Scattered vegetable field remote sensing recognition method and device, electronic equipment and storage medium

Abstract

The invention discloses a sporadic vegetable field remote sensing recognition method and device, electronic equipment and a storage medium. The recognition method comprises the following steps: collecting remote sensing images of a land parcel in a to-be-detected area in a plurality of image periods; identifying based on a periodic change characteristic curve decision tree model to obtain a first identification result; analyzing decision tree model recognition based on the texture features to obtain a second recognition result; identifying based on the spectral feature decision tree model of the images with different resolution scales to obtain a third identification result; if the first recognition result, the second recognition result and the third recognition result are all met, the plot is recognized as a sporadic vegetable plot. According to the method, the sporadic vegetable field planting distribution pattern characteristics and the vegetable planting period change characteristics are fully utilized, the characteristics of target ground feature spectrum, texture, pattern and the like are combined, the sporadic vegetable field identification model is constructed, and therefore efficient and accurate identification of the sporadic vegetable field is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing monitoring of crop planting patterns, and in particular to a remote sensing identification method, device, electronic equipment and storage medium for scattered vegetable fields. Background Art

[0002] Scattered vegetable plots refer to agricultural lands scattered around residents' houses where various types of vegetables are planted. They are characterized by complex and diverse planting types, scattered and dispersed planting scales, and frequent changes in planting conditions. Scattered vegetable plots make full use of idle land in front of and behind houses and deeply tap the potential for vegetable planting, which is an important part of a beautiful vegetable garden. Therefore, efficient and accurate monitoring and identification of scattered vegetable plots will help management departments to understand the distribution status and area base of scattered vegetable plots, better coordinate the planning and construction of beautiful vegetable gardens, and are of great significance to the implementation of the "vegetable basket" project and the realization of stable production and supply of vegetables.

[0003] At present, remote sensing identification technology methods for single crop planting patterns are relatively mature, mainly including decision tree classification, maximum likelihood method, K-means method, multi-scale segmentation and other pixel-based and object-oriented methods. These methods use the spectrum, texture and other characteristics of crops to more accurately identify single crop planting patterns. However, there are currently few remote sensing identification methods for scattered vegetable plots. First, scattered vegetable plots are a new planting pattern that has been promoted and utilized in the construction of beautiful vegetable gardens in recent years, and there are few remote sensing identification methods in this regard; second, the types of vegetables planted in scattered vegetable plots are complex and diverse, the scale of vegetable planting is sporadic and fragmented, and vegetable planting changes frequently, which are difficult to identify simply using spectral texture and other features.

[0004] Based on the characteristics of scattered vegetable fields, such as complex and diverse planting types, scattered and dispersed planting scales, and frequent changes in planting conditions, the present invention proposes a technical method for constructing a remote sensing recognition model of scattered vegetable fields according to the distribution pattern and periodic change characteristics of vegetable planting in scattered vegetable fields by using texture analysis and pixel statistical analysis methods, thereby achieving efficient and accurate identification of scattered vegetable fields.

[0005] The disclosure of the above background technology content is only used to assist in understanding the inventive concept and technical solution of the present invention. It does not necessarily belong to the prior art of this patent application, nor does it necessarily provide technical guidance. In the absence of clear evidence that the above content has been disclosed before the filing date of this patent application, the above background technology should not be used to evaluate the novelty and creativity of the present application. Summary of the invention

[0006] The purpose of the present invention is to provide a remote sensing identification method for scattered vegetable fields, which makes full use of the characteristics of the planting distribution pattern of scattered vegetable fields and the periodic change characteristics of vegetable planting, combines the spectrum, texture, pattern and other characteristics of the target ground objects, and constructs a scattered vegetable plot identification model, thereby realizing efficient and accurate identification of scattered vegetable fields.

