Bare soil identification method and device based on novel remote sensing index and computer program

Abstract

The invention relates to a bare soil identification method and device based on a novel remote sensing index and a computer program, and the method comprises the steps: constructing a bare earth surface remote sensing index based on a visible light-near infrared band combination, dividing a threshold value according to a threshold value segmentation algorithm, and identifying bare earth surface distribution; the index can also be combined with an improved normalized difference water body index to construct a derived wetland combined bare soil index and a non-wetland combined bare soil index for extracting wetland and non-wetland bare soil. According to the bare soil identification method provided by the invention, bare soil and other ground features can be effectively distinguished so as to improve the accuracy and efficiency of bare soil identification, and wetland bare soil and non-wetland bare soil can be further distinguished. The method has good universality, good effects are obtained in multiple regions, rapid and accurate extraction of spatial and temporal changes of bare soil is facilitated, and a technical method is provided for water and soil conservation, natural resource protection, territorial resource planning and the like.

Description

Technical Field

[0001] The invention belongs to the field of land use remote sensing identification, and relates to a bare soil identification method, equipment and computer program based on a novel remote sensing index. Background Art

[0002] Bare land is caused by climate change, human activities and other factors, including bare soil, tidal flats, deserts and other types. Among them, bare soil is the main type of land where soil erosion occurs, which not only leads to a decline in soil fertility and land degradation, but also causes river siltation and flooding when soil enters water bodies. With the continuous intensification of climate warming, soil erosion has become an important problem facing mankind. Therefore, the use of remote sensing technology to quickly and accurately identify bare land is of great significance to soil and water conservation and natural resource protection.

[0003] Remote sensing technology has the characteristics of wide range, long time series, convenient and fast identification of surface features, and has become the main technical method for extracting bare soil information. At present, the remote sensing indices commonly used to extract bare soil include Normalized Difference Soil Index (NDSI), Bareness soil Index (BSI), Normalized Difference Bareness Index (NDBaI), etc. Although remote sensing indices can enhance the characteristics of specific targets, due to the same spectrum of different objects and sensor performance reasons, targets such as impervious surfaces are easy to be confused. Therefore, some scholars have proposed using multiple indices such as impervious surface index and surface temperature to construct a classification tree system to extract bare soil information.

[0004] In summary, the remote sensing indices currently used to extract bare soil are mostly based on dual-band ratios and normalization, or multi-indices are used for step-by-step extraction. There is a lack of bare soil indices that can directly distinguish other land use types. Summary of the invention

[0005] The first purpose of the present invention is to provide a bare soil identification method based on a new remote sensing index. The new remote sensing index is constructed by combining field surveys, visual interpretation of images and measured remote sensing reflectance, and is constructed using band combinations, which can well distinguish bare soil from other ground objects.

[0006] To achieve the above object, the present invention adopts the following technical solution:

[0007] A bare soil remote sensing identification method, comprising:

[0008] Obtain remote sensing images and calculate the BSFI index value of each pixel based on the following formula:

[0009]

[0010] In the formula, is the central wavelength of the remote sensing image The reflectivity of the band; , , , , , are respectively the central wavelength of the red light band, the central wavelength of the blue light band, the central wavelength of the near infrared band, the central wavelength of the short-wave infrared band 1, and the central wavelength of the short-wave infrared band 2;

[0011] The BSFI image is subjected to binary classification through threshold segmentation to extract the bare surface, namely the bare soil.

[0012] As a preferred implementation, the method further includes: assigning the following values ​​to the pixels of the remote sensing image according to the improved normalized difference water index MNDWI:

[0013]

[0014] The WCBSI image was classified into two categories by threshold segmentation to extract the bare soil in the wetland.

[0015] As a preferred implementation, the method further includes: assigning the following values ​​to the pixels of the remote sensing image according to the improved normalized difference water index MNDWI:

[0016]

[0017] The NonWCBSI image was classified into two categories by threshold segmentation to extract non-wetland bare soil.

[0018] Furthermore, in the above method, the Otsu method is used to calculate the image threshold, and binary classification is performed based on the threshold to extract pixels above the image threshold, which are bare soil / wetland bare soil / non-wetland bare soil pixels.

[0019] As a preferred embodiment, the remote sensing image is a Sentinel-2 MSI image. Further, the Surface Reflectance data of the Sentinel-2 MSI L2A remote sensing image is used as a data source. The index of the present invention is developed based on the Sentinel-2 MSI image, and has been verified to have good application effects for the bands of Sentinel-2 MSI.

[0020] As a preferred implementation, the method further includes preprocessing the remote sensing image, including removing cloud-covered areas and removing areas affected by solar flares.

[0021] A second object of the present invention is to provide a computer device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0022] A third object of the present invention is to provide a computer-readable storage medium having a computer program / instruction stored thereon, which implements the steps of the above method when the computer program / instruction is executed by a processor.

