CN116341223B - Novel three-band spectrum index-based rice spike rot disease severity estimation method - Google Patents
Novel three-band spectrum index-based rice spike rot disease severity estimation method Download PDFInfo
- Publication number
- CN116341223B CN116341223B CN202310236635.0A CN202310236635A CN116341223B CN 116341223 B CN116341223 B CN 116341223B CN 202310236635 A CN202310236635 A CN 202310236635A CN 116341223 B CN116341223 B CN 116341223B
- Authority
- CN
- China
- Prior art keywords
- disease
- spectrum
- severity
- rice
- pixels
- 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
- 201000010099 disease Diseases 0.000 title claims abstract description 114
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 114
- 235000007164 Oryza sativa Nutrition 0.000 title claims abstract description 76
- 235000009566 rice Nutrition 0.000 title claims abstract description 76
- 238000001228 spectrum Methods 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 45
- 240000007594 Oryza sativa Species 0.000 title description 2
- 241000209094 Oryza Species 0.000 claims abstract description 75
- 210000005069 ears Anatomy 0.000 claims abstract description 12
- 238000010219 correlation analysis Methods 0.000 claims abstract description 10
- 238000012216 screening Methods 0.000 claims abstract description 6
- 238000000985 reflectance spectrum Methods 0.000 claims abstract description 5
- 238000009395 breeding Methods 0.000 claims description 10
- 230000001488 breeding effect Effects 0.000 claims description 10
- 238000002310 reflectometry Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 7
- 238000005520 cutting process Methods 0.000 claims description 6
- 230000017260 vegetative to reproductive phase transition of meristem Effects 0.000 claims description 6
- 230000005855 radiation Effects 0.000 claims description 5
- 238000009826 distribution Methods 0.000 claims description 4
- 241000196324 Embryophyta Species 0.000 claims description 3
- 239000008186 active pharmaceutical agent Substances 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000003892 spreading Methods 0.000 claims description 3
- 230000007480 spreading Effects 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 abstract description 19
- 238000012544 monitoring process Methods 0.000 abstract description 13
- 230000035558 fertility Effects 0.000 abstract description 7
- 230000004044 response Effects 0.000 abstract description 6
- 238000010276 construction Methods 0.000 abstract description 5
- 230000035945 sensitivity Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 239000000049 pigment Substances 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 235000013339 cereals Nutrition 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 208000032625 disorder of ear Diseases 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 210000000056 organ Anatomy 0.000 description 2
- 208000031888 Mycoses Diseases 0.000 description 1
- 238000012352 Spearman correlation analysis Methods 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 238000000701 chemical imaging Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 238000011081 inoculation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001766 physiological effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/47—Scattering, i.e. diffuse reflection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Geometry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a novel three-band spectral index-based rice spike rot disease severity estimation method, which comprises the following steps of: extracting reflectance spectrum of rice ears and disease severity of a sample; extracting characteristic wave bands of a plurality of spectrum intervals by utilizing spectral response characteristics and correlation analysis of multiple growth periods; creating a three-band spectral index to increase sensitivity of the spectral signature to severity of the condition; and finally, screening out the optimal wave band combination through accuracy comparison to estimate the severity of the spike rot. According to the invention, characteristic wave bands sensitive to disease severity are captured in multiple fertility periods, and disease estimation of construction indexes in different fertility periods is applicable, so that the estimation accuracy of early disease is remarkably improved; the method has simple steps and low operation cost, and the novel spectrum index selected wave band is commonly used for the unmanned aerial vehicle multispectral camera with low cost, so that the method can be widely applied to rice germplasm resource screening and disease monitoring and control work.
Description
Technical Field
The invention belongs to the field of crop disease monitoring, and particularly relates to a monitoring method for constructing a novel disease index and improving the accuracy of estimating the severity of early disease of rice spike rot.
