemical composition.
Invention content
To solve the above problems, the present invention provides a kind of simple grain fixed point identification skills for different cultivars wheat seed
Art.
To achieve the above object, the technical solution that the present invention takes is:
A kind of simple grain for different cultivars wheat seed pinpoints identification technology, includes the following steps:
The acquisition of S1, hyperspectral image data
S11, the hyperspectral image data that wheat seed to be identified is obtained using EO-1 hyperion sorter, and pass through following formula
High spectrum image is corrected using black and white bearing calibration, to eliminate the influence of noise:
In formula, RdThe diffusing reflection image of blackboard;RwThe diffusing reflection image of blank;RsThe original spectrogram that diffuses of sample
Picture;The spectrogram that diffuses after R- corrections;
S12, principal component transform is carried out to the high-spectral data after correction, binaryzation is carried out to the 4th principal component figure (PC-4)
Processing obtains the binary map of wheat seed, as mask template, wherein target area pixel value is 1, remaining is all 0;
S13, it selects Intact wheat seed as area-of-interest, the figure after all masks of sample is obtained using following formula
Picture, i.e., remaining region is 0 entirely in addition to target sample.
BR=OR × Mask (2)
Wherein:BR is obtained image after mask;OR is the high spectrum image after correction;Mask is the 4th principal component of sample
Figure binaryzation;
The pretreatment of S2, spectroscopic data
Spectroscopic data is smoothed using S-G exponential smoothings, wherein smooth window size is 7;
The selection of S3, best band
Using Kennard-Stone methods, selection phase is concentrated according to spectroscopic data of 3: 2 ratios after convolution smoothing processing
The spectroscopic data answered collects respectively as training set and verification, and data are carried out to raw data set using successive projection algorithm (SPA)
Dimension-reduction treatment, which is chosen, is suitable for the optimal wave band that different Hybrid wheat seeds differentiate;
S4, the foundation for identifying model
Training set is marked off using the characteristic wave bands information and Kennard-Stone methods of above-mentioned acquisition and verification collection is established
Support vector machines (SVM) disaggregated model;
S5, seed mixture fixed point identification
In the SVM models that the optimal wave band input of gained is established, prediction result is indicated with different colours, completes mixing
The fixed point of seed identifies.
Preferably, 430nm to 980nm spectrum range data informations are selected in the step S2.
The present invention realizes the on-line checking of the wheat seed of different Hybrids, and realizes the fixed point identification of simple grain.
Description of the drawings
Fig. 1 is the sample disposing way in the embodiment of the present invention.
Fig. 2 is the mask template in the embodiment of the present invention.
Fig. 3 is that the characteristic wavelength in the embodiment of the present invention chooses figure.
Fig. 4 is the visualization mixing sample in the embodiment of the present invention.
Specific implementation mode
In order to make objects and advantages of the present invention be more clearly understood, the present invention is carried out with reference to embodiments further
It is described in detail.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to limit this hair
It is bright.
Embodiment
The selection of sample and hyperspectral image data obtain
The hybrid wheat seed sample (being shown in Table 1) used in the present embodiment is planted in Northeast China San Sheng extensively.Wherein according to
Two large attributes are belonging respectively to according to six kinds of hybrid wheat seeds of attribute, i.e. eastern agriculture winter wheat No. 1 (winter wheat), (winter is small for eastern agriculture winter wheat No. 2
Wheat), in gram spring No. 4 (spring wheat), gram drought No. 16 (spring wheat) cultivates nine No. 10 (spring wheat) and imperial good fortune wheat No. 21 (spring wheat).Its
In each kind samples selection 100,600 samples to be tested in total.Six kinds of hybrid wheats are obtained using EO-1 hyperion sorter
The hyperspectral image data of seed obtains the arrangement mode of the wheat seed of sample as shown in Figure 1, often arranging 8, one time 5 every time
Row.In high spectrum image gatherer process, since various objective condition limit, collected high light spectrum image-forming data include various make an uproar
Sound, if halogen light source light distribution is uneven, there are electronic noises and individual of sample difference to cause light in CCD camera hardware
According to unequal.These noise informations can influence the quality of high spectrum image, and then influence high spectrum image qualitative or quantitative analysis
The precision and stability of model.It is therefore desirable to be corrected to high spectrum image, to eliminate the influence of noise.Research is using black
White bearing calibration, formula are
In formula, RdThe diffusing reflection image of blackboard;RwThe diffusing reflection image of blank;RsThe original spectrogram that diffuses of sample
Picture;The spectrogram that diffuses after R- corrections.
