Remote sensing identification method for urban bare soil

Abstract

The invention discloses a remote sensing identification method for urban bare soil in the technical field of remote sensing identification, which comprises the following steps of: acquiring satellite remote sensing data of a target urban area, synthesizing the remote sensing data into a true color image map, dividing the true color image map into a plurality of bare soil lattice points, the true color image map is visually interpreted to obtain a plurality of bare soil areas. According to the method, remote sensing data is synthesized into a true color image map, a plurality of bare soil areas are interpreted visually, part of urban bare soil under the same space-time condition is selected as an experimental area, and a reasonable low threshold value and a reasonable high threshold value of urban bare soil indexes are determined according to the bare soil indexes of the bare soil areas. And whether the bare soil lattice points are the urban bare soil or not is judged according to the low threshold value and the high threshold value of the urban bare soil index, and the method is simple, high in operability, good in accuracy of judging the urban bare soil, capable of avoiding missing the bare soil lattice points, low in calculation performance requirement and capable of rapidly completing the urban bare soil identification work.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing identification, and specifically to a method for remotely sensing and identifying urban bare soil. Background Art

[0002] Bare soil refers to the soil surface directly exposed to the natural environment without being covered by plants or buildings. The formation of this landform may stem from various factors, including natural causes such as extreme weather conditions like drought, severe cold, strong winds, and heavy snowstorms, or animal activities that cause severe damage to the original ecological community; human factors may involve overgrazing, agricultural cultivation, construction activities, etc., which may directly lead to vegetation damage and thus leave the land bare. The existence of bare soil is prone to causing soil erosion problems: due to the lack of vegetation protection, the exposed soil is easily eroded by rainwater, resulting in soil erosion, which will have an adverse impact on the sustainable use of the land. In order to mitigate the negative impact of bare soil on the environment, it is necessary to effectively monitor and manage bare soil so as to timely discover and solve the bare soil problem, thereby protecting and maintaining the ecological environment. Among them, urban bare soil refers to those lands without vegetation cover or hardening treatment, which are prone to generating dust and exposing soil, becoming an important source of air particulate matter. These urban bare soils not only affect the air quality of the city, but may also cause soil erosion and visual pollution. Therefore, urban bare soil needs to be identified and treated in a timely manner to reduce the generation and spread of soil dust, thereby reducing the concentration of particulate matter in the atmosphere and improving the urban environmental quality.

[0003] Traditional urban bare soil investigation work mainly relies on manual on-site exploration, which is easily restricted by traffic conditions and subjective awareness. Existing satellite remote sensing urban bare soil identification technologies mainly include methods such as manual visual interpretation, supervised classification, and index method, etc. Among them, manual visual interpretation has a large workload and is prone to omissions; the supervised classification method requires constructing a large number of training samples, with a large time cost and high requirements for computer storage and operation; in the index method, it is easy to encounter the situation of no suitable discrimination threshold. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for remotely sensing and identifying urban bare soil to solve the problems of the above-mentioned manual on-site exploration being easily restricted by traffic conditions and subjective awareness, and the large workload and high time cost of existing satellite remote sensing identification.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] A method for remotely sensing and identifying urban bare soil, comprising the following steps:

[0007] Obtain satellite remote sensing data of the target urban area, synthesize a true color image map from the remote sensing data, divide the true color image map into several bare soil grid points, and visually interpret several bare soil areas from the true color image map;

[0008] Calculate the bare soil index of the bare soil area according to the remote sensing data, and determine the low threshold and high threshold of the urban bare soil index according to the distribution of the bare soil index of the bare soil area;

[0009] Calculate the bare soil index of the bare soil grid points according to the remote sensing data. If the bare soil index of the bare soil grid points is within the range of the low threshold and high threshold of the urban bare soil index, then determine that the bare soil grid points are urban bare soil.

[0010] As a further solution of the present invention: the remote sensing data of the satellite includes blue band data (Blue), green band data (Green) and red band data (Red), and the blue band data, green band data and red band data are synthesized into a true color image map.