Background
The rice has important roles in grain production in China, and timely and accurately monitors the occurrence of rice diseases and plays a fundamental role in guaranteeing grain safety in China and reducing medicine and improving efficiency in agricultural production. Rice spike rot is a fungal disease, is influenced by the change of environmental climate, cultivar and planting mode, and is widely generated in main rice producing areas in China in recent years as an emerging disease, so that the quality and the high yield of rice are seriously endangered.
At present, a plurality of scholars use spectroscopic technology to effectively identify and evaluate crop diseases. The usual way is to build a feature. These features mainly include two categories, spectral parameters and vegetation physiological structure parameters. Physiological structural parameters are usually obtained by inversion of vegetation radiation transmission models, and such modes are difficult to apply to rice ears due to the lack of a reliable ear radiation transmission model. For spectral parameters, part of the method requires the use of a continuous piece or a large amount of spectral information. For example, continuous wavelet transform relies on continuous spectral signals for feature construction, is difficult to use on broadband data, and is too costly to apply. The vegetation index method combines a few characteristic wave bands in a specific mathematical form, improves the sensitivity of spectral characteristics, and simultaneously can remarkably reduce the wave band quantity requirement, so that the practical application potential is outstanding. Because of certain specificity of the spectral response characteristics of diseases, the traditional vegetation indexes such as normalized vegetation index NDVI and the like are used for estimating the disease severity with lower accuracy. There is currently a lack of a spectral index specifically for estimating the severity of rice ear rot.
At present, disease monitoring based on a spectrum characteristic technology is seriously dependent on spectrum response caused by change of physiological and biochemical parameters of crops in the disease infection process. However, the physiological properties of the ear organs of crops develop with the period of fertility, and also vary significantly. The existing ear disease spectrum monitoring research is mostly based on the disease monitoring research within days after inoculation or based on a single-period data sample, whether the disease spectrum response rules of different breeding periods are the same or not, and whether the disease severity estimation accuracy is consistent or not is not clear. For the ear disease, there is a certain correlation between the growing period and the disease severity, the rice ear in the heading period is basically in early stage of disease, the disease spectral response in the stage is weak, the influence of the rice ear maturity is more easily caused, and the feasibility and the precision of early monitoring of the rice ear rot are still unclear. In view of this, there is still a lack of a strong-mechanistic index construction method, so that the constructed rice ear rot index can be used for disease estimation in multiple growth periods, and has higher estimation accuracy in early disease.
Disclosure of Invention
The technical problem solved by the invention is to provide a novel three-band spectrum index-based estimation method for the disease severity of rice ear rot, which comprises the steps of extracting the reflection spectrum of rice ears and the disease severity of samples; extracting characteristic wave bands of a plurality of spectrum intervals by utilizing spectral response characteristics and correlation analysis of multiple growth periods; creating a three-band spectral index to increase sensitivity of the spectral signature to severity of the condition; and finally, screening out the optimal wave band combination through accuracy comparison to estimate the severity of the spike rot. According to the invention, characteristic wave bands sensitive to disease severity are captured in multiple fertility periods, and disease estimation of construction indexes in different fertility periods is applicable, so that the estimation accuracy of early disease is remarkably improved; the method has simple steps and low operation cost, and the novel spectrum index selected wave band is commonly used for the unmanned aerial vehicle multispectral camera with low cost, so that the method can be widely applied to rice germplasm resource screening and disease monitoring and control work.
The technical solution for realizing the purpose of the invention is as follows:
a novel three-band spectrum index-based rice spike rot disease severity estimation method comprises the following steps:
step 1: obtaining reflection spectra and digital pictures of different disease severity spike samples in a plurality of growth periods;
step 2: disease severity based on digital photo extraction of a sample of a disease
Step 3: extracting a plurality of characteristic wave bands by utilizing correlation analysis and reflection spectrum waveform analysis;
step 4: and constructing a double-difference spectrum index based on the characteristic wave band, determining an optimal wave band combination, and estimating the severity of the rice spike rot in the multiple growth periods.