[0023] A fourth object of the present invention is to provide a computer program product, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0024] The present invention has the following beneficial effects:

[0025] The present invention develops a bare soil identification method based on a new remote sensing index. The BSFI index amplifies the gap between land types and can significantly distinguish bare soil from other land types. The present invention uses the index to further combine the modified normalized difference water index (MNDWI) to construct the derived wetland combinational bare soil index (WCBSI) and non-wetland combinational bare soil index (NonWCBSI). The WCBSI index and the NonWCBSI index can clearly distinguish wetland bare soil (tidal flat swamp), non-wetland bare soil and other land types, effectively reducing the generation of uncertainty and systematic errors in the identification process. The technology of the present invention has been preliminarily applied in many regions around the world, and it is found that the technology of the method has significant promotion and applicability. The technology discovered provides a basis for quickly identifying different types of bare surfaces based on remote sensing images, which is of great significance to soil and water conservation and natural resource protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in every figure. Embodiments of various aspects of the present invention will now be described by way of example and with reference to the accompanying drawings, in which:

[0027] Figure 1 It is a flow chart of the method of the present invention.

[0028] Figure 2 These are Sentinel-2 MSI images of typical areas such as Zhuhai, Quanzhou, and Los Angeles.

[0029] Figure 3 It is the calculation result of BSFI index of typical regions such as Zhuhai, Quanzhou and Los Angeles.

[0030] Figure 4 These are the identification results of wetland pixels and non-wetland pixels in typical areas such as Zhuhai, Quanzhou, and Los Angeles.

[0031] Figure 5 These are the NonWCBSI calculation results for typical areas such as Zhuhai, Quanzhou, and Los Angeles.

[0032] Figure 6 It is the result of identifying non-wetland bare soil in typical areas such as Zhuhai, Quanzhou, and Los Angeles.

[0033] Figure 7 These are the WCBSI calculation results for typical areas such as Zhuhai, Quanzhou, and Los Angeles.

[0034] Figure 8 It is the identification result of wetland bare soil (swamp mudflat) in typical areas such as Zhuhai, Quanzhou, and Los Angeles.

[0035] Fig. 9 It is a verification point.

[0036] Fig.10 It is the calculation result of typical bare soil index in typical areas such as Zhuhai, Quanzhou and Los Angeles.

[0037] Fig.11 It is a distribution box plot of the typical bare soil index and the index proposed by the present invention.

[0038] In the above-mentioned Figures 1-10, the coordinates, symbols or other expressions expressed in English are all well known in the art and will not be described in detail in this example. DETAILED DESCRIPTION

[0039] In order to better understand the technical content of the present invention, specific embodiments are given and described as follows in conjunction with the accompanying drawings.

[0040] Various aspects of the present invention are described in this disclosure with reference to the accompanying drawings, in which many illustrative embodiments are shown. The embodiments of the present disclosure are not necessarily intended to include all aspects of the present invention. It should be understood that the various concepts and embodiments introduced above, as well as those described in more detail below, can be implemented in any of many ways, and this is because the concepts and embodiments disclosed by the present invention are not limited to any implementation. In addition, some aspects disclosed by the present invention can be used alone or in any appropriate combination with other aspects disclosed by the present invention.

[0041] Example 1

[0042] The method of the present invention is based on the surface reflectance data of Sentinel-2 MSI L2A remote sensing image. , , , and Calculate the bare surface index BSFI. In the SurfaceReflectance data, , , , and They are the reflectances of the bands with a central wavelength of 665 nm, 495 nm, 835 nm, 1610 nm, and 2200 nm. Then, based on the BSFI values ​​of different images, the Otsu threshold is calculated and bare soil and other land types are automatically extracted by binary classification.

[0043] Furthermore, the improved normalized difference water index MNDWI can be combined to calculate the wetland combined bare soil index WCBSI and the non-wetland combined bare soil index NonWCBSI of the image respectively; based on the WCBSI and NonWCBSI of different images, the Otsu threshold is calculated respectively and the wetland bare soil / non-wetland bare soil and other land types are automatically extracted by binary classification.

[0044] As an exemplary description, the following takes the application of Sentinel-2 MSI sensor in Zhuhai, Quanzhou and Los Angeles as an example, and combines the accompanying drawings to specifically illustrate the implementation of the above method. The specific process of the present invention is as follows: Figure 1 As shown, the following steps are included:

[0045] Step 1: Obtain the surface reflectance data of the Sentinel-2 MSI L2A remote sensing image of the study area and extract the required bands , , , and And calculate the bare surface index BSFI.

[0046] like Figure 2 As shown, the images used are located in Zhuhai, Quanzhou and Los Angeles, and the image numbers are 20230129T025959_20230129T030738_T49QGE, 20240508T025551_20240508T030143_T50SMJ and 20221107T024919_20221107T025702_T50SPA respectively.

[0047] Based on the band reflectance extracted from the image data , perform BSFI calculation:

[0048]

[0049] In the formula, The central wavelength is The reflectivity of the band.

[0050] Figure 3 The calculation results of the bare surface index (BSFI) of the applied image are shown. The larger the BSFI, the higher the confidence of the bare surface.

[0051] Step 2: Based on the improved normalized difference water index MNDWI and combined with the BSFI index, the wetland combined bare soil index WCBSI and the non-wetland combined bare soil index NonWCBSI are constructed to further distinguish wetland bare soil and non-wetland bare soil.

[0052] According to the characteristics of wetlands and non-wetlands, and the improved normalized difference water index (MNDWI) to distinguish between wetlands and non-wetland bare soil. , it is judged as a wetland pixel, otherwise it is a non-wetland pixel, such as Figure 4 shown.

[0053] based on And the improved normalized difference water index MNDWI, further calculate the wetland combined bare soil index WCBSI and non-wetland combined bare soil index NonWCBSI:

[0054]

[0055]

[0056]

[0057] The calculation results of non-wetland combined bare soil index NonWCBSI are as follows: Figure 5 As shown in the figure, the calculation results of wetland combined bare soil index WCBSI are as follows: Figure 7 shown.

[0058] The different images with calculated WCBSI and NonWCBSI values ​​were subjected to Otsu threshold calculations and automatically classified into two categories based on the threshold. Pixels greater than the threshold were identified as wetland bare soil (WCBSI image) / non-wetland bare soil (NonWCBSI image), and pixels less than or equal to the threshold were identified as other land types. The Otsu threshold classification based on the non-wetland combined bare soil index NonWCBSI is as follows Figure 6 As shown in Figure 2, the Otsu threshold classification based on the wetland combined bare soil index (WCBSI) is as follows: Figure 8 shown.

[0059] Step 3: In the applied images, visual interpretation and field investigation are combined to set verification points, and the performance of the new bare soil index proposed in the present invention is evaluated by combining typical bare soil indices: bare soil index BSI and drought bare soil index DBSI.

[0060] Fig. 9 The verification points of the images of the three regions are shown. In order to test and compare the recognition and extraction effects of different land types, 9 common land types were selected: non-wetland bare soil, sparse vegetation, dense vegetation, low-rise buildings, high-rise buildings, high-reflectivity impervious surfaces, water bodies, wetland bare soil / tidal flat swamps, and beach deserts.

[0061] Fig.10 The calculation results of typical bare soil indices BSI and DBSI are shown.

[0062] Fig.11 The statistical box plots of two typical bare soil indices, BSI and DBSI, and the new bare soil index proposed in the present invention are shown. It can be seen that both typical bare soil indices are easy to confuse bare soil and other land types, especially BSI has the worst extraction effect. Moreover, it is difficult for the two typical bare soil indices to distinguish between wetland bare soil and non-wetland bare soil. The new bare soil index BSFI proposed in the present invention amplifies the gap between various land types and can significantly distinguish between bare soil and other land types. NonWCBSI and WCBSI of BSFI can also further distinguish between wetland bare soil and non-wetland bare soil, especially the median and average values ​​of non-wetland bare soil and wetland bare soil are significantly different from other land types, which is helpful to quickly extract different types of bare soil in applications.

[0063] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. A person skilled in the art of the present invention may make various modifications and alterations without departing from the spirit and scope of the present invention.

Claims

1. A bare soil remote sensing identification method, characterized in that: include: Obtain remote sensing images and calculate the BSFI index value of each pixel based on the following formula: ; In the formula, is the central wavelength of the remote sensing image The reflectivity of the band; , , , , , are respectively the central wavelength of the red light band, the central wavelength of the blue light band, the central wavelength of the near infrared band, the central wavelength of the short-wave infrared band 1, and the central wavelength of the short-wave infrared band 2; The BSFI image is subjected to binary classification through threshold segmentation to extract the bare surface, namely the bare soil.

2. The method according to claim 1, characterized in that Also includes: According to the improved normalized difference water index MNDWI, the pixels of the remote sensing image are assigned as follows: ; The WCBSI image was classified into two categories by threshold segmentation to extract the bare soil in the wetland.

3. The method according to claim 1, characterized in that: Also includes: According to the improved normalized difference water index MNDWI, the pixels of the remote sensing image are assigned as follows: ; The NonWCBSI image was classified into two categories by threshold segmentation to extract non-wetland bare soil.

4. The method according to any one of claims 1 to 3, characterized in that: The Otsu method was used to calculate the image threshold, and binary classification was performed based on the threshold to extract pixels above the image threshold, which were bare soil / wetland bare soil / non-wetland bare soil pixels.

5. The method according to claim 1, characterized in that The remote sensing image is a Sentinel-2 MSI image.

6. The method according to claim 5, characterized in that The surface reflectance data of Sentinel-2 MSI L2A remote sensing image is used as the data source.

7. The method according to claim 1, characterized in that It also includes pre-processing of remote sensing images, including removing cloud-covered areas and removing areas affected by solar flares.

8. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 3 and 5 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the method described in any one of claims 1 to 3 and 5 to 7 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the method described in any one of claims 1 to 3 and 5 to 7 are implemented.

Images