Mainly by artificially delimiting when the spectroscopic data of tradition extraction specific region, take time and effort without objectivity, very
Hardly possible includes complete information.The present embodiment carries out principal component transform to the high-spectral data after correction first in terms of spectrum extraction,
Since the 4th principal component figure (PC-4) can most distinguish background and target.Therefore binary conversion treatment is carried out to PC-4 and obtains wheat seed
(target area pixel value is 1 to binary map, and 0) remaining is all, as shown in Fig. 2, as mask template.It is selected according to our purpose
Intact wheat seed obtains the image after all masks of sample as area-of-interest using formula (2), that is, removes target sample
Remaining outer region is 0 entirely.The averaged spectrum of relevant range is extracted as the spectroscopic data of this sample using ENVI4.7.Therefore,
Six kinds of hybrid wheat seeds obtain 600 groups of spectroscopic datas, data matrix 600*520 altogether.
BR=OR × Mask (2)
Wherein:BR is obtained image after mask;OR is the high spectrum image after correction;Mask is the 4th principal component of sample
Figure binaryzation.
1 sample message of table
The pretreatment of spectroscopic data
The spectroscopic data obtained is smoothed and can effectively eliminate since sample surface is uneven, noise of instrument
With high-frequency noise caused by the reasons such as random error, be conducive to improve signal-to-noise ratio.In addition to this, right before selecting optimal wavelength
Data carry out pre-processing the difference that can also be protruded between studied sample.(S-G is flat for Savitzky-Golay convolution exponential smoothing
It is sliding), it is the filtering method using polynomial least mean square fitting, is considered as the smooth method of average of Weight, more strong center
The effect of point.For spectrum x at the wavelength m it is smoothed after average value be formula (3).Therefore to initial data, (600*520 is tieed up
The data set of degree) S-G exponential smoothings (wherein smooth window size for 7) are used, spectrum is carried out smooth.And due to CCD camera
Spectral sensitivity near 400nm and 1000nm wavelength have lower signal-to-noise ratio.So selecting 430nm to 980nm spectrum
Interval censored data information is used for further data analysis.Through obtained number of the convolution smoothly and after the rejecting lower wave band of signal-to-noise ratio
According to integrating as 600*470.That is each sample is characterized by 470 wave bands, chooses what 470 wave bands were handled as later data
Spectroscopic data.
Wherein, H is normalization factor, can seek smoothing factor h with fitting of a polynomial according to the principle of least squarei, each
The purpose that absorbance at wavelength is multiplied with smoothing factor, which is to try to reduce, smooths out useful information.
The selection of best band
In order to correctly identify above-mentioned six kinds of hybrid wheat seed classifications, using Kennard-Stone methods, according to 3: 2 ratios
Rate selects 360 groups of spectroscopic datas therein as training set, and 240 groups of spectroscopic datas is used to collect as verification.By being obtained
Wheat seed hyperspectral image data wave band number it is more, the wave band that when on-line checking is obtained is more, and real-time is poorer.