Furthermore, the invention provides a novel three-band spectrum index-based rice spike rot disease severity estimation method, which comprises the following specific steps of:
step 1-1: cutting 5-8 infected rice ears from plants and spreading the cut rice ears on a black reference plate at a certain interval to improve the acquisition efficiency when hyperspectral images and digital photos are acquired each time;
step 1-2: carrying out radiation correction on the obtained organ-scale hyperspectral image by using a standard reflectivity whiteboard to obtain reflectivity data of the hyperspectral image; removing black background with reflectivity lower than 0.2 by using a threshold method;
step 1-3: the ROI tool using ENVI software cuts out the hyperspectral image of each rice spike from it, averages the sample hyperspectral cubes in the spatial dimension, and obtains the average reflectance spectrum for each sample.
Further, the invention provides a method for estimating the disease severity of rice spike rot based on a novel three-band spectrum index, and the specific steps of the step 2 include:
step 2-1: cutting each ear sample from the original digital photograph using an ROI tool of ENVI software; and the reflection spectrums of the samples obtained in the step 1 are in one-to-one correspondence with each other according to the test sequence;
step 2-2: converting the digital photograph of the ear sample into a Lab color space image based on the color space conversion;
step 2-3: generating a threshold value for background removal by combining a 'b' wave band of a Lab image based on a dynamic threshold value method in a skimage tool package in a python environment, setting a block size parameter of the threshold value method to be 2001 pixels, wherein pixels larger than the threshold value are rice spike pixels, and pixels smaller than the threshold value are background noise and remove a background mask;
step 2-4: generating a threshold value for identifying a disease-sensing pixel by combining an 'a' wave band of a Lab image based on a dynamic threshold value method in a skimage tool package in a python environment, wherein a block size parameter of the threshold value method is set to 1001 pixels, pixels larger than the threshold value are disease-sensing pixels, and pixels smaller than the threshold value are healthy pixels;
step 2-5: defining quantitative severity of illness, the calculation formula is as follows:
wherein DS represents the disease spot proportion of the rice spike rot of each disease spike sample, namely the disease severity; n is n d And N represents the total number of the infected pixels and the total number of the rice spike pixels, respectively.
Furthermore, the invention provides a novel three-band spectrum index-based rice spike rot disease severity estimation method, and the specific steps of characteristic band extraction in the step 3 comprise the following steps:
step 3-1: dividing all the spike samples into a plurality of sub-data sets according to the sample growth period, carrying out normal examination on the illness severity of the samples of each sub-data set, using Piercan correlation analysis if the normal distribution is met, using Szechwan correlation analysis if the normal distribution is not met, and calculating the correlation coefficient of each wave band and illness severity;
step 3-2: dividing the reflection spectrum into a positive correlation interval and a negative correlation interval which are associated with the disease severity according to the positive and negative of the correlation coefficient values, selecting a wave band with the highest absolute value of the correlation coefficient of the disease severity in each correlation interval as a characteristic wave band, and respectively extracting the characteristic wave bands in the corresponding sub-data set of each growth period.
Further, the invention provides a novel three-band spectrum index-based rice spike rot disease severity estimation method, wherein in the step 4, characteristic bands are screened and spectrum indexes are constructed:
step 4-1: a double difference (Double difference) form spectral index was constructed, the formula of which is:
wherein R is λ1 ,R λ2 ,and R λ3 The three characteristic wave bands in different steps 3 are respectively arranged according to the wave band numbers in an increasing order.
Step 4-2: selecting a red light interval and a near infrared interval with highest correlation coefficients in a plurality of growth periods from the characteristic wave bands extracted in the step 3; the characteristic wave bands of the red light section and the near infrared section in different breeding periods are unified by approximating the characteristic wave band wavelength of the relevant section in the multiple breeding periods, and the characteristic wave bands are used as two fixed characteristic wave bands of disease indexes. The research shows that the blue light wave band has high importance for early disease monitoring, a plurality of double-difference spectrum indexes are respectively constructed with the fixed wave band according to the characteristic wave band of the residual blue light spectrum interval, and the spectrum index with the optimal performance is selected as the final disease index by comparing the disease estimation performance of the multiple growth periods.