[0011] As a further solution of the present invention: the remote sensing data of the satellite further includes near-infrared band data (NIR), mid-infrared band data (MIR) and far-infrared band data (TIR);

[0012] The bare soil index includes normalized bare soil index (NDSI), normalized water body index (MNDWI) and ratio residential area index (RII);

[0013] The normalized bare soil index The normalized bare soil index is used to enhance the bare soil characteristics;

[0014] The normalized water body index The normalized water body index is used to distinguish the bare soil area and the building area of the city;

[0015] The ratio residential area index The ratio residential area index is used to strengthen the distinction between the bare soil area and the building area of the city.

[0016] As a further solution of the present invention: determine the low threshold and high threshold of the urban bare soil index according to the distribution of the bare soil index of the bare soil area;

[0017] The 5th percentile of the numerical distribution of the normalized bare soil index is its low threshold, and the 95th percentile of the numerical distribution of the normalized bare soil index is its high threshold;

[0018] The 5th percentile of the numerical distribution of the normalized water body index is its low threshold, and the 95th percentile of the numerical distribution of the normalized water body index is its high threshold;

[0019] The 5th percentile of the numerical distribution of the ratio residential area index is its low threshold, and the 95th percentile of the numerical distribution of the ratio residential area index is its high threshold.

[0020] As a further solution of the present invention: if the bare soil index of the bare soil grid point is within the low threshold and high threshold of the urban bare soil index, then it is determined that the bare soil grid point is urban bare soil;

[0021] If the normalized bare soil index, normalized water body index, and ratio residential area index of the bare soil grid point are all within their corresponding low thresholds and high thresholds, then it is determined that the bare soil grid point is urban bare soil.

[0022] As a further solution of the present invention: when it is determined that the bare soil grid point is urban bare soil, the urban bare soil is marked separately to highlight the bare soil grid point.

[0023] As a further solution of the present invention: the number of visually interpreted bare soil areas is at least ten, and the number of visually interpreted bare soil grid points is at least five thousand.

[0024] As a further solution of the present invention: the cloud cover of the target urban area is less than 10% to ensure the accuracy of the bare soil identification result in the urban area.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] 1. In the present invention, by synthesizing remote sensing data into a true color image map, visually interpreting several bare soil areas, selecting a part of urban bare soil under the same spatio-temporal conditions as the experimental area, determining the reasonable low threshold and high threshold of the urban bare soil index according to the bare soil index of several bare soil areas, and using the low threshold and high threshold of the urban bare soil index to judge whether the bare soil grid point is urban bare soil, this method is simple, highly operable, has good accuracy in judging urban bare soil, avoids missing bare soil grid points, and has low computational performance requirements, and can quickly complete the urban bare soil identification work.

[0027] 2. In the present invention, by calculating the normalized bare soil index, normalized water body index, and ratio residential area index of several bare soil grid points, and determining that the bare soil grid point is urban bare soil, the urban bare soil area can be quickly obtained, the calculation is fast, this method is simple and only depends on real-time remote sensing data, without big data modeling or constructing a data set, and can realize satellite remote sensing identification of urban bare soil without a large computer for data storage and operation. Especially the ratio residential area index specially used for separating urban buildings greatly improves the accuracy of urban bare soil extraction and reduces the confusion between urban bare soil and urban buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0029] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0030] Embodiment:

[0031] Please refer to Figure 1 , in the embodiments of the present invention, a remote sensing identification method for urban bare soil includes the following steps:

[0032] S1: Obtain satellite remote sensing data of the target urban area, synthesize a true-color image map from the remote sensing data, divide the true-color image map into several bare soil grid points, and visually interpret several bare soil areas from the true-color image map;

[0033] S2: Calculate the bare soil index of the bare soil area according to the remote sensing data, and determine the low threshold and high threshold of the urban bare soil index according to the distribution of the bare soil index of the bare soil area;

[0034] S3: Calculate the bare soil index of the bare soil grid point according to the remote sensing data. If the bare soil index of the bare soil grid point is within the low threshold and high threshold of the urban bare soil index, it is determined that the bare soil grid point is urban bare soil.