It selects optimal wave band to be beneficial to reduce the complexity of model, improves the speed of service.Therefore difference is suitable for by mathematical method selection
The optimal wave band that Hybrid wheat seed differentiates is further analyzed.Successive projection algorithm (SPA) is Optimal Bands Selection
Forward direction selection method.It can be minimized the synteny between variable.Successive projection algorithm is used to raw data set at this
(SPA) Data Dimensionality Reduction processing is carried out to raw data set, obtains the characteristic data set of 600*12 dimensions.The wavelength chosen such as Fig. 3
Shown, the specific wavelength being related to is:450.12nm, 504.97nm, 535.69nm, 554.21nm, 711.01nm, 751.86nm,
808.51nm, 831.84nm, 877.47nm, 919.49nm, 952.52nm, 993.72nm.
Identify the foundation of model
Using after spectral data collection (600*520), convolution smoothing processing spectroscopic data collection (600*470) and rolling up
The data set (600*12) after SPA optimal screenings is carried out on the basis of product is smooth establishes support vector cassification mould respectively as input
Type.The results are shown in Table 2.By comparing untreated all-wave segment data, after SG processing, the nicety of grading after SG processing+SPA,
It was found that the correct recognition rata of SG pretreatment+SPA characteristic wave bands selection is more preferable.
2 support vector cassification result of table
Experiment six kinds of different hybrid wheats used can be divided into spring wheat and winter wheat according to sample attribute.Each kind
Affiliated type is shown in Table 1.We establish svm classifier model respectively to the hybrid wheat seed of different attribute.Wherein each wheat seed
The number of samples of grain is 100.Classification accuracy is as shown in table 3 between its class, and the classification accuracy between class is higher.
Classification results between 3 different attribute of table
Seed mixture fixed point identification
It is acquired using Gaia Sorter " Gai Ya " EO-1 hyperion sorter that Beijing Zolix Instrument Co., Ltd. provides mixed
The hyperspectral image data of the wheat seed sample of conjunction.Using 4.7 softwares of ENVI and digital image processing techniques is combined to extract height
Average spectral data in spectrum picture, inputs in the S-G+SPA+SVM models of foundation, and prediction result is indicated with different colours.
This example is simulation doping phenomenon, and the hybrid mode of the present embodiment is:4 kind of gram spring, a data are adulterated in eastern agriculture winter wheat No. 1
5 rows 8 row totally 40 samples are obtained, wherein preceding 6 being classified as totally 30, No. 1 sample of eastern agriculture winter wheat from left to right, next two columns are gram spring 4
Totally 10, sample.To this mixing identification acquired results as shown in figure 3, left side represents Dong Nong winter wheat No. 1, right side represents Ke Chun
4.The discrimination of doping 4 sample of gram spring in eastern agriculture winter wheat No. 1 is up to by the S-G+SPA+SVM models of secondary visible foundation
100%.Therefore the average spectral data combination S-G+SPA+SVM models that sample can be obtained by using EO-1 hyperion sorter obtain
Differentiate figure to visualization, can be very good to realize the impurity fixed point identification in doping wheat seed.
Conclusion
The simple grain that the wheat seed to mixing different cultivars may be implemented using Visible-to-Near InfaRed high light spectrum image-forming technology is fixed
Point identification.6 kinds (eastern agriculture winter wheat No. 1, east are collected using based on Visible-to-Near InfaRed (380-1038nm) EO-1 hyperion sorter
Agriculture winter wheat No. 2, gram spring 4, cultivates nine No. 10 and imperial good fortune wheat No. 21 at gram drought 16) hyperspectral image data of hybrid wheat seed.Make
It with Kennard-Stone methods, selects 360 groups of spectroscopic datas therein as training set according to 3: 2 ratios, and uses 240
Group spectroscopic data collects as verification.After analyzing hybrid wheat seed spectroscopic data, with S-G convolution exponential smoothing to spectroscopic data into