Compared with the prior art, the technical scheme provided by the invention has the following technical effects:
1. the rice spike rot disease severity estimation method based on the novel three-band spectrum index has the characteristics of high characteristic band selection speed and simple index construction;
2. according to the novel three-band spectrum index-based rice spike rot disease severity estimation method, the constructed disease spectrum index is suitable for disease estimation precision in a plurality of growth periods after heading, and the disease early estimation precision can be remarkably improved;
3. according to the rice spike rot disease severity estimation method based on the novel three-band spectrum index, the novel index selected band is commonly used for a low-cost unmanned aerial vehicle-mounted multispectral camera, and can be widely used for germ plasm screening and disease monitoring in rice production.
Drawings
FIG. 1 is a flow chart of a method for constructing disease index and improving the accuracy of estimating the disease severity of early stage of rice ear rot disease based on spectral analysis;
FIG. 2 is a schematic diagram of calculating correlation coefficients and characteristic band extraction of each band and severity of illness using Szellman correlation analysis in the example;
FIG. 3 is a graph comparing the prior art index and the spectrum index constructed by the invention with the severity of rice ear rot;
FIG. 4 is a scatter plot of a model for estimating severity of disease for a sample of a prior spectral index and an index obtained according to the invention, validated at different stages of birth.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for explaining the present invention and are not to be construed as limiting the present invention.
The invention is based on the ear samples with different breeding periods and different disease severity under the cell test condition, extracts characteristic wave bands sensitive to diseases according to the acquired reflection spectrum, is used for constructing disease spectrum indexes, is used for estimating the disease severity of the multi-period rice ear rot, and the basic information and data acquisition conditions of specific rice tests are shown in table 1. The number of susceptible rice ears obtained corresponding to the different breeding periods is shown in Table 2.
TABLE 1 basic information of Rice test fields
TABLE 2 number of samples of rice ears at different stages of fertility
| Period of fertility | Number of samples |
| Heading stage | 57 |
| Flowering period | 161 |
| Grouting period | 184 |
And obtaining organ-scale disease-sensitive rice spike reflection spectrum and high-resolution digital pictures from a rice test field in the rice spike stage, the flowering stage and the grouting stage respectively. The experimental data acquisition system is strong, covers common rice varieties in Jiangsu province, comprises the optimal control period and the important disease period, has a large disease severity range and large sample number, and can effectively verify the accuracy and adaptability of estimating the disease severity of the rice spike rot under different cultivars, different breeding periods and different disease sensing periods.
As shown in FIG. 1, the method for estimating the disease severity of the rice spike rot based on the novel three-band spectrum index specifically comprises the following steps:
step 1: and obtaining hyperspectral images and digital photographs of rice ears with the feeling of rice ear rot under organ scale in multiple growth periods.
The method comprises the steps of selecting a heading period, a flowering period and a grouting period of rice, and carrying imaging hyperspectral devices and single phase inverters to acquire spectral data and digital photos before and after midnoon under a clear weather condition. Imaging spectrum measurement adopts GaiaField-V10E push-broom hyperspectral camera manufactured by Jiangsu Shuangli He Spectrum (Dualix Spectral Imaging) company, and the used wave band range is 450-800nm. The digital photograph was taken using an EOS 80D single lens reflex camera manufactured by Canon, inc., with a maximum resolution of 6000 x 4000 pixels. The acquisition mode of the spike image specifically comprises the following steps:
step 1-1: and selecting the rice ears with different severity degrees, cutting off the rice ears from plants, spreading 5-8 samples on a black reference plate each time, and placing the samples under a spectrum camera for data acquisition. The lens was adjusted to 0.4 meters from the sample height and the camera exposure was set to 0.4s to avoid overexposure of the white reference plate.
