Sensitive wave band-based method and device for identifying potatoes in tillering period, and electronic equipment

Abstract

The invention provides a sensitive wave band-based potato identification method and device in a tillering period and electronic equipment, and relates to the technical field of identification, and the method comprises the steps: obtaining a remote sensing image including potatoes in the tillering period of a target region; determining a plurality of random points based on surface reflectance images of a plurality of target wavebands acquired by the remote sensing image; determining a correlation coefficient of the target wave band based on the surface reflectance value of each random point under the target wave band and the total surface reflectance value of each random point under all the target wave bands; based on all the correlation coefficients, determining a sensitive wave band for the potatoes in the tillering period; and identifying a planting area of the potatoes in the tillering period in the target area based on the sensitive wave band. According to the method, the defect that a sample needs to be labeled every time or the sample labeling amount is large is overcome, meanwhile, a tillering stage image with remarkable potato features is taken as a main part, the vegetation features of the potatoes are enhanced, the problems of mixed separation and missing separation of the potatoes and other crops are solved, and the identification accuracy of the potato planting area is improved.

Description

Technical Field

[0001] The present invention relates to the field of identification technology, and in particular to a method, device and electronic equipment for identifying potatoes in the growing period based on a sensitive band. Background Art

[0002] In recent years, potato planting has been widely distributed. Monitoring potato planting areas is the main technical means to quickly obtain the distribution range and area of ​​potato planting. Accurate potato planting distribution range and area are the most basic basis for planting enterprises to make decisions.

[0003] In the related art, a potato recognition model based on a deep learning algorithm is usually used to identify potato planting areas.

[0004] However, in the above-mentioned related technologies, a large number of representative samples are required when training the potato remote sensing recognition model, and the sample annotation is difficult. In addition, the planting structure of the main potato producing areas is relatively complex, with many types of crops. There are obvious phenomena of same objects with different spectra and different objects with the same spectra, which will cause potato confusion and omission, thereby reducing the accuracy of potato planting area identification. Summary of the invention

[0005] The invention provides a method, device and electronic equipment for identifying potatoes in the growing period based on a sensitive band, so as to solve the defect of reducing the accuracy of identifying potato planting areas in the prior art.

[0006] The invention provides a method for identifying potatoes in the growing period based on sensitive bands, comprising the following steps.

[0007] Acquire remote sensing images of target areas including potatoes in the growing period; Acquire surface reflectance images of multiple target bands based on the remote sensing image; Based on the surface reflectance images of the multiple target bands, determining multiple random points representing potatoes at the growing stage; For each of the target bands, determining a correlation coefficient of the target band based on a ground surface reflectance value of each of the random points in the target band and a total ground surface reflectance value of each of the random points in all of the target bands; Based on all the correlation coefficients, determining the sensitive band for potatoes in the growing stage from all the target bands; The planting area of ​​potatoes in the growth stage in the target area is identified based on the sensitive band.

[0008] According to a method for identifying potatoes in the growing period based on sensitive bands provided by the present invention, the sensitive bands for potatoes in the growing period are determined from all the target bands based on all the correlation coefficients, including: Compare each of the correlation coefficients with a preset correlation coefficient to obtain a first comparison result; The target band corresponding to the target correlation coefficient greater than the preset correlation coefficient in the first comparison result is determined as the sensitive band, and the sensitive band includes the 8A band, the 8th band, the 7th band and the 6th band.

[0009] According to a method for identifying potatoes in the growing period based on sensitive bands provided by the present invention, the method for identifying the planting area of ​​potatoes in the growing period in the target area based on the sensitive bands includes: For each of the sensitive bands, normalizing the surface reflectance value of each pixel in the sensitive band to obtain the normalized surface reflectance value of each pixel; For each of the pixels, based on the normalized surface reflectance value of the pixel in each of the sensitive bands, determine the potato spectral index value corresponding to the pixel; Determining a potato spectral index map based on the potato spectral index value corresponding to each of the pixels; Based on the potato spectral index map, the planting area of ​​the potatoes in the growth stage is determined.

[0010] According to a method for identifying potatoes in the growing period based on sensitive bands provided by the present invention, the method of determining the planting area of ​​potatoes in the growing period based on the potato spectral index map comprises: Comparing the potato spectral index value of each pixel in the potato spectral index map with a preset spectral index value to obtain a second comparison result; Determine the pixel corresponding to the potato spectral index value greater than the preset spectral index value in the second comparison result as the target pixel; The area where all the target pixels are located is determined as the planting area for potatoes in the growth stage.

[0011] According to a method for identifying potatoes in the growing period based on sensitive bands provided by the present invention, the potato spectral index value corresponding to the pixel is determined based on the normalized surface reflectance value of the pixel under each of the sensitive bands, including: Determine the square value of the sum of the normalized surface reflectance values ​​of the pixel in the 8A band, the 8th band, the 7th band, and the 6th band; The sum of the square value and the normalized surface reflectance value of the pixel in the 8A band is determined as the potato spectral index value corresponding to the pixel, and the 8A band is the most sensitive band for potatoes in the growing stage.

[0012] According to a method for identifying potatoes in the growing period based on sensitive bands provided by the present invention, the remote sensing images include surface reflectance images of bands 2, 3, 4 and 8 of a first spatial resolution, and surface reflectance images of bands 5, 6, 7, 8A, 11 and 12 of a second spatial resolution, wherein the first spatial resolution is greater than the second spatial resolution; The step of acquiring surface reflectance images of multiple target bands based on the remote sensing image comprises: Resampling the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the second spatial resolution to obtain the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution; The surface reflectance images of the 2nd band, the 3rd band, the 4th band and the 8th band of the first spatial resolution and the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution are superimposed to obtain the surface reflectance images of the multiple target bands.

[0013] According to a method for identifying potatoes in the growing period based on sensitive bands provided by the present invention, the correlation coefficient of the target band is determined based on the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all the target bands, including: Constructing a two-dimensional scatter plot based on the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all the target bands; Based on the two-dimensional scatter plot, determining a linear fitting function; Based on the linear fit function, the correlation coefficient is determined.

[0014] The present invention also provides a device for identifying potatoes in the growing period based on a sensitive band, comprising: A first acquisition unit is used to acquire a remote sensing image of a target area including potatoes in the growing period; A second acquisition unit, used for acquiring surface reflectance images of multiple target bands based on the remote sensing image; A first determination unit is used to determine a plurality of random points representing potatoes at a growing stage based on the surface reflectance images of the plurality of target bands; A second determination unit is used to determine, for each of the target bands, a correlation coefficient of the target band based on a surface reflectance value of each of the random points in the target band and a total surface reflectance value of each of the random points in all of the target bands; A third determining unit is used to determine a sensitive band for potatoes in the growing stage from all the target bands based on all the correlation coefficients; An identification unit is used to identify a planting area of ​​potatoes in the growth stage within the target area based on the sensitive band.

[0015] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for identifying potatoes in the growing period based on sensitive bands as described above is implemented.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for identifying potatoes in the growing period based on sensitive bands as described above is implemented.

[0017] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for identifying potatoes in the growing period based on sensitive bands.

[0018] The present invention provides a method, device and electronic device for identifying potatoes in the growing period based on sensitive bands. The method, device and electronic device obtain surface reflectance images of multiple target bands based on remote sensing images of potatoes in the target area. Based on the surface reflectance images of the multiple target bands, multiple random points representing potatoes in the growing period are determined. For each target band, the correlation coefficient of the target band is determined based on the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all target bands. Based on all the correlation coefficients, the sensitive band for potatoes in the growing period is determined from all the target bands, and then the planting area of ​​potatoes in the growing period in the target area is identified based on the sensitive bands. It can be seen that the present invention can determine the sensitive bands for potatoes in the growing stage from all target bands based on the correlation coefficients of each target band, and then identify the planting area of ​​potatoes in the growing stage in the target area based on the sensitive bands, overcoming the defects of the related art that samples need to be labeled or the sample labeling amount is large each time. At the same time, the present invention mainly uses the images of the growing stage with significant potato features to enhance the potato vegetation features, overcomes the problem of mixed classification and missed classification of potatoes and other crops, and thus improves the accuracy of potato planting area identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 This is one of the flow charts of the method for identifying potatoes in the growing period based on sensitive bands provided by an embodiment of the present invention.

[0021] Figure 2 It is a two-dimensional scatter plot of the second band provided by an embodiment of the present invention.

[0022] Figure 3 It is a two-dimensional scatter plot of the third band provided by an embodiment of the present invention.

[0023] Figure 4 It is a two-dimensional scatter plot of the 4th band provided by an embodiment of the present invention.

[0024] Figure 5 It is a two-dimensional scatter plot of the 5th band provided by an embodiment of the present invention.

[0025] Figure 6 It is a two-dimensional scatter plot of the 6th band provided by an embodiment of the present invention.

[0026] Figure 7 It is a two-dimensional scatter plot of the 7th band provided by an embodiment of the present invention.

[0027] Figure 8 It is a two-dimensional scatter plot of the 8th band provided by an embodiment of the present invention.

[0028] Fig. 9 It is a two-dimensional scatter plot of the 8A band provided by an embodiment of the present invention.

[0029] Fig.10 It is a two-dimensional scatter plot of the 11th band provided by an embodiment of the present invention.

[0030] Fig.11 It is a two-dimensional scatter plot of the 12th band provided by an embodiment of the present invention.

[0031] Fig.12 This is the second flow chart of the method for identifying potatoes in the growing period based on sensitive bands provided by an embodiment of the present invention.

[0032] Fig.13 It is an overall schematic diagram of a method for identifying potatoes in the growing period based on sensitive bands provided by an embodiment of the present invention.

[0033] Fig.14It is a schematic diagram of the structure of a potato identification device in the growing period based on a sensitive band provided by an embodiment of the present invention.

[0034] Fig.15 It is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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 are within the scope of protection of the present invention.

[0036] At present, monitoring of potato planting areas through remote sensing technology is generally carried out by supervised classification. It is necessary to select a certain number of potato samples and select classifiers of machine learning algorithms such as maximum likelihood, Mahalanobis distance, neural network (Back Propagation, BP), support vector machine (SVM), random forest, etc. for crop classification, or use deep learning algorithms to train potato recognition models, and obtain information such as potato planting areas and planting areas through potato recognition models. However, when using a potato recognition model using supervised classification, a large number of representative samples are required, and the sample annotation is difficult. In addition, the planting structure of the main potato producing areas is relatively complex, with many types of crops, and there are obvious phenomena of different spectra for the same object and the same spectra for different objects, which will cause potato mixed classification and missed classification, thereby reducing the accuracy of potato planting area identification.

[0037] Based on this, the present invention provides a method for identifying potatoes in the growing stage based on sensitive bands, which can determine the sensitive bands for potatoes in the growing stage from all target bands based on the correlation coefficients of the target bands, and then identify the planting areas of potatoes in the growing stage in the target area based on the sensitive bands, overcoming the defects of the related technology that samples need to be labeled or the sample labeling amount is large each time. At the same time, the present invention mainly uses images of the growing stage with significant potato features to enhance potato vegetation features, overcomes the problem of mixed classification and missed classification of potatoes and other crops, and thus improves the accuracy of potato planting area recognition.

[0038] Combine the following Figure 1-Figure 13The present invention describes a method for identifying potatoes in the growing period based on sensitive bands. The execution subject of the method for identifying potatoes in the growing period based on sensitive bands can be an electronic device such as a terminal, a tablet computer, a computer, or a device for identifying potatoes in the growing period based on sensitive bands set in the electronic device. The device for identifying potatoes in the growing period based on sensitive bands can be implemented by software, hardware, or a combination of the two.

[0039] Figure 1 FIG. 1 is one of the flow charts of a method for identifying potatoes in the growing period based on sensitive bands provided by an embodiment of the present invention. Figure 1 As shown, the method for identifying potatoes in the growing period based on sensitive bands includes the following steps: Step 101: Acquire a remote sensing image of a target area including potatoes in the growing period.

[0040] The target area may be any area where potato planting areas need to be counted.

[0041] For example, taking XX city in XX province as the target area, XX city in XX province is a typical potato planting area, and the potato planting period in spring in XX city is in XX month. The images acquired this time take the remote sensing images of the target area in XX city as an example, totaling 4 images. By combining these 4 images, we can obtain the remote sensing images of the target area including potatoes in the planting period.

[0042] It should be noted that the remote sensing image of the entire target area including potatoes in the growing stage can also be acquired at one time, and the present invention is not limited to this.

[0043] Step 102: Acquire surface reflectance images of multiple target bands based on the remote sensing image.

[0044] For example, when a remote sensing image is obtained, the remote sensing image is subjected to radiometric calibration, atmospheric correction, geometric registration, orthorectification and other preprocessing through the atmospheric correction model (Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes, FLAASH) to obtain surface reflectance images of multiple target bands, where the target bands include band 2, band 3, band 4, band 8, band 5, band 6, band 7, band 8A, band 11 and band 12, a total of 10 bands. Table 1 is a schematic diagram of the description of each band, excluding band 1 (coast or aerosol band), band 9 (water vapor band), and band 10 (60-meter short-wave infrared band), that is, a combination of visible light, near infrared, and short-wave infrared images is used for analysis.

[0045] Table 1

[0046] Step 103: Based on the surface reflectance images of the multiple target bands, determine multiple random points representing potatoes in the growing period.

[0047] For example, when surface reflectance images of multiple target bands are obtained, visual interpretation is performed based on the surface reflectance images of multiple target bands, and multiple (for example, 300) random points with the ground feature category of potatoes in the growing period are randomly obtained, and band sensitivity analysis is performed using the multiple random points representing potatoes in the growing period.

[0048] Step 104 : for each of the target bands, determine the correlation coefficient of the target band based on the surface reflectance value of each of the random points in the target band and the total surface reflectance value of each of the random points in all of the target bands.

[0049] Among them, the surface reflectance value usually refers to the reflectivity of surface objects to solar radiation. It is a physical quantity used to describe the ratio of solar radiation energy reflected by the surface of an object to the incident solar radiation energy. In the field of remote sensing, the surface reflectance value is often used to evaluate the spectral characteristics of surface objects and is an important basis for remote sensing image interpretation and classification.

[0050] Optionally, a two-dimensional scatter plot is constructed based on the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all the target bands; a linear fitting function is determined based on the two-dimensional scatter plot; and the correlation coefficient is determined based on the linear fitting function.

[0051] For example, for each target band, the surface reflectance value of each random point in the target band is extracted from the surface reflectance images of multiple target bands, and the total surface reflectance value of each random point in all target bands is determined. A two-dimensional scatter plot of the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all target bands is established through the matplotlib library. The surface reflectance value of each random point in the target band is used as the abscissa of the two-dimensional scatter plot, and the total surface reflectance value of each random point in all target bands is used as the ordinate of the two-dimensional scatter plot. Finally, the two-dimensional scatter plots corresponding to the 10 target bands are obtained. Figure 2 is a two-dimensional scatter plot of the second band provided by an embodiment of the present invention, Figure 3 is a two-dimensional scatter plot of the third band provided by an embodiment of the present invention, Figure 4 is a two-dimensional scatter plot of the fourth band provided by an embodiment of the present invention, Figure 5 is a two-dimensional scatter plot of the fifth band provided by an embodiment of the present invention, Figure 6 is a two-dimensional scatter plot of the sixth band provided by an embodiment of the present invention, Figure 7is a two-dimensional scatter plot of the seventh band provided by an embodiment of the present invention, Figure 8 is a two-dimensional scatter plot of the 8th band provided by an embodiment of the present invention, Fig. 9 is a two-dimensional scatter plot of the 8A band provided by an embodiment of the present invention, Fig.10 is a two-dimensional scatter plot of the 11th band provided by an embodiment of the present invention, Fig.11 is a two-dimensional scatter plot of the 12th band provided by an embodiment of the present invention. The linregress function of scipy is used to calculate the parameters (slope and intercept) of the linear fitting, and then the linear fitting function is obtained, for example, Figure 2 The linear fitting function of the two-dimensional scatter plot of the second band shown is , and then use the scipy linregress function to get the correlation coefficient R 2 .

[0052] Finally, the correlation coefficient R of each target band is obtained 2 They are: The correlation coefficient R of the second band 2 The correlation coefficient R of the third band is 0.0014. 2 The correlation coefficient R of the fourth band is 0.0622. 2 is 2E-05, and the correlation coefficient R of the 5th band 2 The correlation coefficient R of the sixth band is 0.1989. 2 The correlation coefficient R of the seventh band is 0.7867. 2 The correlation coefficient R of the 8th band is 0.8385. 2 The correlation coefficient R of the 8A band is 0.825. 2 The correlation coefficient R of the 11th band is 0.8603. 2 The correlation coefficient R of the 12th band is 0.3277. 2 It is 0.0059.

[0053] Step 105: Based on all the correlation coefficients, determine the sensitive band for potatoes in the growing stage from all the target bands.

[0054] Optionally, each of the correlation coefficients is compared with a preset correlation coefficient to obtain a first comparison result; the target band corresponding to the target correlation coefficient represented by the first comparison result being greater than the preset correlation coefficient is determined as the sensitive band, and the sensitive band includes band 8A, band 8, band 7 and band 6.

[0055] The preset correlation coefficient may be an empirical value, for example, the preset correlation coefficient is 0.6.

[0056] For example, when the correlation coefficients of all target bands are obtained, each correlation coefficient is compared with the preset correlation coefficient to obtain a first comparison result, and the target band corresponding to the target correlation coefficient greater than the preset correlation coefficient in the first comparison result is determined as a sensitive band sensitive to potatoes in the planting period. Thus, the sensitive bands of the potato remote sensing image in the planting period are band 8A, band 7, band 8 and band 6, and the correlation coefficients of these four sensitive bands decrease in sequence. The determination of the sensitive bands provides a solid foundation for the construction of the potato spectral index model PSI.

[0057] Step 106: Identify the planting area of ​​potatoes in the growth stage in the target area based on the sensitive band.

[0058] For example, when all sensitive bands are obtained, the pixel values ​​of the pixels of all sensitive bands are analyzed, and the planting area of ​​potatoes in the growth stage in the target area is identified based on the analysis results.

[0059] The invention provides a method for identifying potatoes at the growing stage based on sensitive bands. The method obtains surface reflectance images of multiple target bands based on remote sensing images of potatoes at the growing stage in a target area. Based on the surface reflectance images of the multiple target bands, multiple random points representing potatoes at the growing stage are determined. For each target band, a correlation coefficient of the target band is determined based on the surface reflectance value of each random point under the target band and the total surface reflectance value of each random point under all target bands. Based on all correlation coefficients, a sensitive band for potatoes at the growing stage is determined from all target bands, and then a planting area of ​​potatoes at the growing stage in the target area is identified based on the sensitive bands. It can be seen that the present invention can determine the sensitive bands for potatoes in the growing stage from all target bands based on the correlation coefficients of each target band, and then identify the planting area of ​​potatoes in the growing stage in the target area based on the sensitive bands, overcoming the defects of the related art that samples need to be labeled or the sample labeling amount is large each time. At the same time, the present invention mainly uses the images of the growing stage with significant potato features to enhance the potato vegetation features, overcomes the problem of mixed classification and missed classification of potatoes and other crops, and thus improves the accuracy of potato planting area identification.

[0060] In one embodiment, Fig.12 FIG. 2 is a flow chart of a method for identifying potatoes in the growing period based on sensitive bands provided by an embodiment of the present invention. Fig.12 As shown, the above step 106 identifies the planting area of ​​potatoes in the growth period in the target area based on the sensitive band, which can be specifically achieved by the following steps: Step 1061: for each of the sensitive bands, normalize the surface reflectance value of each pixel in the sensitive band to obtain the normalized surface reflectance value of each pixel.

[0061] For example, the surface reflectance values ​​of each pixel in band 6, band 7, band 8 and band 8A are normalized to adjust the range of the surface reflectance values ​​to between 0 and 1. Specifically, the normalization process can be performed using the following formula (1):

[0062] in, Indicates the surface reflectance value of the pixel in the sensitive band. Represents the normalized surface reflectance value of the pixel in the sensitive band. Indicates the minimum value of the surface reflectance of all pixels in the sensitive band. Indicates the maximum value of the surface reflectance of all pixels in the sensitive band.

[0063] Step 1062: for each pixel, based on the normalized surface reflectance value of the pixel in each sensitive band, determine the potato spectral index value corresponding to the pixel.

[0064] Optionally, the square value of the sum of the normalized surface reflectance values ​​of the pixel in the 8A band, the 8th band, the 7th band and the 6th band is determined; the sum of the square value and the normalized surface reflectance value of the pixel in the 8A band is determined as the potato spectral index value corresponding to the pixel, and the 8A band is the most sensitive band for potatoes in the growing stage.

[0065] For example, a potato spectral index model (PSI) is constructed based on the following formula (2). The normalized surface reflectance value of the pixel in each sensitive band is substituted into formula (2) to obtain the potato spectral index value corresponding to the pixel:

[0066] in, It represents the normalized surface reflectance value of the pixel in the 6th band. It represents the normalized surface reflectance value of the pixel in the 7th band. It represents the normalized surface reflectance value of the pixel in the 8th band. It represents the normalized surface reflectance value of the pixel in the 8A band. The contrast between potato features and other ground features is increased by the square of the sum value. The potato features are further highlighted by superimposing the 8A band, which is most sensitive to potatoes in the planting stage.

[0067] Step 1063: Determine a potato spectral index map based on the potato spectral index value corresponding to each pixel.

[0068] For example, when the potato spectral index of each pixel in the sensitive band is obtained, a potato spectral index map can be obtained based on the potato spectral index value of each pixel in the sensitive band.

[0069] Step 1064: Determine the planting area of ​​the potatoes in the growth stage based on the potato spectral index map.

[0070] For example, the area with high brightness value in the potato spectral index map is determined as the planting area of ​​potatoes in the growth stage.

[0071] In this embodiment, based on the sensitive bands for potatoes in the growing period, the potato spectral index value of each pixel in the sensitive band is determined to obtain a potato spectral index map, and then the planting area of ​​potatoes in the growing period is determined based on the analysis of the potato spectral index map. Since the potato spectral index model is constructed based on the sensitive bands for potatoes in the growing period, the potato spectral features are amplified by the operation of the potato spectral index model, and the difference between potatoes and other types of land objects is enhanced, thereby improving the accuracy of identifying potato planting areas through the potato spectral index model.

[0072] In one embodiment, the above step 1064 determines the planting area of ​​the potatoes in the growth period based on the potato spectral index map, which can be specifically implemented in the following manner: The potato spectral index value of each pixel in the potato spectral index map is compared with a preset spectral index value to obtain a second comparison result; the pixel corresponding to the potato spectral index value greater than the preset spectral index value in the second comparison result is determined as the target pixel; and the area where all the target pixels are located is determined as the planting area of ​​potatoes in the growth period.

[0073] Among them, the preset spectral index value can be determined based on the surface reflectance values ​​of all the above-mentioned random points and the image characteristics of the surface reflectance images of multiple target bands, or it can be a value determined based on experience. For example, the preset spectral index value is 0.121, and the present invention is not limited to this.

[0074] For example, after obtaining the potato spectral index value of each pixel, for each pixel, the potato spectral index value of the pixel is compared with the preset spectral index value. When the potato spectral index value of the pixel is greater than the preset spectral index value, it indicates that the ground object type of the pixel is potato. According to the same method, a second comparison result of the potato spectral index value of each pixel and the preset spectral index value can be obtained. Finally, the pixels corresponding to the potato spectral index values ​​greater than the preset spectral index values ​​in the second comparison result are all determined as target pixels, and the area where all the target pixels are located is determined as the planting area of ​​potatoes in the planting period. The area of ​​the area where all the target pixels are located is calculated, and finally the planting area of ​​potatoes in the planting period is obtained.

[0075] In this embodiment, the potato spectral index value of each pixel in the potato spectral index map is compared with the preset spectral index value to obtain a second comparison result, and the pixel corresponding to the potato spectral index value greater than the preset spectral index value in the second comparison result is determined as the target pixel, and the area where all the target pixels are located is determined as the planting area of ​​potatoes in the growing period. Since the potato spectral index model is constructed based on the sensitive bands of potatoes in the growing period, the potato spectral characteristics are amplified by the operation of the potato spectral index model, and the distinction between potatoes and other types of land objects is enhanced, thereby improving the accuracy of identifying potato planting areas through the potato spectral index model.

[0076] In one embodiment, the remote sensing image includes surface reflectance images of bands 2, 3, 4 and 8 of a first spatial resolution, and surface reflectance images of bands 5, 6, 7, 8A, 11 and 12 of a second spatial resolution, and the first spatial resolution is greater than the second spatial resolution; the above step 102 acquires surface reflectance images of multiple target bands based on the remote sensing image, which can be specifically implemented in the following manner: Resampling the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the second spatial resolution to obtain the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution; The surface reflectance images of the 2nd band, the 3rd band, the 4th band and the 8th band of the first spatial resolution and the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution are superimposed to obtain the surface reflectance images of the multiple target bands.

[0077] Among them, the first spatial resolution is 10 meters, and the second spatial resolution is 20 meters.

[0078] For example, the surface reflectance images of bands 5, 6, 7, 8A, 11 and 12 with a spatial resolution of 20 meters are resampled to obtain surface reflectance images of bands 5, 6, 7, 8A, 11 and 12 with a spatial resolution of 10 meters, the surface reflectance images of bands 2, 3, 4 and 8 with a spatial resolution of 10 meters and the surface reflectance images of bands 5, 6, 7, 8A, 11 and 12 with a spatial resolution of 10 meters are superimposed and combined to obtain a combined image, and the combined image is cropped using the boundary of the target area to obtain a 10-band surface reflectance image of the target area.

[0079] In this embodiment, the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the second spatial resolution are resampled to obtain the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution, and then the surface reflectance images of the 2nd band, the 3rd band, the 4th band and the 8th band of the first spatial resolution and the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution are superimposed to obtain surface reflectance images of multiple target bands, so as to facilitate the subsequent identification of potato planting areas based on the surface reflectance images of the multiple target bands.

[0080] Fig.13 FIG. 1 is an overall schematic diagram of a method for identifying potatoes in the growing period based on sensitive bands provided by an embodiment of the present invention. Fig.13 As shown, a remote sensing image of a target area including potatoes in the growing period is obtained, the remote sensing image is preprocessed, and the bands are cropped and superimposed to obtain surface reflectance images of multiple target bands, and the surface reflectance images of multiple target bands are visually interpreted to obtain multiple random points representing potatoes in the growing period, and the multiple random points are used as sample points, and multiple random verification points representing potatoes in the growing period are obtained, and the surface reflectance values ​​of the multiple random verification points are analyzed for potato sensitive bands to determine the sensitive bands for potatoes in the growing period, and the potato spectral index value is determined based on the surface reflectance value of each pixel under the sensitive band, and the preset spectral index value is determined, and the potato spectral index value is compared with the preset spectral index value, and finally the planting area of ​​potatoes in the growing period is determined. Finally, the planting area of ​​potatoes in the growing period is evaluated for accuracy based on multiple random verification points to further improve the accuracy of determining the planting area of ​​potatoes in the growing period.

[0081] For example, based on the 10-band surface reflectance image of XX City, the present invention randomly selects 200 random verification points for manual visual interpretation and accuracy evaluation. Table 2 shows the results of visual interpretation: Table 2

[0082] Among them, TP (True Positive) is the number of samples predicted as positive and actually positive; TN (True Negative) is the number of samples predicted as negative and actually negative; FP (False Positive) is the number of samples predicted as positive but actually negative, i.e. false positive; FN (False Negative) is the number of samples predicted as negative but actually positive, i.e. false negative. The indicators used for accuracy evaluation include Precision, Recall and F1. Precision can be expressed by the following formula (3), Recall can be expressed by the following formula (4), and F1 can be expressed by the following formula (5):

[0083]

[0084]

[0085] After calculation, the accuracy of potato planting areas and planting areas identified by the potato spectral index model PSI was 95.83%, the recall rate was 92%, and the F1 value was 93.88%.

[0086] The accuracy evaluation results show that the potato spectral index model PSI can extract the potato planting area and planting area with high precision, effectively avoiding mixed classification and missed classification. Since the potato spectral index model PSI is constructed according to the sensitive band of potato, it has good universality. The potato spectral index model PSI is used to identify the potato planting area. It does not need to perform sample labeling every time like the supervised classification algorithm, nor does it need to perform a large amount of sample labeling like the deep learning algorithm training model. It realizes the convenient and rapid identification of potato planting areas through the spectral index threshold method, thereby reducing the overall technical difficulty and cost investment of identifying potatoes through remote sensing technology, and greatly improving the recognition efficiency.

[0087] The following is a description of a potato identification device in the growing period based on a sensitive band provided by the present invention. The potato identification device in the growing period based on a sensitive band described below and the potato identification method in the growing period based on a sensitive band described above can be referred to each other.

[0088] Fig.14 FIG. 1 is a schematic diagram of a structure of a potato identification device in the growth period based on a sensitive band according to an embodiment of the present invention. Fig.14 As shown, the apparatus 1400 for identifying potatoes in the growing period based on sensitive bands includes a first acquisition unit 1401, a second acquisition unit 1402, a first determination unit 1403, a second determination unit 1404, a third determination unit 1405 and an identification unit 1406, wherein: The first acquisition unit 1401 is used to acquire a remote sensing image of a target area including potatoes in the growing period; A second acquisition unit 1402 is used to acquire surface reflectance images of multiple target bands based on the remote sensing image; The first determining unit 1403 is used to determine a plurality of random points representing potatoes in the growing period based on the surface reflectance images of the plurality of target bands; The second determining unit 1404 is used to determine, for each of the target bands, a correlation coefficient of the target band based on a surface reflectance value of each of the random points in the target band and a total surface reflectance value of each of the random points in all of the target bands; The third determining unit 1405 is used to determine the sensitive band for potatoes in the growing period from all the target bands based on all the correlation coefficients; The identification unit 1406 is used to identify the planting area of ​​potatoes in the growth stage in the target area based on the sensitive band.

[0089] The invention provides a potato identification device at the growing stage based on sensitive bands. The device obtains surface reflectance images of multiple target bands based on remote sensing images of target areas including potatoes at the growing stage. Based on the surface reflectance images of the multiple target bands, multiple random points representing potatoes at the growing stage are determined. For each target band, the correlation coefficient of the target band is determined based on the surface reflectance value of each random point under the target band and the total surface reflectance value of each random point under all target bands. Based on all the correlation coefficients, the sensitive band for potatoes at the growing stage is determined from all the target bands, and then the planting area of ​​potatoes at the growing stage in the target area is identified based on the sensitive bands. It can be seen that the present invention can determine the sensitive bands for potatoes in the growing stage from all target bands based on the correlation coefficients of each target band, and then identify the planting area of ​​potatoes in the growing stage in the target area based on the sensitive bands, overcoming the defects of the related art that samples need to be labeled or the sample labeling amount is large each time. At the same time, the present invention mainly uses the images of the growing stage with significant potato features to enhance the potato vegetation features, overcomes the problem of mixed classification and missed classification of potatoes and other crops, and thus improves the accuracy of potato planting area identification.

[0090] Based on any of the foregoing embodiments, the third determining unit 1405 is specifically configured to: Compare each of the correlation coefficients with a preset correlation coefficient to obtain a first comparison result; The target band corresponding to the target correlation coefficient greater than the preset correlation coefficient in the first comparison result is determined as the sensitive band, and the sensitive band includes the 8A band, the 8th band, the 7th band and the 6th band.

[0091] Based on any of the above embodiments, the identification unit 1406 is specifically used for: For each of the sensitive bands, normalizing the surface reflectance value of each pixel in the sensitive band to obtain the normalized surface reflectance value of each pixel; For each of the pixels, based on the normalized surface reflectance value of the pixel in each of the sensitive bands, determine the potato spectral index value corresponding to the pixel; Determining a potato spectral index map based on the potato spectral index value corresponding to each of the pixels; Based on the potato spectral index map, the planting area of ​​the potatoes in the growth stage is determined.

[0092] Based on any of the above embodiments, the identification unit 1406 is further specifically configured to: Comparing the potato spectral index value of each pixel in the potato spectral index map with a preset spectral index value to obtain a second comparison result; Determine the pixel corresponding to the potato spectral index value greater than the preset spectral index value in the second comparison result as the target pixel; The area where all the target pixels are located is determined as the planting area for potatoes in the growth stage.

[0093] Based on any of the above embodiments, the identification unit 1406 is further specifically configured to: Determine the square value of the sum of the normalized surface reflectance values ​​of the pixel in the 8A band, the 8th band, the 7th band, and the 6th band; The sum of the square value and the normalized surface reflectance value of the pixel in the 8A band is determined as the potato spectral index value corresponding to the pixel, and the 8A band is the most sensitive band for potatoes in the growing stage.

[0094] Based on any of the above embodiments, the remote sensing image includes surface reflectance images of bands 2, 3, 4 and 8 of a first spatial resolution, and surface reflectance images of bands 5, 6, 7, 8A, 11 and 12 of a second spatial resolution, and the first spatial resolution is greater than the second spatial resolution; the second acquisition unit 1802 is specifically used to: Resampling the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the second spatial resolution to obtain the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution; The surface reflectance images of the 2nd band, the 3rd band, the 4th band and the 8th band of the first spatial resolution and the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution are superimposed to obtain the surface reflectance images of the multiple target bands.

[0095] Based on any of the foregoing embodiments, the second determining unit 1404 is specifically configured to: Constructing a two-dimensional scatter plot based on the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all the target bands; Based on the two-dimensional scatter plot, determining a linear fitting function; Based on the linear fit function, the correlation coefficient is determined.

[0096] Fig.15 is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention, such as Fig.15 As shown, the electronic device may include: a processor 1510, a communications interface 1520, a memory 1530 and a communication bus 1540, wherein the processor 1510, the communications interface 1520 and the memory 1530 communicate with each other through the communication bus 1540. The processor 1510 may call the logic instructions in the memory 1530 to execute a method for identifying potatoes in the growing period based on sensitive bands, the method comprising: acquiring a remote sensing image of a target area including potatoes in the growing period; Acquire surface reflectance images of multiple target bands based on the remote sensing image; Based on the surface reflectance images of the multiple target bands, determining multiple random points representing potatoes at the growing stage; For each of the target bands, determining a correlation coefficient of the target band based on a ground surface reflectance value of each of the random points in the target band and a total ground surface reflectance value of each of the random points in all of the target bands; Based on all the correlation coefficients, determining the sensitive band for potatoes in the growing stage from all the target bands; The planting area of ​​potatoes in the growth stage in the target area is identified based on the sensitive band.

[0097] In addition, the logic instructions in the above-mentioned memory 1530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0098] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the method for identifying potatoes in the growing period based on sensitive bands provided by the above methods, the method comprising: acquiring a remote sensing image of a target area including potatoes in the growing period; Acquire surface reflectance images of multiple target bands based on the remote sensing image; Based on the surface reflectance images of the multiple target bands, determining multiple random points representing potatoes at the growing stage; For each of the target bands, determining a correlation coefficient of the target band based on a ground surface reflectance value of each of the random points in the target band and a total ground surface reflectance value of each of the random points in all of the target bands; Based on all the correlation coefficients, determining the sensitive band for potatoes in the growing stage from all the target bands; The planting area of ​​potatoes in the growth stage in the target area is identified based on the sensitive band.

[0099] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for identifying potatoes in the growing period based on sensitive bands provided by the above methods is implemented, and the method comprises: acquiring a remote sensing image of a target area including potatoes in the growing period; Acquire surface reflectance images of multiple target bands based on the remote sensing image; Based on the surface reflectance images of the multiple target bands, determining multiple random points representing potatoes at the growing stage; For each of the target bands, determining a correlation coefficient of the target band based on a ground surface reflectance value of each of the random points in the target band and a total ground surface reflectance value of each of the random points in all of the target bands; Based on all the correlation coefficients, determining the sensitive band for potatoes in the growing stage from all the target bands; The planting area of ​​potatoes in the growth stage in the target area is identified based on the sensitive band.

[0100] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0101] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying potatoes in the growing period based on sensitive bands, characterized in that: include: Acquire remote sensing images of target areas including potatoes in the growing period; Acquire surface reflectance images of multiple target bands based on the remote sensing image; Based on the surface reflectance images of the multiple target bands, determining multiple random points representing potatoes at the growing stage; For each of the target bands, determining a correlation coefficient of the target band based on a ground surface reflectance value of each of the random points in the target band and a total ground surface reflectance value of each of the random points in all of the target bands; Based on all the correlation coefficients, determining the sensitive band for potatoes in the growing stage from all the target bands; The planting area of ​​potatoes in the growth stage in the target area is identified based on the sensitive band.

2. The method for identifying potatoes in the growing period based on sensitive bands according to claim 1, characterized in that: The step of determining the sensitive waveband for potatoes in the growing period from all the target wavebands based on all the correlation coefficients comprises: Compare each of the correlation coefficients with a preset correlation coefficient to obtain a first comparison result; The target band corresponding to the target correlation coefficient greater than the preset correlation coefficient in the first comparison result is determined as the sensitive band, and the sensitive band includes the 8A band, the 8th band, the 7th band and the 6th band.

3. The method for identifying potatoes at the growing stage based on sensitive bands according to claim 2, characterized in that: The step of identifying the planting area of ​​potatoes in the target area based on the sensitive band includes: For each of the sensitive bands, normalizing the surface reflectance value of each pixel in the sensitive band to obtain the normalized surface reflectance value of each pixel; For each of the pixels, based on the normalized surface reflectance value of the pixel in each of the sensitive bands, determine the potato spectral index value corresponding to the pixel; Determining a potato spectral index map based on the potato spectral index value corresponding to each of the pixels; Based on the potato spectral index map, the planting area of ​​the potatoes in the growth stage is determined.

4. The method for identifying potatoes at the growing stage based on sensitive bands according to claim 3, characterized in that: The step of determining the planting area of ​​the potatoes in the growth period based on the potato spectral index map includes: Comparing the potato spectral index value of each pixel in the potato spectral index map with a preset spectral index value to obtain a second comparison result; Determine the pixel corresponding to the potato spectral index value greater than the preset spectral index value in the second comparison result as the target pixel; The area where all the target pixels are located is determined as the planting area for potatoes in the growth stage.

5. The method for identifying potatoes at the growing stage based on sensitive bands according to claim 3, characterized in that: The step of determining the potato spectral index value corresponding to the pixel based on the normalized surface reflectance value of the pixel in each sensitive band includes: Determine the square value of the sum of the normalized surface reflectance values ​​of the pixel in the 8A band, the 8th band, the 7th band, and the 6th band; The sum of the square value and the normalized surface reflectance value of the pixel in the 8A band is determined as the potato spectral index value corresponding to the pixel, and the 8A band is the most sensitive band for potatoes in the growing stage.

6. The method for identifying potatoes in the growing period based on sensitive bands according to claim 1, characterized in that: The remote sensing images include surface reflectance images of bands 2, 3, 4 and 8 of a first spatial resolution, and surface reflectance images of bands 5, 6, 7, 8A, 11 and 12 of a second spatial resolution, wherein the first spatial resolution is greater than the second spatial resolution; The step of acquiring surface reflectance images of multiple target bands based on the remote sensing image comprises: Resampling the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the second spatial resolution to obtain the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution; The surface reflectance images of the 2nd band, the 3rd band, the 4th band and the 8th band of the first spatial resolution and the surface reflectance images of the 5th band, the 6th band, the 7th band, the 8A band, the 11th band and the 12th band of the first spatial resolution are superimposed to obtain the surface reflectance images of the multiple target bands.

7. The method for identifying potatoes in the growing period based on sensitive bands according to claim 1, characterized in that: Determining the correlation coefficient of the target band based on the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all the target bands includes: Constructing a two-dimensional scatter plot based on the surface reflectance value of each random point in the target band and the total surface reflectance value of each random point in all the target bands; Based on the two-dimensional scatter plot, determining a linear fitting function; Based on the linear fit function, the correlation coefficient is determined.

8. A device for identifying potatoes in the growing period based on sensitive bands, characterized in that: include: A first acquisition unit is used to acquire a remote sensing image of a target area including potatoes in the growing period; A second acquisition unit, used for acquiring surface reflectance images of multiple target bands based on the remote sensing image; A first determination unit is used to determine a plurality of random points representing potatoes at a growing stage based on the surface reflectance images of the plurality of target bands; A second determination unit is used to determine, for each of the target bands, a correlation coefficient of the target band based on a surface reflectance value of each of the random points in the target band and a total surface reflectance value of each of the random points in all of the target bands; A third determining unit is used to determine a sensitive band for potatoes in the growing stage from all the target bands based on all the correlation coefficients; An identification unit is used to identify a planting area of ​​potatoes in the growth stage within the target area based on the sensitive band.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the method for identifying potatoes in the growing stage based on sensitive bands as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying potatoes in the growing stage based on sensitive bands as described in any one of claims 1 to 7 is implemented.

Images