Row pretreatment excludes due to the bigger wave band of spectral noise signal caused by unstable at the beginning and end of instrument, chooses 470
The spectroscopic data that wave band is handled as later data.12 characteristic wave bands conducts are chosen again by successive projection algorithm (SPA)
Input, and combination supporting vector machine (SVM) model predicts calibration set six kinds of hybrid classification accuracy can reach
84.58%, pairwise classification accuracy highest can reach 100% between class.And gram spring is adulterated in eastern agriculture winter wheat No. 1 by simulation
4 kind wheats, as a result may indicate that can be real using Visible-to-Near InfaRed high light spectrum image-forming technology combination digital image processing techniques
Now the simple grain of the wheat seed to mixing different cultivars pinpoints identification.Side can be provided for quickly and easily identification wheat breed
It helps.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the principle of the present invention, it can also make several improvements and retouch, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (2)
1. a kind of simple grain for different cultivars wheat seed pinpoints identification technology, which is characterized in that include the following steps:
The acquisition of S1, hyperspectral image data
S11, the hyperspectral image data that wheat seed to be identified is obtained using EO-1 hyperion sorter, and used by following formula
Black and white bearing calibration is corrected high spectrum image, to eliminate the influence of noise:
In formula, RdThe diffusing reflection image of blackboard;RwThe diffusing reflection image of blank;RsThe original diffusing reflection spectrum image of sample;
The spectrogram that diffuses after R- corrections;
S12, principal component transform is carried out to the high-spectral data after correction, binary conversion treatment is carried out to the 4th principal component figure (PC-4)
The binary map of wheat seed is obtained, as mask template, wherein target area pixel value is 1, remaining is all 0;
S13, it selects Intact wheat seed as area-of-interest, the image after all masks of sample is obtained using following formula,
Remaining region is 0 entirely i.e. in addition to target sample;
BR=OR × Mask (2)
Wherein:BR is obtained image after mask;OR is the high spectrum image after correction;Mask is the 4th principal component figure two of sample
Value;
The pretreatment of S2, spectroscopic data
Spectroscopic data is smoothed using S-G exponential smoothings, wherein smooth window size is 7;
The selection of S3, best band
Using Kennard-Stone methods, concentrate selection corresponding according to spectroscopic data of 3: 2 ratios after convolution smoothing processing
Spectroscopic data collects respectively as training set and verification, and Data Dimensionality Reduction is carried out to raw data set using successive projection algorithm (SPA)
Processing, which is chosen, is suitable for the optimal wave band that different Hybrid wheat seeds differentiate;
S4, the foundation for identifying model
Training set is marked off using the characteristic wave bands information and Kennard-Stone methods of above-mentioned acquisition and verification collection is established and supported
Vector machine (SVM) disaggregated model;
S5, seed mixture fixed point identification
In the SVM models that the optimal wave band input of gained is established, prediction result is indicated with different colours, completes seed mixture
Fixed point identification.
2. a kind of simple grain for different cultivars wheat seed according to claim 1 pinpoints identification technology, feature exists
In selection 430nm to 980nm spectrum range data informations in the step S2.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810435100.5A CN108734205A (en) | 2018-04-28 | 2018-04-28 | A kind of simple grain for different cultivars wheat seed pinpoints identification technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810435100.5A CN108734205A (en) | 2018-04-28 | 2018-04-28 | A kind of simple grain for different cultivars wheat seed pinpoints identification technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108734205A true CN108734205A (en) | 2018-11-02 |
Family
ID=63938119
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810435100.5A Pending CN108734205A (en) | 2018-04-28 | 2018-04-28 | A kind of simple grain for different cultivars wheat seed pinpoints identification technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108734205A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109657653A (en) * | 2019-01-21 | 2019-04-19 | 安徽大学 | A kind of wheat seed head blight recognition methods based on Imaging Hyperspectral Data |
CN110095436A (en) * | 2019-05-30 | 2019-08-06 | 江南大学 | Apple slight damage classification method |
CN110763698A (en) * | 2019-10-12 | 2020-02-07 | 仲恺农业工程学院 | Hyperspectral citrus leaf disease identification method based on characteristic wavelength |
CN110837823A (en) * | 2019-12-17 | 2020-02-25 | 华南农业大学 | Method for generating seed variety identification model, identification method and device |
CN111144502A (en) * | 2019-12-30 | 2020-05-12 | 中国科学院长春光学精密机械与物理研究所 | Hyperspectral image classification method and device |
CN111272668A (en) * | 2020-01-22 | 2020-06-12 | 中国农业科学院农产品加工研究所 | Construction method of wheat variety identification model |
CN112052363A (en) * | 2020-09-02 | 2020-12-08 | 安阳工学院 | Grain identification method and system |
CN112147083A (en) * | 2020-10-14 | 2020-12-29 | 武汉轻工大学 | Seed purity detection method, detection device and computer readable storage medium |
CN114136920A (en) * | 2021-12-02 | 2022-03-04 | 华南农业大学 | Hyperspectrum-based single-grain hybrid rice seed variety identification method |
WO2023115682A1 (en) * | 2021-12-24 | 2023-06-29 | 湖南大学 | Hyperspectral traditional chinese medicine identification method based on adaptive random block convolutional kernel network |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103344602A (en) * | 2013-07-04 | 2013-10-09 | 中国科学院合肥物质科学研究院 | Nondestructive testing method for rice idioplasm authenticity based on near infrared spectrum |
CN103822879A (en) * | 2014-02-24 | 2014-05-28 | 西北农林科技大学 | Nondestructive detection method of swelled kiwi fruits based on hyperspectral imaging technology |
CN106290238A (en) * | 2016-08-09 | 2017-01-04 | 西北农林科技大学 | A kind of apple variety method for quick identification based on high light spectrum image-forming |
CN106295498A (en) * | 2016-07-20 | 2017-01-04 | 湖南大学 | Remote sensing image target area detection apparatus and method |
CN107527326A (en) * | 2017-08-17 | 2017-12-29 | 安徽农业大学 | A kind of wheat scab diagnostic method based on high light spectrum image-forming |
-
2018
- 2018-04-28 CN CN201810435100.5A patent/CN108734205A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103344602A (en) * | 2013-07-04 | 2013-10-09 | 中国科学院合肥物质科学研究院 | Nondestructive testing method for rice idioplasm authenticity based on near infrared spectrum |
CN103822879A (en) * | 2014-02-24 | 2014-05-28 | 西北农林科技大学 | Nondestructive detection method of swelled kiwi fruits based on hyperspectral imaging technology |
CN106295498A (en) * | 2016-07-20 | 2017-01-04 | 湖南大学 | Remote sensing image target area detection apparatus and method |
CN106290238A (en) * | 2016-08-09 | 2017-01-04 | 西北农林科技大学 | A kind of apple variety method for quick identification based on high light spectrum image-forming |
CN107527326A (en) * | 2017-08-17 | 2017-12-29 | 安徽农业大学 | A kind of wheat scab diagnostic method based on high light spectrum image-forming |
Non-Patent Citations (3)
Title |
---|
刘晶晶,刘付龙,史铁,孙超,张晋宝,门洪: "高光谱图像技术检测苹果外部损伤", 《中国食品学报》 * |
孙俊: "《光谱技术在农作物/农产品信息无损检测中的应用》", 30 June 2017 * |
赵明富,刘自迪,邹雪,吴亮,张峰,龙洁: "基于高光谱成像技术对马铃薯外部缺陷的识别", 《激光杂志》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109657653B (en) * | 2019-01-21 | 2022-10-04 | 安徽大学 | Wheat grain gibberellic disease identification method based on imaging hyperspectral data |
CN109657653A (en) * | 2019-01-21 | 2019-04-19 | 安徽大学 | A kind of wheat seed head blight recognition methods based on Imaging Hyperspectral Data |
CN110095436A (en) * | 2019-05-30 | 2019-08-06 | 江南大学 | Apple slight damage classification method |
CN110763698A (en) * | 2019-10-12 | 2020-02-07 | 仲恺农业工程学院 | Hyperspectral citrus leaf disease identification method based on characteristic wavelength |
CN110837823A (en) * | 2019-12-17 | 2020-02-25 | 华南农业大学 | Method for generating seed variety identification model, identification method and device |
CN111144502A (en) * | 2019-12-30 | 2020-05-12 | 中国科学院长春光学精密机械与物理研究所 | Hyperspectral image classification method and device |
CN111144502B (en) * | 2019-12-30 | 2023-02-10 | 中国科学院长春光学精密机械与物理研究所 | Hyperspectral image classification method and device |
CN111272668A (en) * | 2020-01-22 | 2020-06-12 | 中国农业科学院农产品加工研究所 | Construction method of wheat variety identification model |
CN112052363A (en) * | 2020-09-02 | 2020-12-08 | 安阳工学院 | Grain identification method and system |
CN112147083A (en) * | 2020-10-14 | 2020-12-29 | 武汉轻工大学 | Seed purity detection method, detection device and computer readable storage medium |
CN112147083B (en) * | 2020-10-14 | 2023-07-11 | 武汉轻工大学 | Seed purity detection method, detection device and computer readable storage medium |
CN114136920A (en) * | 2021-12-02 | 2022-03-04 | 华南农业大学 | Hyperspectrum-based single-grain hybrid rice seed variety identification method |
WO2023115682A1 (en) * | 2021-12-24 | 2023-06-29 | 湖南大学 | Hyperspectral traditional chinese medicine identification method based on adaptive random block convolutional kernel network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108734205A (en) | A kind of simple grain for different cultivars wheat seed pinpoints identification technology | |
CN110763698B (en) | Hyperspectral citrus leaf disease identification method based on characteristic wavelength | |
Gao et al. | Application of hyperspectral imaging technology to discriminate different geographical origins of Jatropha curcas L. seeds | |
US11682203B2 (en) | Feature extraction method, model training method, detection method of fruit spectrum | |
CN110596280B (en) | Rapid detection method for wheat vomitoxin pollution level based on hyperspectral image and spectrum information fusion | |
CN105067532B (en) | A kind of method for differentiating sclerotinia sclerotiorum and gray mold early stage scab | |
CN112699756B (en) | Hyperspectral image-based tea origin identification method and system | |
CN111443043B (en) | Hyperspectral image-based walnut kernel quality detection method | |
CN109001218A (en) | Apple surface defect quick nondestructive recognition methods based on high light spectrum image-forming technology | |
CN107679569A (en) | Raman spectrum substance automatic identifying method based on adaptive hypergraph algorithm | |
WO2024021359A1 (en) | Built environment dominant color measurement method and system based on image eeg sensitivity data | |
CN102706813A (en) | Poa pratensis variety identification method based on hyper-spectral image | |
CN113570538A (en) | Blade RGB image skewed distribution parameter information acquisition and analysis method | |
CN115494008A (en) | Tobacco leaf quality detection method and system combining near-infrared spectrometer and machine vision | |
CN117132843B (en) | Wild ginseng, under-forest mountain ginseng and garden ginseng in-situ identification method, system and related equipment | |
Sola-Guirado et al. | A smart system for the automatic evaluation of green olives visual quality in the field | |
CN110533102A (en) | Single class classification method and classifier based on fuzzy reasoning | |
CN104990891B (en) | A kind of seed near infrared spectrum and spectrum picture qualitative analysis model method for building up | |
CN107271394A (en) | A kind of fuzzy Kohonen differentiates the tealeaves infrared spectrum sorting technique of clustering network | |
CN109685099A (en) | A kind of apple variety discriminating conduct of the preferred fuzzy clustering of spectral band | |
CN115494009A (en) | Intelligent discrimination method and system for tobacco leaf maturity | |
CN114689526A (en) | Banana wilt nondestructive detection method, device and detection equipment | |
CN114778457A (en) | Method and device for detecting aflatoxin B1 content in grains | |
CN113390799A (en) | Method for identifying and detecting stems in tobacco leaves | |
Friska et al. | Random Forest (RF) based identification of rice powder mixture using terahertz spectroscopy |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181102 |