Step 1-2: carrying out radiation correction on the obtained organ-scale hyperspectral image by using a standard reflectivity whiteboard to obtain reflectivity data of the hyperspectral image; removing black background with reflectivity lower than 0.2 by using a threshold method;
step 1-3: the ROI tool using ENVI software cuts out the hyperspectral image of each rice spike from it, averages the sample hyperspectral cubes in the spatial dimension, and obtains the average reflectance spectrum for each sample.
Step 2: disease severity of the samples of the illness were extracted based on digital photographs:
step 2-1: cutting each ear sample from the original digital photograph using an ROI tool of ENVI software; and the hyperspectral images of the samples obtained in the step 1 are in one-to-one correspondence with each other according to the test sequence;
step 2-2: converting the digital photograph of the ear sample into a Lab color space image based on the color space conversion;
step 2-3: generating a threshold value for background removal by combining a 'b' wave band of a Lab image based on a dynamic threshold value method in a skimage tool package in a python environment, setting a block size parameter of the threshold value method to be 2001 pixels, wherein pixels larger than the threshold value are rice spike pixels, and pixels smaller than the threshold value are background noise and remove a background mask;
step 2-4: generating a threshold value for identifying a disease-sensing pixel by combining an 'a' wave band of a Lab image based on a dynamic threshold value method in a skimage tool package in a python environment, wherein a block size parameter of the threshold value method is set to 1001 pixels, pixels larger than the threshold value are disease-sensing pixels, and pixels smaller than the threshold value are healthy pixels;
step 2-5: defining quantitative severity of illness, the calculation formula is as follows:
wherein DS represents the disease spot proportion of the rice spike rot of each disease spike sample, namely the disease severity; n is n d And N represents the total number of the infected pixels and the total number of the rice spike pixels, respectively.
Step 3: as shown in fig. 2, a plurality of characteristic bands are extracted using correlation analysis and reflectance spectrum waveform analysis:
step 3-1: dividing all the spike samples into a plurality of sub-data sets according to the sample growth period, wherein the disease severity of the example spike samples does not pass normal examination, and calculating the correlation coefficient of each wave band and the disease severity by using spearman correlation analysis;
step 3-2: dividing the reflection spectrum into a positive correlation interval and a negative correlation interval which are associated with the disease severity according to the positive and negative of the correlation coefficient values, selecting a wave band with the highest absolute value of the correlation coefficient of the disease severity in each correlation interval as a characteristic wave band, and respectively extracting the characteristic wave bands in the corresponding sub-data set of each growth period.
Step 4: constructing a double-difference spectrum index based on characteristic wave bands, determining an optimal wave band combination, and estimating the disease severity of the rice spike rot in multiple growth periods:
step 4-1: a double difference (Double difference) form spectral index was constructed, the formula of which is:
wherein R is λ1 ,R λ2 ,and R λ3 The three characteristic wave bands in different steps 3 are respectively arranged according to the wave band numbers in an increasing order.
Step 4-2: selecting a red light interval and a near infrared interval with highest correlation coefficients in a plurality of growth periods from the characteristic wave bands extracted in the step 3; the characteristic wave bands of the red light section and the near infrared section in different breeding periods are unified by approximating the characteristic wave band wavelength of the relevant section in the multiple breeding periods, and the characteristic wave bands are used as two fixed characteristic wave bands of disease indexes. The characteristic wave bands of the red light range and the near infrared range selected by the example are 675nm and 740nm respectively. The research shows that the blue light wave band has high importance for early disease monitoring, a plurality of double-difference spectrum indexes are respectively constructed with the fixed wave band according to the characteristic wave band of the residual blue light spectrum interval, and the spectrum index with the optimal performance is selected as the final disease index by comparing the disease estimation performance of the multiple growth periods. The characteristic band of the blue light interval selected for the example was 453nm.
The method provided by the invention is compared with normalized pigment index NPCI, photochemical index PRI and normalized vegetation index NDVI, and the estimated severity of the rice ear rot disease is represented in early onset and multiple growth periods.