[0007] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0008] A remote sensing identification method for scattered vegetable fields comprises the following steps:

[0009] Collect remote sensing images of the area to be measured at multiple imaging periods;

[0010] The first recognition result is obtained based on the decision tree model recognition of the periodic variation characteristic curve: the NDVI of the plots in the area to be measured is calculated according to the remote sensing image, the NDVI periodic variation characteristic curve of the plots in the area to be measured is drawn, and the NDVI of the trough value of the characteristic curve is calculated. min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Compare the result with the corresponding preset first threshold value T1 and second threshold value T2 to obtain a first recognition result;

[0011] The second recognition result is obtained based on texture feature analysis decision tree model recognition: the remote sensing image is analyzed by using a texture analysis method to select the mean texture as a feature parameter, the mean texture of each plot pixel in the measured area is calculated, the pixel mean texture standard deviation σ is calculated based on the mean texture, and the pixel mean texture standard deviation σ is calculated and compared with a preset third threshold value T3 to obtain a second recognition result;

[0012] A third recognition result is obtained based on the decision tree model recognition of spectral characteristics of images of different resolution scales: the NDVI of the plots in the measured area on the images of different resolution scales is calculated according to the remote sensing image, the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions are calculated, and the comparison results of the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions with the corresponding preset fourth threshold value T4 and fifth threshold value T5 are calculated to obtain the third recognition result;

[0013] If the first recognition result, the second recognition result and the third recognition result are all satisfied, the plot of land is identified as the target plot of land.

[0014] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the NDVI is obtained by superposition operation;

[0015] The superposition operation is to calculate the near-infrared band reflectance value and the infrared band reflectance value of the remote sensing image according to the formula: NDVI = (NIR-R) / (NIR+R);

[0016] Among them, NIR is the reflection value of the near infrared band, and R is the reflection value of the infrared band.

[0017] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the NDVI of the calculated characteristic curve trough value min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Comparing the result with the corresponding preset first threshold value T1 and second threshold value T2 to obtain a first recognition result, further comprising:

[0018] The remote sensing image is a medium-resolution remote sensing image;

[0019] The first identification result needs to meet the following requirements: NDVI min ≥T1 and f NDVI ≤T2.

[0020] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the periodic variation characteristic curve decision tree model also includes calculating the peak NDVI of the characteristic curve max ;

[0021] Trough NDVI min , peak NDVI max and the fluctuation amplitude f NDVI and the fluctuation amplitude f NDVI Calculated by the following formula:

[0022] NDVI max = max{NDVI1, NDVI2…NDVI n}

[0023] NDVI min =min{NDVI1, NDVI2…NDVI n}

[0024]

[0025] Among them, n is the number of remote sensing images, NDVI n is the NDVI value of the nth image.

[0026] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the remote sensing image is analyzed by using a texture analysis method to select the mean texture as a feature parameter, and the mean texture of each plot pixel in the measured area is calculated, which also includes:

[0027] Obtain medium-resolution remote sensing images of the land in the area to be measured, select the near-infrared band, use the texture analysis method to perform analysis and calculation, and obtain the mean texture value as the feature parameter value.

[0028] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the standard deviation σ of the pixel feature parameter is calculated according to the mean texture, and the standard deviation σ of the pixel feature parameter is calculated and compared with a preset third threshold T3 to obtain a second recognition result, and further includes:

[0029] The second recognition result satisfies: the characteristic parameter standard deviation σ≥T3.

[0030] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the NDVI of the plots in the measured area on images of different resolution scales is calculated, and the NDVI standard deviations of different resolutions and the difference of NDVI standard deviations of different resolutions are calculated, and the method further includes:

[0031] The NDVI standard deviations of different resolutions are the medium-resolution image NDVI standard deviation σNDVI1, the high-resolution image NDVI standard deviation σNDVI2, and the difference between the two Δσ NDVI , Δσ NDVI =σNDVI2-σNDVI1.

[0032] Further, based on any one of the above-mentioned technical solutions or a combination of multiple technical solutions, the calculation of the comparison results of the NDVI standard deviations at different resolutions and the NDVI standard deviation differences at different resolutions with the corresponding preset fourth threshold value T4 and fifth threshold value T5 to obtain a third recognition result also includes:

[0033] The third recognition result satisfies: σNDVI2≥T4 and Δσ NDVI ≥T5.