[0035] This method synthesizes a true-color image map from remote sensing data, visually interprets several bare soil areas, selects a part of urban bare soil under the same space-time conditions as the experimental area, determines reasonable low and high thresholds of the urban bare soil index according to the bare soil index of several bare soil areas, and uses the low and high thresholds of the urban bare soil index to judge whether the bare soil grid point is urban bare soil. This method is simple, highly operable, has good accuracy in judging urban bare soil, avoids missing bare soil grid points, and has low computational performance requirements, can quickly complete the identification of urban bare soil, is conducive to promoting the scientific management and efficient treatment of urban bare soil dust sources, and provides targeted technical support for the small-scale refined bare soil management of environmental management departments.

[0036] Among them, the bare soil grid point usually refers to the area extracted and analyzed by satellite remote sensing technology. Using a high-resolution true-color image map, the distribution area of bare soil can be quickly and accurately located, and then on-site investigation and rectification work can be carried out. The government and relevant departments can effectively find and rectify these bare soil areas, reducing the impact of dust on the environment and public health.

[0037] In this embodiment, the remote sensing data of the satellite includes blue band data (Blue), green band data (Green), and red band data (Red). The blue band data, green band data, and red band data are synthesized into a true color image map. By synthesizing the true color image map with the blue band data, green band data, and red band data, the colors of the ground objects in the synthesized color image of the true color image map are close to or the same as the actual ground object colors, which is convenient for visually interpreting several bare soil areas from the true color image map.

[0038] In this embodiment, the remote sensing data of the satellite also includes near-infrared band data (NIR), mid-infrared band data (MIR), and far-infrared band data (TIR);

[0039] The bare soil indices include the Normalized Difference Bare Soil Index (NDSI), the Modified Normalized Difference Water Index (MNDWI), and the Ratio of Built-up Index (RII);

[0040] Normalized Difference Bare Soil Index The Normalized Difference Bare Soil Index is used to enhance the bare soil features;

[0041] Modified Normalized Difference Water Index The Modified Normalized Difference Water Index is used to distinguish between bare soil areas and urban built-up areas;

[0042] Ratio of Built-up Index The Ratio of Built-up Index is used to further distinguish between bare soil areas and urban built-up areas.

[0043] Among them, the remote sensing data of the satellite includes the blue band, green band, red band, near-infrared band, mid-infrared band, and far-infrared band. If a single satellite cannot meet the requirements, multiple satellites can be used in combination, but it is necessary to ensure that the transit dates differ by within 15 days. This is mainly to ensure the continuity of Earth observation and the timeliness of data to meet different application requirements. The transit dates of multiple satellites differing by within 15 days are for maintaining continuous observation of the Earth's surface and atmosphere. Different satellites need to transit at similar time points in order to obtain a continuous data sequence. The transit dates of multiple satellites differing by within 15 days can ensure that the acquired data has high timeliness, which is crucial for monitoring changes on the Earth's surface.

[0044] The Normalized Difference Bare Soil Index (NDSI), the Modified Normalized Difference Water Index (NDISI), and the Ratio of Built-up Index (RII) are selected for remote sensing identification of urban bare soil.

[0045] The Normalized Difference Bare Soil Index enhances the bare soil features by normalizing the near-infrared and mid-infrared bands based on the characteristic that the reflectance of bare soil is the highest in the mid-infrared band. It has a good effect on extracting bare soil from natural underlying surfaces such as water bodies and vegetation.

[0046] The Normalized Bare Soil Index enhances both bare soil and built-up land simultaneously, which is prone to confusion. Then, the Normalized Water Index is introduced to strengthen the distinction between the two and separate bare soil and buildings.

[0047] Due to the large urban area and the variety of built-up land types, using only the Normalized Water Index to distinguish between bare soil and built-up land has limited effect. Then, the Ratio Residential Land Index is added for distinction. This index is designed using the strongest and weakest reflection bands of the impervious surface index that are different from other land types and is mainly used for the identification of built-up land. The introduction of this index greatly strengthens the separation of bare soil and buildings.

[0048] Calculating the above three indices requires obtaining the reflectance of the blue band (Blue), green band (Green), near-infrared band (NIR), mid-infrared band (MIR), and thermal infrared band (TIR) at each grid point in the study area for calculation.

[0049] In this embodiment, the low threshold and high threshold of the urban bare soil index are determined according to the distribution of the bare soil index in the bare soil area;

[0050] The 5th percentile of the numerical distribution of the Normalized Bare Soil Index is its low threshold, and the 95th percentile of the numerical distribution of the Normalized Bare Soil Index is its high threshold;

[0051] The 5th percentile of the numerical distribution of the Normalized Water Index is its low threshold, and the 95th percentile of the numerical distribution of the Normalized Water Index is its high threshold;

[0052] The 5th percentile of the numerical distribution of the Ratio Residential Land Index is its low threshold, and the 95th percentile of the numerical distribution of the Ratio Residential Land Index is its high threshold.

[0053] By calculating the Normalized Bare Soil Index, Normalized Water Index, and Ratio Residential Land Index in the bare soil area, and according to the distribution of the Normalized Bare Soil Index, Normalized Water Index, and Ratio Residential Land Index in the bare soil area, the low threshold and high threshold of the Normalized Bare Soil Index, the low threshold and high threshold of the Normalized Water Index, and the low threshold and high threshold of the Ratio Residential Land Index are calculated correspondingly, so as to quickly obtain reasonable low and high thresholds of the indices, and use the corresponding low and high thresholds of the indices as the judgment benchmark.

[0054] In this embodiment, if the bare soil index of the bare soil grid point is within the low threshold and high threshold of the urban bare soil index, then it is determined that the bare soil grid point is urban bare soil;

[0055] If the Normalized Bare Soil Index, Normalized Water Index, and Ratio Residential Land Index of the bare soil grid point are all within their corresponding low threshold and high threshold ranges, then it is determined that the bare soil grid point is urban bare soil.

[0056] The normalized bare soil index, normalized water body index, and ratio residential area index of several bare soil grid points are calculated, and the calculated results are compared with the low and high thresholds of the normalized bare soil index, the low and high thresholds of the normalized water body index, and the low and high thresholds of the ratio residential area index obtained above. If the normalized bare soil index, normalized water body index, and ratio residential area index of the current bare soil grid point are all within the low and high threshold ranges of the corresponding index, it is determined that the bare soil grid point is urban bare soil. The urban bare soil area can be quickly obtained, the calculation is fast, the method is simple, only relies on real-time remote sensing data, does not require big data modeling or building a dataset, and does not require a large computer for data storage and operation to realize satellite remote sensing identification of urban bare soil. Moreover, the normalized bare soil index, normalized water body index, and ratio residential area index are used as remote sensing indices for extracting urban bare soil. Especially the ratio residential area index specifically used for separating urban buildings greatly improves the accuracy of urban bare soil extraction and reduces the confusion between urban bare soil and urban buildings.

[0057] In this embodiment, it is determined that the bare soil grid point is urban bare soil, and the urban bare soil is marked separately to highlight the bare soil grid point. After determining the area where the bare soil grid point is urban bare soil, the corresponding urban bare soil is marked separately, which is convenient for identifying the area of urban bare soil and convenient for observation.

[0058] In this embodiment, the number of visually interpreted bare soil areas is at least ten, and the number of visually interpreted bare soil grid points is at least five thousand. Through at least ten bare soil areas, sufficient bare soil area samples are provided, which is convenient for calculating multiple samples of the normalized bare soil index (NDSI), normalized water body index (NDISI), and ratio residential area index (RII) of the bare soil area, so as to obtain reasonable low and high threshold values of the index, accurately judge subsequent urban bare soil, and improve the recognition accuracy of bare soil. At least five thousand bare soil grid points are convenient for fine segmentation of the experimental area to improve the resolution accuracy of bare soil.

[0059] In this embodiment, the cloud cover in the target urban area is less than 10% to ensure accurate identification results of bare soil in the urban area. To ensure data quality, images with a cloud cover below 10% are usually required. When the cloud cover in the urban area is below 10%, it has a positive impact on the quality of remote sensing images and the accuracy of information extraction. In this case, the cloud layer blocks less of the surface features, enabling the remote sensing image to more clearly reflect the actual situation of the surface. Therefore, selecting image data with a cloud cover below 10% for remote sensing analysis can improve the accuracy of information extraction, especially in applications that require detailed observation of surface features. In addition, images with low cloud cover are also helpful for temporal analysis to observe changes in the same area at different time points. This temporal requirement ensures that we can more accurately monitor vegetation growth, land use changes, natural disasters, etc. Therefore, remote sensing images with a cloud cover below 10% not only provide high-quality information in the current state but also contribute to long-term monitoring and analysis.

[0060] As described above, the above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A remote sensing identification method for urban bare soil, characterized in that: The following steps are involved: Acquire satellite remote sensing data of a target urban area, synthesize the remote sensing data into a true color image, divide the true color image into a number of bare soil grid points, and interpret the true color image into a number of bare soil areas by visual interpretation; Calculating a bare soil index of the bare soil area according to the remote sensing data, and determining a low threshold and a high threshold of the urban bare soil index according to the distribution of the bare soil index of the bare soil area; The bare soil index of the bare soil grid point is calculated according to the remote sensing data. If the bare soil index of the bare soil grid point is within the range of a low threshold and a high threshold of the urban bare soil index, the bare soil grid point is determined to be urban bare soil.

2. The remote sensing identification method of urban bare soil according to claim 1, characterized in that: The remote sensing data of the satellite includes blue band data (Blue), green band data (Green) and red band data (Red), and the blue band data, green band data and red band data are synthesized into a true color image.

3. The remote sensing identification method of urban bare soil according to claim 2, characterized in that: The remote sensing data of the satellite also includes near infrared band data (NIR), mid infrared band data (MIR) and far infrared band data (TIR); The bare soil index includes the normalized bare soil index (NDSI), the normalized difference water index (MNDWI) and the ratio settlement index (RII); Normalized Bare Soil Index The normalized bare soil index is used to enhance bare soil characteristics; The normalized water index The normalized water index is used to distinguish between bare soil areas and built-up areas in cities; The ratio of residential area index The ratio residential area index is used to enhance the distinction between bare land areas and urban built-up areas.

4. The remote sensing identification method of urban bare soil according to claim 3 is characterized in that: Determine a low threshold and a high threshold of the urban bare soil index according to the distribution of the bare soil index in the bare soil area; The 5th percentile of the numerical distribution of the normalized bare soil index is its lower threshold, and the 95th percentile of the numerical distribution of the normalized bare soil index is its upper threshold; The 5th percentile of the numerical distribution of the normalized water index is its lower threshold, and the 95th percentile of the numerical distribution of the normalized water index is its upper threshold; The 5th percentile of the numerical distribution of the ratio residential area index is its lower threshold, and the 95th percentile of the numerical distribution of the ratio residential area index is its upper threshold.

5. The remote sensing identification method of urban bare soil according to claim 4 is characterized in that: If the bare soil index of the bare soil grid point is within the range of the lower threshold and the upper threshold of the urban bare soil index, the bare soil grid point is determined to be urban bare soil; If the normalized bare soil index, the normalized water body index and the ratio residential area index of the bare soil grid point are all within the corresponding low threshold and high threshold ranges, the bare soil grid point is determined to be urban bare soil.

6. The remote sensing identification method of urban bare soil according to claim 5, characterized in that: The bare soil grid point is determined to be urban bare soil, and the urban bare soil is separately marked to highlight the bare soil grid point.

7. The remote sensing identification method of urban bare soil according to claim 6 is characterized in that: The number of bare soil areas for visual interpretation is at least ten, and the number of bare soil grid points for visual interpretation is at least five thousand.

8. The remote sensing identification method of urban bare soil according to claim 1, characterized in that: The cloud cover in the target urban area is less than 10% to ensure accurate bare soil identification results in the urban area.

Images