The relation between the existing index and the spectrum index constructed by the present invention and the severity of rice ear rot is shown in fig. 3 (wherein the first row to the fourth row correspond to the disease severity estimation models based on the rice ear rot index RSRI, the normalized pigment index NPCI, the photochemical index PRI670 and the NDVI, respectively; the rice ear rot disease severity estimation models corresponding to the heading period, the flowering period and the grouting period, respectively, from the first column to the third column), and the model verification scatter diagram constructed by each method is shown in fig. 4 (wherein the first row to the fourth row correspond to the scatter diagram verified by the disease severity estimation models based on the rice ear rot index RSRI, the normalized pigment index NPCI, the photochemical index PRI670 and the NDVI, respectively; the scatter diagram verified by the rice ear rot disease severity estimation models corresponding to the heading period, the flowering period and the grouting period, respectively, from the first column to the third column). As can be obtained from fig. 3 and fig. 4, the spectrum index generated by the method is superior to other spectrum indexes in terms of disease early disease severity estimation and disease severity estimation in multiple growth periods, and the method can remarkably improve the disease estimation precision in the heading period (same as the early disease), simultaneously gives consideration to the disease estimation precision in other growth periods, eliminates the influence of the growth period on the monitoring of the rice spike rot, and improves the adaptability of the rice spike disease monitoring in actual production.
While only a few embodiments of the present invention have been described, it should be appreciated by those skilled in the art that modifications could be made without departing from the principles of the present invention, which would be within the scope of the invention.
Claims (3)
1. A novel three-band spectrum index-based rice spike rot disease severity estimation method comprises the following steps:
step 1: obtaining reflection spectra and digital pictures of different disease severity spike samples in a plurality of growth periods;
step 2: extracting severity of illness from the sample based on the digital photograph;
step 3: extracting a plurality of characteristic wave bands by utilizing correlation analysis and reflection spectrum waveform analysis; the specific steps of extracting the characteristic wave band in the step 3 include:
step 3-1: dividing all the spike samples into a plurality of sub-data sets according to the sample growth period, carrying out normal examination on the illness severity of the samples of each sub-data set, using Piercan correlation analysis if the normal distribution is met, using Szechwan correlation analysis if the normal distribution is not met, and calculating the correlation coefficient of each wave band and illness severity;
step 3-2: dividing the reflection spectrum into a positive correlation interval and a negative correlation interval which are associated with the disease severity according to the positive and negative of the correlation coefficient values, selecting a wave band with the highest absolute value of the correlation coefficient of the disease severity in each correlation interval as a characteristic wave band, and respectively extracting the characteristic wave bands in the corresponding sub-data set of each growth period;
step 4: constructing a double-difference double spectrum form spectrum index based on the characteristic wave band, determining an optimal wave band combination, and estimating the severity of the rice spike rot in a multiple growth period; screening characteristic wave bands and constructing spectrum indexes in the step 4:
step 4-1: the double difference spectrum index is constructed, and the formula is as follows:
wherein R is λ1 ,R λ2 ,R λ3 The three characteristic wave bands in different steps 3 are respectively arranged according to the wavelength increasing sequence of the wave band numbers;
step 4-2: selecting a red light interval and a near infrared interval with highest correlation coefficients in a plurality of growth periods from the characteristic wave bands extracted in the step 3; approximating according to the wavelength of the characteristic wave bands of the relevant sections in multiple breeding periods, thereby unifying the characteristic wave bands of the red light section and the near infrared section in different breeding periods, and taking the characteristic wave bands as two fixed characteristic wave bands of disease indexes; and respectively establishing a plurality of double difference value double spectrum form spectrum indexes with the two fixed characteristic wave bands according to the characteristic wave bands of the residual blue spectrum interval, comparing the disease estimation performance of the double difference value double spectrum form spectrum indexes in the multiple growth periods, and selecting the spectrum index with the highest estimation accuracy sum in the heading period, the flowering period and the grouting period as the final disease index.