[0034] Furthermore, based on any one of the technical solutions or the combination of multiple technical solutions mentioned above, the method also includes the steps of: pre-establishing a periodic variation characteristic curve decision tree model, a texture feature analysis decision tree model and a spectral feature decision tree model for images of different resolution scales, collecting scattered vegetable fields as sample points, analyzing and calculating the sample points, obtaining characteristic parameters representing the sample points, inputting the above models, and obtaining characteristic parameters T1, T2, T3, T4, and T5.

[0035] According to another aspect of the present invention, the present invention provides a remote sensing identification device for scattered vegetable fields, comprising the following modules:

[0036] A remote sensing image acquisition module is used to acquire remote sensing images of the plots in the area to be measured at multiple image periods;

[0037] The characteristic curve recognition module is used to calculate the NDVI of the plots in the area to be measured at each image period based on the remote sensing image, draw the NDVI periodic variation characteristic curve of the plots in the area to be measured, and calculate the NDVI of the trough value of the characteristic curve min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Compare the result with the corresponding preset first threshold and second threshold to obtain a first recognition result;

[0038] A texture feature analysis and recognition module is used to select a mean texture as a feature parameter for analyzing the remote sensing image using a texture analysis method, obtain feature parameters of each plot pixel in the measured area, calculate a standard deviation σ of the pixel feature parameters according to the feature parameters, calculate a comparison result of the standard deviation σ of the pixel feature parameters and a preset third threshold value, and obtain a second recognition result;

[0039] A spectral feature recognition module is used to calculate the NDVI of the plots in the measured area on images of different resolution scales according to the remote sensing image, calculate the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions, calculate the comparison results of the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions with the corresponding preset fourth threshold and fifth threshold, and obtain a third recognition result;

[0040] A target plot identification module is used to obtain a target plot according to the first identification result, the second identification result and the third identification result.

[0041] According to another aspect of the present invention, the present invention provides an electronic device, a memory for storing a computer program;

[0042] The processor is used to execute the steps of the remote sensing identification method of scattered vegetable fields as described in any of the above items when executing the computer program stored in the memory.

[0043] According to another aspect of the present invention, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the remote sensing identification method for scattered vegetable fields as described in any one of the above items are performed.

[0044] The beneficial effects brought by the technical solution provided by the present invention are as follows:

[0045] a. Based on the characteristics of the distribution pattern of scattered vegetable fields and the characteristics of the changes in the vegetable planting cycle, scattered vegetable fields can be identified efficiently and accurately, which helps the management department to understand the distribution status and area of ​​scattered vegetable fields, and plays an important role in further promoting and carrying out the construction of beautiful vegetable gardens and ensuring the stable production and supply of vegetables;

[0046] b. Combining the spectral, texture, and pattern characteristics of the target objects, a scattered vegetable plot identification model was constructed. The remote sensing identification method and workflow for scattered vegetable plots supplemented and enriched the current research results in this field, and provided reference experience and technical reserves for more precise research in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A schematic diagram of a remote sensing identification method for scattered vegetable fields provided as an exemplary embodiment of the present invention;

[0049] Figure 2 A graph showing periodic variation characteristics of spectra of different plots provided for an exemplary embodiment of the present invention;

[0050] Figure 3 A scatter plot of characteristic parameter values ​​of different plots provided for an exemplary embodiment of the present invention;

[0051] Figure 4 An NDVI scatter plot of scattered vegetable fields on images of different resolution scales provided by an exemplary embodiment of the present invention;

[0052] Figure 5 A scattered vegetable field identification result diagram provided for an exemplary embodiment of the present invention;

[0053] Figure 6 A schematic diagram of a remote sensing identification device for scattered vegetable fields is provided as an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0055] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0056] In one embodiment of the present invention, a remote sensing identification method for scattered vegetable fields is provided. Figure 1 , the identification method comprises the following steps:

[0057] Step 1: Collect remote sensing images of the area to be measured at multiple imaging periods;

[0058] Step 2: Obtain the first recognition result based on the decision tree model recognition of the periodic variation characteristic curve: Calculate the NDVI of the plots in the area to be measured in each image period according to the remote sensing image, and draw the NDVI periodic variation characteristic curve of the plots in the area to be measured, such as Figure 2 As shown, calculate the trough value NDVI of the characteristic curve min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Compare the result with the corresponding preset first threshold value T1 and second threshold value T2 to obtain a first recognition result;

[0059] Step 3: Obtain the second recognition result based on texture feature analysis decision tree model recognition: The remote sensing image is analyzed using the texture analysis method to select the mean texture as the feature parameter, and a scatter plot of the mean texture feature parameter values ​​of different plots is drawn, such as Figure 3 As shown, the horizontal axis represents the axial direction of the distribution, and the vertical axis represents the mean texture value, the mean texture of each plot pixel in the test area is calculated, the pixel mean texture standard deviation σ is calculated according to the mean texture, and the pixel mean texture standard deviation σ is calculated and compared with the preset third threshold value T3 to obtain a second recognition result;

[0060] Step 4: Obtain the third recognition result based on the decision tree model recognition of the spectral characteristics of images of different resolution scales: Calculate the NDVI of the plots in the measured area on images of different resolution scales based on the remote sensing images, and draw the following Figure 4The NDVI scatter plot shown in the figure has a horizontal axis representing the axial direction of the distribution and a vertical axis representing the NDVI value; calculating the NDVI standard deviations at different resolutions and the NDVI standard deviation differences at different resolutions, and calculating the comparison results of the NDVI standard deviations at different resolutions and the NDVI standard deviation differences at different resolutions with the corresponding preset fourth threshold value T4 and fifth threshold value T5 to obtain a third recognition result;

[0061] Step 5: If the first recognition result, the second recognition result, and the third recognition result are all satisfied, the plot of land is identified as the target plot of land.

[0062] The remote sensing identification method for scattered vegetable plots provided by the present invention makes full use of the characteristics of the planting distribution pattern of scattered vegetable plots and the characteristics of the vegetable planting cycle, combines the spectrum, texture, pattern and other characteristics of the target object, and constructs three scattered vegetable plot recognition models. By inputting the characteristic parameters into the above three models, efficient and accurate recognition of scattered vegetable plots can be achieved. Figure 5 The areas marked with closed contour lines represent the identified scattered vegetable plots.

[0063] It is worth noting that step 2, step 3, and step 4 can be performed simultaneously without any sequential relationship. The embodiment of the present invention is described based on the remote sensing images of the whole year as an example.

[0064] In remote sensing images, NDVI stands for Normalized Difference Vegetation Index, which is one of the important parameters that reflect the growth and nutritional information of crops.

[0065] In one embodiment of the present invention, a periodic variation characteristic curve decision tree model, a texture feature analysis decision tree model and a different resolution scale image spectral feature decision tree model are pre-established, and scattered vegetable fields are collected as sample points, and the sample points are analyzed and calculated to obtain characteristic parameters representing the sample points. The above models are input to obtain characteristic parameters T1, T2, T3, T4, and T5.

[0066] The characteristic parameter thresholds T1, T2, T3, T4, and T5 are the values ​​that best represent the unified characteristics of the sample points as the characteristic parameter thresholds; for example, the calculation process of the characteristic parameter threshold T1 is: first calculate the trough value NDVI of all sample points min , the lower edge value of all the above sample points is calculated through the box plot, and the lower edge value is used as the NDVI that can reflect the trough value of all sample points min , then set the current value to the model trough value NDVI min Threshold T1.

[0067] In one embodiment of the present invention, the NDVI is obtained by superposition operation;

[0068] The superposition operation is to calculate the near-infrared band reflectance value and the infrared band reflectance value of the remote sensing image according to the formula: NDVI = (NIR-R) / (NIR+R);

[0069] Among them, NIR is the reflection value of the near infrared band, and R is the reflection value of the infrared band.

[0070] In one embodiment of the present invention, the NDVI of the characteristic curve is calculated. min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Comparing the result with the corresponding preset first threshold value T1 and second threshold value T2 to obtain a first recognition result, further comprising:

[0071] The remote sensing image is a medium-resolution remote sensing image;

[0072] The first identification result needs to meet the following requirements: NDVI min ≥T1 and f NDVI ≤T2.

[0073] In one embodiment of the present invention, the periodic variation characteristic curve decision tree model further includes calculating the peak NDVI of the characteristic curve. max ; Trough value NDVI min , peak NDVI max and the fluctuation amplitude f NDVI and the fluctuation amplitude f NDVI Calculated by the following formula:

[0074] NDVI max = max{NDVI1, NDVI2…NDVI n}

[0075] NDVI min =min{NDVI1, NDVI2…NDVI n}

[0076]

[0077] Among them, n is the number of remote sensing images, NDVI n is the NDVI value of the nth image.

[0078] In one embodiment of the present invention, the remote sensing image of the area to be measured is collected, and the mean texture is selected as the feature parameter for analysis using a texture analysis method, which also includes:

[0079] The medium-resolution remote sensing images of the area to be measured are collected, the near-infrared band is selected, and the texture analysis method is used for analysis and calculation to obtain the mean texture value as the characteristic parameter value.

[0080] The characteristic parameters of each plot pixel in the test area are obtained by an extraction tool. The extraction tool can be exemplarily selected from the "ExtractValuesToPoints" tool in the "Spatial Analyst" tool in the ArcGIS software, which can extract the value of the image pixel corresponding to the sample point.

[0081] In one embodiment of the present invention, the pixel mean texture standard deviation σ is calculated according to the characteristic parameter, and the pixel mean texture standard deviation σ is compared with a preset third threshold value T3 to obtain a second recognition result, further comprising:

[0082] The second recognition result satisfies: the mean texture σ≥T3.

[0083] In one embodiment of the present invention, the step of calculating the NDVI of the plots in the area to be measured on images of different resolution scales and calculating the standard deviation of the NDVI of different resolutions further includes:

[0084] The NDVI standard deviations of different resolutions are the medium-resolution image NDVI standard deviation σNDVI1, the high-resolution image NDVI standard deviation σNDVI2, and the difference between the two Δσ NDVI .

[0085] In one embodiment of the present invention, the third recognition result is obtained by comparing the NDVI standard deviation of different resolutions with the corresponding medium resolution image NDVI standard deviation threshold T4 and the medium resolution and high resolution image NDVI standard deviation difference threshold T5, further comprising:

[0086] The third recognition result satisfies: σNDVI2≥T4 and Δσ NDVI ≥T5;

[0087] Among them, Δσ NDVI The calculation formula is:

[0088] Δσ NDVI =σNDVI2-σNDVI1.

[0089] In one embodiment of the present invention, after calculating the NDVI of the plots in the area to be measured on images of different resolution scales, the method further includes using an extraction tool to obtain the NDVI of pixels in different plots in the area to be measured on images of different resolution scales. The extraction tool can exemplarily be selected from the "ExtractValuesToPoints" in the "Spatial Analyst" tool in ArcGIS software.

[0090] In one embodiment of the present invention, a remote sensing identification device for scattered vegetable plots is provided. Figure 6 As shown, the device includes the following modules:

[0091] A remote sensing image acquisition module is used to acquire remote sensing images of the plots in the area to be measured at multiple image periods;

[0092] The characteristic curve recognition module is used to calculate the NDVI of the plots in the area to be measured at each image period based on the remote sensing image, draw the NDVI periodic variation characteristic curve of the plots in the area to be measured, and calculate the NDVI of the trough value of the characteristic curve min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Compare the result with the corresponding preset first threshold and second threshold to obtain a first recognition result;

[0093] A texture feature analysis and recognition module is used to select a mean texture as a feature parameter for analyzing the remote sensing image using a texture analysis method, obtain feature parameters of each plot pixel in the measured area, calculate a standard deviation σ of the pixel feature parameters according to the feature parameters, calculate a comparison result of the standard deviation σ of the pixel feature parameters and a preset third threshold value, and obtain a second recognition result;

[0094] The spectral feature recognition module is used to calculate the NDVI of the plots in the measured area on images of different resolution scales based on the remote sensing images, calculate the NDVI standard deviation of different resolutions and the difference of NDVI standard deviations of different resolutions, calculate the comparison results of the NDVI standard deviation of different resolutions and the difference of NDVI standard deviations of different resolutions with the corresponding preset fourth threshold and fifth threshold, and obtain a third recognition result.

[0095] A target plot identification module is used to obtain a target plot according to the first identification result, the second identification result and the third identification result.

[0096] In one embodiment of the present invention, an electronic device is provided, including:

[0097] Memory, used to store computer programs;

[0098] The processor is used to execute the steps of the remote sensing identification method of scattered vegetable fields as described in any of the above items when executing the computer program stored in the memory.

[0099] In one embodiment of the present invention, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of the remote sensing identification method for scattered vegetable fields as described in any one of the above items are performed.

[0100] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0101] The above is only a specific implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A remote sensing identification method for scattered vegetable fields, characterized in that: The following steps are involved: Collect remote sensing images of the area to be measured at multiple imaging periods; The first recognition result is obtained based on the decision tree model recognition of the periodic variation characteristic curve: the NDVI of the plots in the area to be measured is calculated according to the remote sensing image, the NDVI periodic variation characteristic curve of the plots in the area to be measured is drawn, and the NDVI of the trough value of the characteristic curve is calculated. min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Compare the result with the corresponding preset first threshold value T1 and second threshold value T2 to obtain a first recognition result; The second recognition result is obtained based on texture feature analysis decision tree model recognition: the remote sensing image is analyzed by using a texture analysis method to select the mean texture as a feature parameter, the mean texture of each plot pixel in the measured area is calculated, the pixel mean texture standard deviation σ is calculated based on the mean texture, and the pixel mean texture standard deviation σ is calculated and compared with a preset third threshold value T3 to obtain a second recognition result; A third recognition result is obtained based on the decision tree model recognition of spectral characteristics of images of different resolution scales: the NDVI of the plots in the measured area on the images of different resolution scales is calculated according to the remote sensing image, the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions are calculated, and the comparison results of the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions with the corresponding preset fourth threshold value T4 and fifth threshold value T5 are calculated to obtain the third recognition result; If the first recognition result, the second recognition result and the third recognition result are all satisfied, the plot of land is identified as the target plot of land.

2. The remote sensing identification method for scattered vegetable fields according to claim 1 is characterized in that: The NDVI is obtained by superposition operation; The superposition operation is to calculate the near-infrared band reflectance value and the infrared band reflectance value of the remote sensing image according to the formula: NDVI = (NIR-R) / (NIR+R); Among them, NIR is the reflection value of the near infrared band, and R is the reflection value of the infrared band.

3. The remote sensing identification method for scattered vegetable fields according to claim 1 or 2, characterized in that: The calculation characteristic curve trough value NDVI min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Comparing the result with the corresponding preset first threshold value T1 and second threshold value T2 to obtain a first recognition result, further comprising: The remote sensing image is a medium-resolution remote sensing image; The first identification result needs to meet the following requirements: NDVI min ≥T1 and f NDVI ≤T2.

4. The remote sensing identification method for scattered vegetable fields according to claim 3 is characterized in that: The periodic variation characteristic curve decision tree model also includes calculating the peak NDVI of the characteristic curve max ; Trough NDVI min , peak NDVI max and the fluctuation amplitude f NDVI and the fluctuation amplitude f NDVI Calculated by the following formula: NDVI max =max{NDVI1,NDVI2…NDVI n } NDVI min =min{NDVI1, NDVI2…NDVI n } Among them, n is the number of remote sensing images, NDVI n is the NDVI value of the nth image.

5. The remote sensing identification method for scattered vegetable fields according to claim 1 is characterized in that: The remote sensing image is analyzed by using a texture analysis method to select a mean texture as a feature parameter, and the mean texture of each plot pixel in the area to be measured is calculated, which also includes: Obtain medium-resolution remote sensing images of the land in the area to be measured, select the near-infrared band, use the texture analysis method to perform analysis and calculation, and obtain the mean texture value.

6. The remote sensing identification method for scattered vegetable fields according to claim 1 or 5, characterized in that: Calculating the NDVI standard deviations at different resolutions and the difference between the NDVI standard deviations at different resolutions, calculating the comparison results between the NDVI standard deviations at different resolutions and the difference between the NDVI standard deviations at different resolutions and the corresponding preset fourth threshold value T4 and fifth threshold value T5, and obtaining a third recognition result, further comprising: The second recognition result satisfies: the mean texture σ≥T3.

7. The remote sensing identification method for scattered vegetable fields according to claim 1 is characterized in that: Calculating the NDVI of the plots in the measured area on images of different resolution scales, calculating the NDVI standard deviations of different resolutions and the difference of the NDVI standard deviations of different resolutions, further comprising: The NDVI standard deviations of different resolutions are the medium-resolution image NDVI standard deviation σNDVI1, the high-resolution image NDVI standard deviation σNDVI2, and the difference between the two Δσ NDVI , Δσ NDVI =σNDVI2-σNDVI1.

8. The remote sensing identification method for scattered vegetable fields according to claim 7 is characterized in that: The calculating of the comparison results of the NDVI standard deviations at different resolutions and the NDVI standard deviation differences at different resolutions with the corresponding preset fourth threshold value T4 and fifth threshold value T5 to obtain a third recognition result also includes: The third recognition result satisfies: σNDVI2≥T4 and Δσ NDVI ≥T5.

9. The remote sensing identification method for scattered vegetable fields according to claim 1, characterized in that: The method also includes the steps of pre-establishing a periodic variation characteristic curve decision tree model, a texture feature analysis decision tree model and a spectral feature decision tree model for images of different resolution scales, collecting scattered vegetable fields as sample points, analyzing and calculating the sample points, obtaining characteristic parameters representing the sample points, inputting the above models, and obtaining characteristic parameters T1, T2, T3, T4 and T5.

10. A remote sensing identification device for scattered vegetable fields, characterized in that: Includes the following modules: A remote sensing image acquisition module is used to acquire remote sensing images of the plots in the area to be measured at multiple image periods; The characteristic curve recognition module is used to calculate the NDVI of the plots in the area to be measured at each image period based on the remote sensing image, draw the NDVI periodic variation characteristic curve of the plots in the area to be measured, and calculate the NDVI of the trough value of the characteristic curve min and the fluctuation amplitude f NDVI , calculate the trough value NDVI min and the fluctuation amplitude f NDVI Compare the result with the corresponding preset first threshold and second threshold to obtain a first recognition result; A texture feature analysis and recognition module is used to select a mean texture as a feature parameter for analyzing the remote sensing image using a texture analysis method, obtain feature parameters of each plot pixel in the measured area, calculate a standard deviation σ of the pixel feature parameters according to the feature parameters, calculate a comparison result of the standard deviation σ of the pixel feature parameters and a preset third threshold value, and obtain a second recognition result; A spectral feature recognition module is used to calculate the NDVI of the plots in the measured area on images of different resolution scales according to the remote sensing image, calculate the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions, calculate the comparison results of the NDVI standard deviation of different resolutions and the difference of the NDVI standard deviation of different resolutions with the corresponding preset fourth threshold and fifth threshold, and obtain a third recognition result; A target plot identification module is used to obtain a target plot according to the first identification result, the second identification result and the third identification result.

11. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the steps of the method according to any one of claims 1 to 9 when executing the computer program stored in the memory.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are performed.

Images