CN101699315A - Monitoring device and method for crop growth uniformity - Google Patents

Abstract

The invention discloses a monitoring device and a method for crop growth uniformity degree, aiming at the problem of low efficiency for manually investigating data. The device comprises a remote sensing image processing module, a plot vector data processing module, a vegetation index processing module and a growth uniformity processing module. The method comprises: obtaining the remote sensing images of a satellite; carrying out radiometric ratification, atmospheric correction and geometric correction according to the remote sensing images of the satellite; classifying crops in the remote sensing images of a satellite to obtain a spatial distribution map; converting a grid classification result in the spatial distribution map into facet vector data; processing the spatial distribution map to correct the plot boundary of the crops; according to the spectral characteristics of the plot in the remote sensing image, calculating the vegetation index of the plot; and calculating the growth uniformity degree index according to the vegetation index of each plot. The invention uses the integration of the grid and the vector data to monitor the growth uniformity degree of crops in natural plots and performs the advantages of remote sensing data.

Description

A kind of monitoring device of crop growth uniformity and method

Technical field

The present invention relates to a kind of monitoring device and method of crop growth uniformity.

Background technology

Crop growth monitoring refers to the macroscopic view monitoring of the growth of cereal crop seedlings, upgrowth situation and variation thereof to crop, and requirement can reflect agricultural feelings in time comprehensively.The growth uniformity of crop is an important indicator estimating the crop growing state quality always, and this index not only can embody the difference of different farmland massifs basis soil fertility, landform, also can reflect different tillers' management level.

How the tradition crop growth uniformity is determined by the field sample survey.As the patriotic basic seedling uniformity coefficient of the winter wheat index P that waits the calendar year 2001 proposition of horse, adopt the method for grab sample exactly.This method is counted according to definite investigation of each block area, then the seedling amount of every each points for investigation of ground is arranged from big to small.If the first half points for investigation seedling amount sum is ∑ a, half points for investigation seedling amount sum of back is ∑ b, calculates the basic seedling uniformity coefficient P that obtains each plot correspondence then in the substitution formula (1-1):

$P=[1-\frac{\mathrm{\Σa}\÷\frac{1}{2}n-\mathrm{\Σb}\÷\frac{1}{2}n}{\left(\frac{\mathrm{\Σa}+\mathrm{\Σb}}{n}\right)}]\×100\%---(1-1)$

Wherein, n is that the plot investigation is counted.

In addition, calendar year 2001 such as Zhong Qizhuan has also proposed cotton uniformity coefficient index.Defined according to the growing of cotton, formalness feature and cultivation management measure comprise the regularity of emerging, stay the seedling uniformity coefficient, a series of evaluation indexes such as the regularity of buddingging, the regularity of blooming, blow-of-cottons regularity and plant height uniformity coefficient, and proposed to improve the technical measures of uniformity coefficient at these indexs.

Yang Lihua etc. proposed the definition of plant field distribution consistency degree in 2006, plant is defined as plant field distribution consistency degree in the degree of the approximate honeycomb of field layout, and has provided the computing formula (1-2) of plant field distribution consistency degree:

$\mathrm{UD}=\frac{1}{(1+\mathrm{\Σ\η}/m)N}---(1-2)$

Wherein, UD is a plant field distribution consistency degree, 0＜UD≤1, and it is worth more near 1, and plant distributes even more; M counts for observation; N is average cave strain number; η (i=1,2 ... ..m) be each observation station from equal coefficient, can calculate by formula (1-3):

$\mathrm{\η}=\sqrt{\frac{\mathrm{\Σ}{(P_i-P)}^2}{12}}/p---(1-3)$

P in the formula _iBe 12 plant observed readings of a honeycomb observation station, p is a nominal linewidth.

In actual applications, obtaining of the monitoring of crop growth uniformity and index of correlation need be expended lot of manpower and material resources.Owing to need growing way difference, so there are shortcomings such as workload is big, automaticity is low, the update cycle is long to open-air on-site inspection crops.And when needs carried out the investigation of large-scale crop growth uniformity, it was measured difficulty and also will strengthen.Simultaneously, the difference of different farmland massif crop growing states is again ubiquitous, relies on the manual research data merely, can not satisfy the needs of modern agriculture management.

Summary of the invention

At defective that exists in the prior art and deficiency, the purpose of this invention is to provide a kind of monitoring device and method of crop growth uniformity, remote sensing via satellite obtains growth uniformity.

For achieving the above object, the present invention proposes a kind of monitoring device of crop growth uniformity, comprising:

Remote sensing image processing module, described remote sensing image processing module are carried out radiation correcting, atmosphere correction and geometric correction according to the remote sensing images that obtain to remote sensing images;

Plot vector data processing module, described plot vector data processing module is according to classifying to the crops in the described remote sensing images, to obtain the spatial distribution map of target crops; And the grid classification result in the sorted remote sensing images is converted into facet vector data; Then the ground block boundary of described spatial distribution map is revised;

Vegetation index processing module, described vegetation parameter processing module are calculated the vegetation index NDVI (NormalizedDifference Vegetation Index) in this plot according to the intramassif spectral signature in the described remote sensing images:

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Growth uniformity processing module, described growth uniformity processing module are calculated the growth uniformity index GUI (Growth Uniformity Index) in this plot according to vegetation index NDVI;

$\mathrm{GUI}=1-\frac{{\mathrm{NDVI}}_{\mathrm{CV}}}{({\mathrm{NDVI}}_{\mathrm{CV}\min }+{\mathrm{NDVI}}_{\mathrm{CV}\max })}$

Wherein:

NDVI _CVIt is the coefficient of variation of each pairing NDVI in plot;

NDVI _{CV min}For with the minimum value in all plot NDVI coefficient of variation mutually for the moment;

NDVI _{CV max}For with the maximal value in all plot NDVI coefficient of variation mutually for the moment.

Wherein, described plot vector data processing module comprises:

The spatial distribution map processing sub, described spatial distribution map processing sub is classified to the crops in the described remote sensing images, to obtain the spatial distribution map of target crops; And the classification of the grid of described spatial distribution map is converted into facet vector data;

The land use data processing sub; Described land use data processing sub is according to satellite remote sensing images, and the phase soil utilized thematic data when described remote sensing images were carried out reference that visual interpretation obtains data over the years;

Plot boundary treatment submodule, described plot boundary treatment submodule with described facet vector data with described with reference to the time after the soil utilizes thematic data stack mutually, cut the back by polar plot layer Intersect algorithm and extract the ground block boundary; And utilize the satellite remote sensing images in this year, and carry out ground block boundary correction by visual interpretation, obtain final crops plot data boundary.

Wherein, described vegetation index processing module comprises:

The spectral signature processing sub, described spectral signature processing sub is extracted the remotely-sensed data of described plot correspondence, and obtains crops not phase, different-waveband spectral signature information simultaneously according to remotely-sensed data;

The vegetation parameter processing sub, described vegetation parameter processing sub is carried out the wave band computing to spectral signature information, obtains vegetation parameter NDVI;

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

The wave band calculating sub module, described wave band calculating sub module is to not phase, different-waveband simultaneously, and spectral signature information is carried out the wave band computing, obtains the NDVI minimum value, maximal value, average, standard deviation, coefficient of variation NDVI _CV

Wherein, described device also comprises:

Crop spectral information module, described crop spectral information module be at each plot, extracts this corresponding all pixel NDVI values of plot institute, calculates this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CVWith winter wheat growth uniformity index GUI, and this information added to new field in the record of plot vector file correspondence, generate crop growing state spectral information knowledge base.

Wherein, described device also comprises:

Thematic map is made module, and described thematic map is made the crop growing state spectral information knowledge base that module generates according to described crop spectral information module, and the growth uniformity of the target crops in all plot is set up thematic map.

Simultaneously, the invention allows for a kind of monitoring method of crop growth uniformity, comprising:

Step 1, obtain satellite remote sensing images, and at described satellite remote sensing images carry out radiation correcting, atmosphere is corrected and geometric correction;

Step 2, the crops in the satellite remote sensing images are classified, to obtain the spatial distribution map of target crops;

Step 3, the grid classification result in the spatial distribution map is converted into facet vector data; And described spatial distribution map handled with the ground block boundary to crops revise;

Step 4, calculate the vegetation index NDVI in this plot according to the intramassif spectral signature in the described remote sensing images:

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Step 5, calculate growth uniformity index GUI according to the vegetation index NDVI in each plot;

$\mathrm{GUI}=1-\frac{{\mathrm{NDVI}}_{\mathrm{CV}}}{({\mathrm{NDVI}}_{\mathrm{CV}\min }+{\mathrm{NDVI}}_{\mathrm{CV}\max })}$

Wherein:

NDVI _CVIt is the coefficient of variation of each pairing NDVI in plot;

NDVI _{CV min}For with the minimum value in all plot NDVI coefficient of variation mutually for the moment;

NDVI _{CV max}For with the maximal value in all plot NDVI coefficient of variation mutually for the moment.

Wherein, described step 3 is specially:

Step 31, the classification of the grid of described spatial distribution map is converted into facet vector data;

Step 32, by the high resolution ratio satellite remote-sensing image, to remote sensing images over the years carry out visual interpretation obtain with reference to the time phase soil utilize thematic data;

Step 33, with after the stack of step 31 and step 32 gained data, by polar plot layer Intersect algorithm the two-layer vector data of step 31 and step 32 gained is cut the back and extracts the ground block boundary; And utilize the satellite remote sensing images in this year, and carry out ground block boundary correction by visual interpretation, obtain final crops plot data boundary.

Wherein, described step 4 is specially:

Step 41, extract the remotely-sensed data of plot correspondence, and obtain crops not phase, different-waveband spectral signature information simultaneously according to remotely-sensed data;

Step 42, spectral signature information is carried out the wave band computing, obtain vegetation parameter NDVI;

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Step 43, to not phase, different-waveband simultaneously, spectral signature information is carried out the wave band computing, obtains the NDVI minimum value, maximal value, average, standard deviation, coefficient of variation NDVI _CV

Wherein, described method also comprises:

Step 6, at each plot, extract corresponding all the pixel NDVI values of this plot institute, calculate this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CVWith winter wheat growth uniformity index GUI, and this information added to new field in the record of plot vector file correspondence, generate crop growing state spectral information knowledge base.

Wherein, described method also comprises:

Step 7, the crop growing state spectral information knowledge base that generates according to described crop spectral information module are set up thematic map to the growth uniformity of the target crops in all plot.

Technique scheme has following advantage: the present invention can obtain remote sensing images by remote sensing technology, and obtain the vector data in described plot by the ground block boundary, vector data according to this plot obtains vegetation parameter again, calculates crops growth uniformity degree index thereby calculate.The present invention can solve shortcomings such as workload is big in the prior art, automaticity is low, the update cycle is long, increases work efficiency.Integrated remote sensing of the present invention and GIS technology, utilize grid and vector data integrated technique to realize monitoring at the crops growth uniformity in natural plot, given full play to the advantage of remotely-sensed data in crop condition monitoring, crops growth uniformity evaluation index has been proposed, made up crop growth uniformity knowledge base based on remote sensing features information, and realized in real time, the remote sensing monitoring of crops growth uniformity fast and accurately, improved the precision of crop growth uniformity investigation.

Description of drawings

Fig. 1 is the structural representation of the monitoring device of the crop growth uniformity of the present invention's proposition;

Fig. 2 is the schematic flow sheet of the monitoring method of the crop growth uniformity of the present invention's proposition;

Fig. 3 is the spatial distribution map of the target crops of the extraction in specific embodiment of the present invention;

Fig. 4 is for being converted to this raster data among Fig. 3 the plot image behind the vector data;

Fig. 5 is the revised plot of Fig. 4 process image.

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used to illustrate the present invention, but are not used for limiting the scope of the invention.

Embodiment 1

First preferred embodiment of the invention has proposed a kind of monitoring device of crop growth uniformity, and its structure comprises as shown in Figure 1:

Remote sensing image processing module, described remote sensing image processing module are carried out radiation correcting, atmosphere correction and geometric correction according to the remote sensing images that obtain to remote sensing images;

Plot vector data processing module, described plot vector data processing module is according to classifying to the crops in the described remote sensing images, to obtain the spatial distribution map of target crops; And the grid classification result in the sorted remote sensing images is converted into facet vector data; Then the ground block boundary of described spatial distribution map is revised;

Vegetation index processing module, described vegetation parameter processing module are calculated the vegetation index NDVI in this plot according to the intramassif spectral signature in the described remote sensing images:

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Growth uniformity processing module, described growth uniformity processing module are calculated the growth uniformity index GUI in this plot according to vegetation index NDVI;

$\mathrm{GUI}=1-\frac{{\mathrm{NDVI}}_{\mathrm{CV}}}{({\mathrm{NDVI}}_{\mathrm{CV}\min }+{\mathrm{NDVI}}_{\mathrm{CV}\max })}$

Wherein:

NDVI _CVIt is the coefficient of variation of each pairing NDVI in plot;

NDVI _{CV min}For with the minimum value in all plot NDVI coefficient of variation mutually for the moment;

NDVI _{CV max}For with the maximal value in all plot NDVI coefficient of variation mutually for the moment.First preferred embodiment of the invention can be determined the plot of target crops by remote sensing images, and calculates the vegetation index NDVI and the growth uniformity index GUI in this plot by the intramassif spectral signature in the intramassif remote sensing images.First preferred embodiment of the invention utilizes grid and vector data integrated technique to realize monitoring at the crops growth uniformity in natural plot, has given full play to the advantage of remotely-sensed data in crop condition monitoring.The present invention realized in real time, the remote sensing monitoring of crops growth uniformity fast and accurately, improved the precision of crop growth uniformity investigation.

Embodiment 2

Second preferred embodiment of the invention is to improve on the first preferred embodiment basis, and promptly described plot vector data processing module comprises:

The spatial distribution map processing sub, described spatial distribution map processing sub is classified to the crops in the described remote sensing images, to obtain the spatial distribution map of target crops; And the classification of the grid of described spatial distribution map is converted into facet vector data;

The land use data processing sub; Described land use data processing sub is according to satellite remote sensing images, and the phase soil utilized thematic data when described remote sensing images were carried out reference that visual interpretation obtains data over the years;

Plot boundary treatment submodule, described plot boundary treatment submodule with described facet vector data with described with reference to the time after the soil utilizes thematic data stack mutually, cut the back by polar plot layer Intersect algorithm and extract the ground block boundary; And utilize the satellite remote sensing images in this year, and carry out ground block boundary correction by visual interpretation, obtain final crops plot data boundary.

In the second preferred embodiment of the invention, the phase soil utilizes superpose back cutting of thematic data to extract the ground block boundary during by reference, carries out more accurate correction with block boundary over the ground, improves the accuracy of Monitoring Data.

Wherein " with reference to the time phase soil utilize thematic data " be meant and utilize high-resolution remote sensing image over the years or the historical land use data of study area, with the farmland massif data of obtaining.Because historical data and up-to-date soil utilize the situation may be slightly different, thus utilize historical data in conjunction with current data over the ground block boundary revise, can improve the accuracy of plot data boundary.

Embodiment 3

Third preferred embodiment of the invention is to improve on the basis of first preferred embodiment and second preferred embodiment, and promptly described vegetation index processing module comprises:

The spectral signature processing sub, described spectral signature processing sub is extracted the remotely-sensed data of described plot correspondence, and obtains crops not phase, different-waveband spectral signature information simultaneously according to remotely-sensed data;

The vegetation parameter processing sub, described vegetation parameter processing sub is carried out the wave band computing to spectral signature information, obtains vegetation parameter NDVI;

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

The wave band calculating sub module, described wave band calculating sub module is to not phase, different-waveband simultaneously, and spectral signature information is carried out the wave band computing, obtains the NDVI minimum value, maximal value, average, standard deviation, coefficient of variation NDVI _CV

Wherein vegetation parameter NDVI is that vegetation index NDVI is the normalized ratio of red wave band of visible light and near-infrared band two wave bands, it can reflect the photosynthetic effective radiation absorbing state of vegetation on the one hand, can reflect crop growing state, leaf area index LAI etc. on the other hand, be the vegetation index that is most widely used at present.In the third preferred embodiment of the invention, use remote sensing images and accurate revised ground block boundary, obtain intramassif spectral signature information, and, can improve the accuracy of NDVI according to spectral signature information calculations vegetation parameter NDVI.

Embodiment 4

Four preferred embodiment of the invention is to improve on the basis of above-mentioned three preferred embodiments, and promptly described device also comprises:

Crop spectral information module, described crop spectral information module be at each plot, extracts this corresponding all pixel NDVI values of plot institute, calculates this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CVWith winter wheat growth uniformity index GUI, and this information added to new field in the record of plot vector file correspondence, generate crop growing state spectral information knowledge base.

Four preferred embodiment of the invention utilizes NDVI to set up crop growing state spectral information knowledge base,, the effectively monitoring of crop growing state spectral information long-term to improve.

Embodiment 5

Fifth preferred embodiment of the invention is to improve on the basis of above-mentioned four preferred embodiments, and promptly described device also comprises:

Thematic map is made module, and described thematic map is made the crop growing state spectral information knowledge base that module generates according to described crop spectral information module, and the growth uniformity of the target crops in all plot is set up thematic map.

Thematic map can obtain more intuitive image data according to crop growing state spectral information knowledge base.

Embodiment 6

The monitoring method of a kind of crop growth uniformity that the present invention proposes, its preferred embodiment flow process comprises as shown in Figure 2:

Step 1, obtain satellite remote sensing images, and at described satellite remote sensing images carry out radiation correcting, atmosphere is corrected and geometric correction;

Step 2, the crops in the satellite remote sensing images are classified, to obtain the spatial distribution map of target crops;

Step 3, described spatial distribution map handled with the ground block boundary to crops revise, obtain the vector data in plot;

Step 4, calculate the vegetation index NDVI in this plot according to the intramassif spectral signature in the described remote sensing images:

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Step 5, calculate growth uniformity degree index GUI according to the vegetation index NDVI in each plot;

$\mathrm{GUI}=1-\frac{{\mathrm{NDVI}}_{\mathrm{CV}}}{({\mathrm{NDVI}}_{\mathrm{CV}\min }+{\mathrm{NDVI}}_{\mathrm{CV}\max })}$

Wherein:

NDVI _CVIt is the coefficient of variation of each pairing NDVI in plot;

NDVI _{CV min}For with the minimum value in all plot NDVI coefficient of variation mutually for the moment;

NDVI _{CV max}For with the maximal value in all plot NDVI coefficient of variation mutually for the moment.

Sixth preferred embodiment of the invention can be determined the plot of target crops by remote sensing images, and calculates the vegetation index NDVI and the growth uniformity GUI in this plot by the intramassif spectral signature in the intramassif remote sensing images.Sixth preferred embodiment of the invention utilizes grid and vector data integrated technique to realize monitoring at the crops growth uniformity in natural plot, has given full play to the advantage of remotely-sensed data in crop condition monitoring.The present invention realized in real time, the remote sensing monitoring of crops growth uniformity fast and accurately, improved the precision of crop growth uniformity investigation.

Embodiment 7

Seventh preferred embodiment of the invention is to improve on the basis of above-mentioned the 6th preferred embodiment, and promptly described step 3 is specially:

Step 31, the classification of the grid of described spatial distribution map is converted into facet vector data;

Step 32, by the high resolution ratio satellite remote-sensing image, to remote sensing images over the years carry out visual interpretation obtain with reference to the time phase soil utilize thematic data;

Step 33, with after the stack of step 31 and step 32 gained data, by polar plot layer Intersect algorithm the two-layer vector data of step 31 and step 32 gained is cut the back and extracts the ground block boundary; And utilize the satellite remote sensing images in this year, and carry out ground block boundary correction by visual interpretation, obtain final crops plot data boundary.

In the seventh preferred embodiment of the invention, the phase soil utilizes superpose back cutting of thematic data to extract the ground block boundary during by reference, carries out more accurate correction with block boundary over the ground, improves the accuracy of Monitoring Data.

Embodiment 8

Eighth preferred embodiment of the invention be the above-mentioned the 6th or the basis of the 7th preferred embodiment on improve, promptly described step 4 is specially:

Step 41, extract the remotely-sensed data of plot correspondence, and obtain crops not phase, different-waveband spectral signature information simultaneously according to remotely-sensed data;

Step 42, spectral signature information is carried out the wave band computing, obtain vegetation parameter NDVI;

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Step 43, to not phase, different-waveband simultaneously, spectral signature information is carried out the wave band computing, obtains the NDVI minimum value, maximal value, average, standard deviation, coefficient of variation NDVI _CV

Wherein vegetation parameter NDVI is that vegetation index NDVI is the normalized ratio of red wave band of visible light and near-infrared band two wave bands, it can reflect the photosynthetic effective radiation absorbing state of vegetation on the one hand, can reflect crop growing state, leaf area index LAI etc. on the other hand, be the vegetation index that is most widely used at present.In the sixth preferred embodiment of the invention, use remote sensing images and accurate revised ground block boundary, obtain intramassif spectral signature information, and, can improve the accuracy of NDVI according to spectral signature information calculations vegetation parameter NDVI.

Embodiment 9

Eighth preferred embodiment of the invention be the above-mentioned the 6th or the basis of the 7th or the 8th preferred embodiment on improve, promptly described method also comprises:

Step 6, at each plot, extract corresponding all the pixel NDVI values of this plot institute, calculate this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CVWith winter wheat growth uniformity index GUI, and this information added to new field in the record of plot vector file correspondence, generate crop growing state spectral information knowledge base.

Four preferred embodiment of the invention utilizes NDVI to set up crop growing state spectral information knowledge base, long-term to improve, effective crop growing state spectral information.

Embodiment 10

Eighth preferred embodiment of the invention be the above-mentioned the 6th or the basis of the 7th or the 8th or the 9th preferred embodiment on improve, promptly described method also comprises:

Step 7, the crop growing state spectral information knowledge base that generates according to described crop spectral information module are set up thematic map to the growth uniformity of the target crops in all plot.

Thematic map can obtain more intuitive image data according to crop growing state spectral information knowledge base.

The present invention will be described below by a specific embodiment.

(1) remote sensing images obtain and handle

2008 years, obtain study area Landsat TM remote sensing images 3 scapes season altogether in winter wheat growth, to obtain the date to be respectively March 27, April 28 and May 30, corresponding respectively winter wheat is stood up phase, boot stage and winter wheat milk stage.In addition, also obtained study area India star ISP6 image one scape on July 12nd, 08.All Landsat images adopt the dark goal method under the 6S model supports to carry out the atmosphere correction, and the ISP6 image adopts ENVI software FLAASH module to carry out the atmosphere correction, has obtained the earth surface reflection rate of all images.The geometric correction of image adopts image image to be chosen the method for ground control point, every scape image is chosen and is surpassed 300 ground control points, in addition, the satellite differential GPS reference mark that is obtained during according to factual survey is revised entire image, and the precision of images of process geometric correction is controlled within the pixel.

(2) the target crops are extracted

Utilize 2008 March 27 year, April 28 and Landsat5 TM on the 30th winter wheat growth in May season remote sensing images, and harvested remote sensing images of winter wheat on July 12 in 2008, adopt the decision tree classification method, area, Tongzhou winter wheat growing area is extracted.Extract the result as shown in Figure 3.

Utilize ENVI software classification post-processing function that this raster data is converted to vector data, the result as shown in Figure 4.Wherein, ENVI (The Environment for Visualizing Images) is the remote sensing image processing software of American I TT Visual Information Solutions company.

(3) target crops plot vector data obtains

In ARCVIEW3.3 software, divide the cutting computing that superposes of type polar plot by the farmland, Beijing area in 2006 that obtains with high resolution ratio satellite remote-sensing image visual interpretation, obtained meticulousr farmland massif vector data boundary, winter wheat remote sensing images in 2008 information extraction simultaneously superposes.By visual interpretation, determine area, 2008 Beijing Tongzhou winter wheat planting site block boundary.Winter wheat ground blocks of data is through after revising, as shown in Figure 5.

Totally 1105 in winter wheat plantation in 2008 plot, area, Tongzhou, the total area is 17791 hectares (26.7 ten thousand mus), wherein, area has 526 less than the wheat plot of 10 hectares (150 mus), the wheat plot of 10-30 hectare (150-450 mu) has 394, the wheat plot of 30-70 hectare (450-1050 mu) has 130, has 2 greater than the wheat plot of 70 hectares (1050 mus).

(4) for each plot, utilize VB to extract corresponding all the pixel NDVI values of plot institute in conjunction with GIS secondary development control MO programming, calculate this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CV, and this information added to new field in the record of plot vector file correspondence, generate crop spectral information knowledge base.

(5) determining based on the crop growth uniformity index in plot

For the growing way situation to all plot is carried out integrated survey, this research is according to the coefficient of variation NDVI in all plot mutually for the moment _CVMade up the size of winter wheat growth uniformity index GUI (Growth Uniformity Index) according to this index, estimate the growing way of different plot winter wheat, shown in GUI is defined as follows, the value of GUI is between 0-1, the GUI value is big more, explanatorily the block length gesture is good more, and NDVI value height and growing way are even more.

$\mathrm{GUI}=1-\frac{{\mathrm{NDVI}}_{\mathrm{CV}}}{({\mathrm{NDVI}}_{\mathrm{CV}\min }+{\mathrm{NDVI}}_{\mathrm{CV}\max })}$

Wherein,

NDVI _CVIt is the coefficient of variation of each pairing NDVI in plot;

NDVI _{CV min}For with the minimum value in all plot NDVI coefficient of variation mutually for the moment;

NDVI _{CV max}For with the maximal value in all plot NDVI coefficient of variation mutually for the moment;

(6) generate crop spectral information knowledge base;

For each plot, utilize VB to extract corresponding all the pixel NDVI values of plot institute in conjunction with GIS secondary development control MO programming, calculate this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CVWith winter wheat growth uniformity index GUI, and this information added to new field in the record of plot vector file correspondence, generate crop spectral information knowledge base.

(7) generate thematic map;

Phase crop growth uniformity thematic map during according to interior different plot, zone difference.

The method that this example utilizes the present invention to propose, realized remote sensing monitoring based on the crops growth uniformity of farmland massif, the technical scheme that the present invention proposes has made full use of remote sensing image data can be repeatedly, instantaneous, the harmless characteristics of obtaining on a large scale " planar " object spectrum information, carry out the evaluation of crop growth uniformity at natural plot, having overcome the investigation of crop growth uniformity in the past wastes time and energy, inefficient shortcoming, increasing work efficiency, when alleviating working strength, effectively raise crop growth uniformity monitoring accuracy and precision on a large scale.

The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the technology of the present invention principle; can also make some improvement and modification, these improve and modification also should be considered as protection scope of the present invention.

Claims (10)

1. the monitoring device of a crop growth uniformity is characterized in that, comprising:

Remote sensing image processing module, described remote sensing image processing module are carried out radiation correcting, atmosphere correction and geometric correction according to the remote sensing images that obtain to remote sensing images;

Plot vector data processing module, described plot vector data processing module is according to classifying to the crops in the described remote sensing images, to obtain the spatial distribution map of target crops; And the grid classification result in the sorted remote sensing images is converted into facet vector data; Then the ground block boundary of described spatial distribution map is revised;

Vegetation index processing module, described vegetation parameter processing module are calculated the vegetation index NDVI in this plot according to the intramassif spectral signature in the described remote sensing images:

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Growth uniformity processing module, described growth uniformity processing module are calculated the growth uniformity index GUI in this plot according to vegetation index NDVI;

$\mathrm{GUI}=1-\frac{{\mathrm{NDVI}}_{\mathrm{CV}}}{({\mathrm{NDVI}}_{\mathrm{CV}\min }+{\mathrm{NDVI}}_{\mathrm{CV}\max })}$

Wherein:

NDVI _CVIt is the coefficient of variation of each pairing NDVI in plot;

NDVI _CVminFor with the minimum value in all plot NDVI coefficient of variation mutually for the moment;

NDVI _CVmaxFor with the maximal value in all plot NDVI coefficient of variation mutually for the moment.

2. the monitoring device of crop growth uniformity according to claim 1 is characterized in that, described plot vector data processing module comprises:

The spatial distribution map processing sub, described spatial distribution map processing sub is classified to the crops in the described remote sensing images, to obtain the spatial distribution map of target crops; And the classification of the grid of described spatial distribution map is converted into facet vector data;

The land use data processing sub; Described land use data processing sub is according to satellite remote sensing images, described remote sensing images carried out visual interpretation utilize thematic data in the phase soil when obtaining reference based on data over the years;

Plot boundary treatment submodule, described plot boundary treatment submodule with described facet vector data with described with reference to the time after the soil utilizes thematic data stack mutually, cut the back by polar plot layer Intersect algorithm and extract the ground block boundary; And utilize the satellite remote sensing images in this year, and carry out ground block boundary correction by visual interpretation, obtain final crops plot data boundary.

3. the monitoring device of crop growth uniformity according to claim 1 and 2 is characterized in that, described vegetation index processing module comprises:

The spectral signature processing sub, described spectral signature processing sub is extracted the remotely-sensed data of described plot correspondence, and obtains crops not phase, different-waveband spectral signature information simultaneously according to remotely-sensed data;

The vegetation parameter processing sub, described vegetation parameter processing sub is carried out the wave band computing to spectral signature information, obtains vegetation parameter NDVI;

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

The wave band calculating sub module, described wave band calculating sub module is to not phase, different-waveband simultaneously, and spectral signature information is carried out the wave band computing, obtains the NDVI minimum value of remote sensing images, maximal value, average, standard deviation, coefficient of variation NDVI _CV

4. the monitoring device of crop growth uniformity according to claim 1 is characterized in that, described device also comprises:

Crop spectral information module, described crop spectral information module be at each plot, extracts this corresponding all pixel NDVI values of plot institute, calculates this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CVWith winter wheat growth uniformity index GUI, and this information added to new field in the record of plot vector file correspondence, generate crop growing state spectral information knowledge base.

5. the monitoring device of crop growth uniformity according to claim 4 is characterized in that, described device also comprises:

Thematic map is made module, and described thematic map is made the crop growing state spectral information knowledge base that module generates according to described crop spectral information module, and the growth uniformity of the target crops in all plot is set up thematic map.

6. the monitoring method of a crop growth uniformity comprises:

Step 1, obtain satellite remote sensing images, and at described satellite remote sensing images carry out radiation correcting, atmosphere is corrected and geometric correction;

Step 2, the crops in the satellite remote sensing images are classified, to obtain the spatial distribution map of target crops;

Step 3, the grid classification result in the spatial distribution map is converted into facet vector data; And described spatial distribution map handled with the ground block boundary to crops revise;

Step 4, calculate the vegetation index NDVI in this plot according to the intramassif spectral signature in the described remote sensing images:

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Step 5, calculate growth uniformity index GUI according to the vegetation index NDVI in each plot,

$\mathrm{GUI}=1-\frac{{\mathrm{NDVI}}_{\mathrm{CV}}}{({\mathrm{NDVI}}_{\mathrm{CV}\min }+{\mathrm{NDVI}}_{\mathrm{CV}\max })}$

Wherein:

NDVI _CVIt is the coefficient of variation of each pairing NDVI in plot;

NDVI _CVminFor with the minimum value in all plot NDVI coefficient of variation mutually for the moment;

NDVI _CVmaxFor with the maximal value in all plot NDVI coefficient of variation mutually for the moment.

7. the monitoring method of a kind of crop growth uniformity according to claim 6 is characterized in that, described step 3 is specially:

Step 31, the classification of the grid of described spatial distribution map is converted into facet vector data;

Step 32, by the high resolution ratio satellite remote-sensing image, to remote sensing images over the years carry out visual interpretation obtain with reference to the time phase soil utilize thematic data;

Step 33, with after the stack of step 31 and step 32 gained data, by polar plot layer Intersect algorithm the two-layer vector data of step 31 and step 32 gained is cut the back and extracts the ground block boundary; And utilize the satellite remote sensing images in this year, and carry out ground block boundary correction by visual interpretation, obtain final crops plot data boundary.

8. according to the monitoring method of claim 6 or 7 described a kind of crop growth uniformities, it is characterized in that described step 4 is specially:

Step 41, extract the remotely-sensed data of plot correspondence, and obtain crops not phase, different-waveband spectral signature information simultaneously according to remotely-sensed data;

Step 42, spectral signature information is carried out the wave band computing, obtain vegetation parameter NDVI;

$\mathrm{NDVI}=\frac{(R_{\mathrm{nir}}-R_{\mathrm{red}})}{(R_{\mathrm{nir}}+R_{\mathrm{red}})}$

Wherein, R _NirThe reflectivity that refers to the near-infrared band of remote sensing images; R _RedThe reflectivity that refers to the red spectral band of remote sensing images;

Step 43, to not phase, different-waveband simultaneously, spectral signature information is carried out the wave band computing, obtains the NDVI minimum value, maximal value, average, standard deviation, coefficient of variation NDVI _CV

9. the monitoring method of a kind of crop growth uniformity according to claim 6 is characterized in that, described method also comprises:

Step 6, at each plot, extract corresponding all the pixel NDVI values of this plot institute, calculate this plot not phase NDVI minimum value, maximal value, average, standard deviation and coefficient of variation NDVI simultaneously _CVWith winter wheat growth uniformity index GUI, and this information added to new field in the record of plot vector file correspondence, generate crop growing state spectral information knowledge base.

10. the monitoring method of a kind of crop growth uniformity according to claim 9 is characterized in that, described method also comprises:

Step 7, the crop growing state spectral information knowledge base that generates according to described crop spectral information module are set up thematic map to the growth uniformity of the target crops in all plot.

Images