2. The method for estimating the disease severity of rice spike rot based on the novel three-band spectrum index as claimed in claim 1, wherein the specific steps of acquiring data of different disease severity in multiple growth periods in step 1 comprise:
step 1-1: cutting 5-8 infected rice ears from plants and spreading the cut rice ears on a black reference plate at a certain interval to improve the acquisition efficiency when hyperspectral images and digital photos are acquired each time;
step 1-2: carrying out radiation correction on the obtained organ-scale hyperspectral image by using a standard reflectivity whiteboard to obtain reflectivity data of the hyperspectral image; removing black background with reflectivity lower than 0.2 by using a threshold method;
step 1-3: the ROI tool using ENVI software cuts out the hyperspectral image of each rice spike from it, averages the sample hyperspectral cubes in the spatial dimension, and obtains the average reflectance spectrum for each sample.
3. The method for estimating the disease severity of rice ear rot based on the novel three-band spectrum index as claimed in claim 1, wherein the specific steps of the step 2 include:
step 2-1: cutting each ear sample from the original digital photograph using an ROI tool of ENVI software; and the reflection spectrums of the samples obtained in the step 1 are in one-to-one correspondence with each other according to the test sequence;
step 2-2: converting the digital photograph of the ear sample into a Lab color space image based on the color space conversion;
step 2-3: generating a threshold value for background removal by combining a 'b' wave band of a Lab image based on a dynamic threshold value method in a skimage tool package in a python environment, setting a block size parameter of the threshold value method to be 2001 pixels, wherein pixels larger than the threshold value are rice spike pixels, and pixels smaller than the threshold value are background noise and remove a background mask;
step 2-4: generating a threshold value for identifying a disease-sensing pixel by combining an 'a' wave band of a Lab image based on a dynamic threshold value method in a skimage tool package in a python environment, wherein a block size parameter of the threshold value method is set to 1001 pixels, pixels larger than the threshold value are disease-sensing pixels, and pixels smaller than the threshold value are healthy pixels;
step 2-5: defining quantitative severity of illness, the calculation formula is as follows:
wherein DS represents the disease spot proportion of the rice spike rot of each disease spike sample, namely the disease severity; n is n d And N represents the total number of the infected pixels and the total number of the rice spike pixels, respectively.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2022115682667 | 2022-12-07 | ||
| CN202211568266 | 2022-12-07 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116341223A CN116341223A (en) | 2023-06-27 |
| CN116341223B true CN116341223B (en) | 2024-01-16 |
Family
ID=86887049
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310236635.0A Active CN116341223B (en) | 2022-12-07 | 2023-03-13 | Novel three-band spectrum index-based rice spike rot disease severity estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116341223B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016000088A1 (en) * | 2014-07-02 | 2016-01-07 | 中国林业科学研究院林业研究所 | Hyperspectral waveband extraction method based on optimal index factor-correlation coefficient method |
| CN110132862A (en) * | 2019-05-30 | 2019-08-16 | 安徽大学 | Wheat scab detects exclusive disease index construction method and its application |
| CN110132856A (en) * | 2019-05-18 | 2019-08-16 | 安徽大学 | Wheat scab catch an illness seed identification spectrum disease index construction and application |
| CN111832507A (en) * | 2020-07-20 | 2020-10-27 | 安徽大学 | Wheat head top spectrum information-based wheat scab remote sensing identification method |
| CN113155747A (en) * | 2021-01-15 | 2021-07-23 | 西安科技大学 | Method for constructing fractional order differential spectral index and monitoring wheat stripe rust thereof |
| CN114694023A (en) * | 2022-03-18 | 2022-07-01 | 河南农业大学 | Cotton aphid severity grading method based on spectral index data reconstruction |
-
2023
- 2023-03-13 CN CN202310236635.0A patent/CN116341223B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016000088A1 (en) * | 2014-07-02 | 2016-01-07 | 中国林业科学研究院林业研究所 | Hyperspectral waveband extraction method based on optimal index factor-correlation coefficient method |
| CN110132856A (en) * | 2019-05-18 | 2019-08-16 | 安徽大学 | Wheat scab catch an illness seed identification spectrum disease index construction and application |
| CN110132862A (en) * | 2019-05-30 | 2019-08-16 | 安徽大学 | Wheat scab detects exclusive disease index construction method and its application |
| CN111832507A (en) * | 2020-07-20 | 2020-10-27 | 安徽大学 | Wheat head top spectrum information-based wheat scab remote sensing identification method |
| CN113155747A (en) * | 2021-01-15 | 2021-07-23 | 西安科技大学 | Method for constructing fractional order differential spectral index and monitoring wheat stripe rust thereof |
| CN114694023A (en) * | 2022-03-18 | 2022-07-01 | 河南农业大学 | Cotton aphid severity grading method based on spectral index data reconstruction |
Non-Patent Citations (3)
| Title |
|---|
| Multispectral Detection of Citrus Canker Using Hyperspectral Band Selection;Jianwei Qin et al.;Transactions of the ASABE;第54卷(第6期);第2331-2341页 * |
| New Optimized Spectral Indices for Identifying and Monitoring Winter Wheat Diseases;Wenjiang Huang et al.;IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing;第7卷(第6期);第2516-2524页 * |
| 基于新型光谱指数的小麦赤霉病检测研究;王倩;中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑;第2022卷(第03期);第C028-334页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116341223A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11029251B2 (en) | Method for estimating aboveground biomass of rice based on multi-spectral images of unmanned aerial vehicle | |
| CN110132862B (en) | Construction method and application of disease index special for wheat scab detection | |
| CN102495005B (en) | Method for diagnosing crop water deficit through hyperspectral image technology | |
| CN110132860B (en) | Winter wheat scab hyperspectral remote sensing monitoring method based on wheat ear scale analysis | |
| CN109389049A (en) | Crop Classification in Remote Sensing Image method based on multidate SAR data and multispectral data | |
| CN114821362B (en) | Multi-source data-based rice planting area extraction method | |
| CN110763698A (en) | Hyperspectral citrus leaf disease identification method based on characteristic wavelength | |
| CN109827957B (en) | Rice leaf SPAD value estimation method and system based on computer vision | |
| CN101692037A (en) | Method for analyzing chlorophyll distribution on surface of leaves of plant by hyperspectral image and independent component | |
| CN110647932B (en) | Planting crop structure remote sensing image classification method and device | |
| CN108195784A (en) | A kind of multispectral estimating and measuring method of tomato chlorophyll content | |
| CN112287886B (en) | Wheat plant nitrogen content estimation method based on hyperspectral image fusion map features | |
| CN111553922A (en) | Automatic cloud detection method for satellite remote sensing image | |
| KR102095539B1 (en) | Method for measuring growth amount by image analyzing ginseng | |
| CN115271217A (en) | Wheat yield prediction method based on multi-source remote sensing data of unmanned aerial vehicle | |
| CN112446298A (en) | Hyperspectral nondestructive testing method for wheat scab | |
| CN114140695B (en) | Prediction method and system for tea tree nitrogen diagnosis and quality index determination based on unmanned aerial vehicle multispectral remote sensing | |
| CN114778483A (en) | Method for correcting terrain shadow of remote sensing image near-infrared wave band for monitoring mountainous region | |
| CN106568730B (en) | A kind of rice yin-yang leaf fringe recognition methods based on Hyperspectral imaging near the ground | |
| CN113570538A (en) | Blade RGB image skewed distribution parameter information acquisition and analysis method | |
| CN115424006A (en) | Multi-source multi-level data fusion method applied to crop phenotypic parameter inversion | |
| CN112861712B (en) | Agricultural pest and disease monitoring method based on artificial intelligence and multi-temporal remote sensing | |
| CN111242934A (en) | Wheat nitrogen concentration inversion method based on unmanned aerial vehicle image spectrum and texture features | |
| CN112329733B (en) | Winter wheat growth monitoring and analyzing method based on GEE cloud platform | |
| CN116341223B (en) | Novel three-band spectrum index-based rice spike rot disease severity estimation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |