An intelligent extraction and evaluation method for soil and water conservation measures based on multispectral imager

Abstract

The present invention provides an intelligent extraction and evaluation method for soil and water conservation measures based on a multispectral imager, which obtains multispectral information of soil and water conservation measures through a multispectral imager, adopts an object-oriented method to comprehensively and accurately identify soil and water conservation measures, and classifies and maps them, and obtains the quantity and specification parameters of soil and water conservation measures through geographic information and remote sensing algorithms, and evaluates the completion and quality of soil and water conservation measures. The method integrates ground data collection, classification, mapping, parameter extraction, quantity accounting, quality assessment, etc. of soil and water conservation measures to form a set of intelligent methods, which can effectively reduce the time and subjective errors of manual extraction and evaluation of soil and water conservation measures, and improve the efficiency, accuracy and intelligent level of soil and water conservation measures extraction.

Description

Technical Field

[0001] The invention relates to the technical field of soil and water conservation information management, and specifically to an intelligent extraction and evaluation method for soil and water conservation measures based on a multi-spectral imager. Background Art

[0002] In the process of industrialization and urbanization in my country, the phenomenon of soil erosion caused and aggravated by human activities, mainly production and construction projects, is relatively common, and some even cause serious soil erosion hazards. At present, carrying out soil and water conservation supervision monitoring of production and construction projects has become a statutory and regular task for water administration departments at all levels. In addition, water administration departments at all levels have implemented a large number of comprehensive soil and water erosion control projects in order to control soil erosion. However, production and construction projects and soil and water erosion control projects are characterized by many points, wide areas, and large quantities. Traditional on-site monitoring methods have many disadvantages such as long cycles, limited accuracy, and terrain restrictions. The existing technologies are all artificial recognition of soil and water conservation measures for images, and there is no intelligent recognition of soil and water conservation measures for images. After recognition, it is impossible to quantitatively evaluate the implementation effect of soil and water conservation measures. Therefore, it is urgent to adopt information-based data collection and processing methods to improve work efficiency.

[0003] The inventors of this application have discovered through research during the process of realizing the present invention that multispectral imagers have the characteristics of high resolution, low cost, high efficiency, high timeliness, and are not affected by terrain. They can monitor the soil and water loss conditions of production and construction projects, the implementation of soil and water conservation measures, etc. in a timely manner, and provide technical support for soil and water conservation monitoring and supervision of production and construction projects, and the evaluation of soil and water loss control projects.

[0004] When identifying soil and water conservation measures for images acquired by multispectral imagers, object-oriented interpretation algorithms can avoid misclassification and omission caused by salt and pepper noise, different spectra of the same object, and different objects of the same spectrum compared to pixel-oriented methods. Object-oriented interpretation algorithms classify images based on characteristic parameters such as band information, shape characteristics, and texture characteristics. Object-oriented interpretation algorithms require more characteristic parameters to meet the classification accuracy requirements. Therefore, multispectral drones that integrate multispectral sensors with as many bands as possible (blue light, green light, red light, red edge, and near infrared, etc.) can obtain more band information of aerial objects, so as to more accurately obtain the number and characteristics of soil and water conservation measures for production and construction projects; then, based on the obtained soil and water conservation measures, the implementation degree and effect of soil and water conservation measures in the monitoring area can be quantitatively evaluated. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that the prior art cannot intelligently extract the quantity and characteristics of soil and water conservation measures, and proposes an intelligent extraction and evaluation method for soil and water conservation measures based on a multispectral imager. The multispectral imager can be carried out by aerial photography of the monitoring area by unmanned aerial vehicles, satellites or other aviation or spacecraft, and the multispectral orthophotos can be obtained by processing. The object-oriented method is used to obtain the quantity and specification parameters of soil and water conservation measures, and the completion status and quality status of soil and water conservation measures are evaluated, which can solve the above problems well.

[0006] An intelligent extraction and evaluation method for soil and water conservation measures based on a multispectral imager comprises the following steps:

[0007] Step 1: Use a drone equipped with a multispectral imager to take aerial photos of the soil and water conservation work area to obtain multispectral information on soil and water conservation measures;

[0008] Step 2: Use aerial or spacecraft image processing software to process the multispectral information of soil and water conservation measures obtained in step 1, generate a digital surface model (DSM) by analyzing the image through aerial photogrammetry, terrain modeling, and splicing, and perform geometric correction, spectral correction, image registration, and image fusion to generate a multispectral orthophoto map (DOM);

[0009] Step 3: Import the multispectral orthophoto DOM obtained in step 2 into an image processing software with object-oriented functions, use an object-oriented interpretation algorithm, set the optimal multi-scale segmentation parameters for segmentation, calculate the band information and shape characteristics of the segmented multispectral image spots, perform nearest neighbor supervised classification according to land use type, and extract soil and water conservation measures;

[0010] Step 4: Import the soil and water conservation measures extracted from the multispectral orthophoto DOM and the digital surface model DSM obtained in step 2 into the geographic information system or remote sensing image processing software, respectively extract the length, width, quantity, area, vegetation survival rate, preservation rate and coverage of different soil and water conservation measures obtained after supervised classification, obtain the elevation information of the digital surface model DSM in the geographic information system or remote sensing image processing software, calculate the slope, slope length, height, depth, thickness, and volume characteristics of the area where the soil and water conservation measures are located, and calculate the NDVI and vegetation coverage in the geographic information system or remote sensing image processing software;

[0011] Step 5: Compare the characteristics of various soil and water conservation measures extracted and calculated in step 4 with the soil and water conservation plan and design documents, and conduct quantitative and qualitative evaluation of the effects of soil and water conservation measures based on the Standard for Soil and Water Loss Prevention and Control of Production and Construction Projects (GB / T 50434-2018).

[0012] Furthermore, the aviation or spacecraft image processing software used in step 2 is Agisoft Photoscan Professional software.

[0013] Furthermore, the image processing software with object-oriented function used in step three is eCognition software.

[0014] Furthermore, the geographic information system or remote sensing image processing software used in step 4 is ArcGIS.

[0015] Furthermore, in step three, the band information of the multispectral image spots calculated includes R band, G band, B band, NIR band, normalized vegetation index NDVI, and vegetation coverage; the shape characteristics include aspect ratio; and the land use types include paddy fields, dry land, woodland, grassland, residential buildings, roads, bare land, and water areas.

[0016] Furthermore, the soil and water conservation measures in step three include engineering measures, plant measures and temporary measures.

[0017] Furthermore, engineering measures include drainage ditches, mortar-made stone slope protection with grass, mortar-made stone retaining walls, land reclamation, topsoil stripping, topsoil recovery, and terrace construction; plant measures include trees, shrubs, and grass planting; temporary measures include temporary drainage ditches, temporary siltation ponds, temporary barriers, temporary thatch covers, and temporary vehicle washing pools.

[0018] 10. Further, the soil and water conservation measures extracted in step 4 and their corresponding characteristics are shown in the following table:

[0019] Table 1

[0020]

[0021]

[0022] Furthermore, step five conducts quantitative evaluation on the effects of soil and water conservation measures, including:

[0023] (1) Evaluate the implementation of the slope-to-terrace system by comparing the terrace area and vegetation coverage with the implementation plan;

[0024] (2) The topsoil protection rate is evaluated by dividing the volume of topsoil protected by the temporary covering area and topography by the volume of topsoil stripped and stored;

[0025] Topsoil protection rate = volume of protected topsoil / volume of stripped and stored topsoil

[0026] (3) Evaluate the forest and grass vegetation recovery rate and forest and grass coverage index through the number, area, vegetation survival rate, vegetation preservation rate and vegetation coverage of afforestation and grass planting measures:

[0027] Forest and grass vegetation recovery rate = current afforestation and grass vegetation area / designed afforestation and grass vegetation area

[0028] Forest and grass coverage rate = current afforestation and grass planting area / planned afforestation and grass planting area

[0029] (4) The interception rate index is evaluated by dividing the actual amount of permanent waste and temporary soil piles obtained by the length of the retaining wall and the area of ​​the retaining wall by the total amount of permanent waste and temporary soil piles:

[0030] Slag interception rate = actual intercepted slag or soil pile / total slag or soil pile of the project

[0031] (5) The ratio of the sum of the engineering measures area, plant measures area and hardened area in the above table to the disturbed land area is obtained through drone photography, and the disturbed land remediation rate is calculated:

[0032] Disturbed land remediation rate = (engineering measures area + plant measures area + hardening area) / disturbed land area.

[0033] Furthermore, step five conducts a qualitative evaluation of the effects of soil and water conservation measures, specifically including: comparing the results obtained from the quantitative calculation with the design requirements, and those within the preset threshold of deviation are considered to meet the standard, and those exceeding the threshold are considered to fail to meet the standard.

[0034] Beneficial effects of the present invention: The present invention obtains multispectral information of soil and water conservation measures through a multispectral imager, and adopts an object-oriented method to classify soil and water conservation measures, which can accurately identify and extract soil and water conservation measures. By analyzing the information such as the quantity, specification and parameters of the extracted soil and water conservation measures, the implementation of soil and water conservation measures can be quantitatively and qualitatively evaluated. The present invention integrates ground data collection, classification extraction, parameter calculation, quality assessment, etc. of soil and water conservation measures to form a set of intelligent methods, which can effectively reduce the subjective errors and costs of manual monitoring, and improve the efficiency, accuracy and intelligence level of soil and water conservation measures extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of one embodiment of the method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager of the present invention;

[0036] Figure 2 is a schematic diagram of processing multispectral images of unmanned aerial vehicles to generate a digital surface model DSM in an embodiment of the present invention;

[0037] Figure 3 It is a schematic diagram of processing a multispectral image of a drone to generate a multispectral orthophoto map DOM in an embodiment of the present invention;

[0038] Figure 4It is a schematic diagram of the nearest neighbor supervised classification according to the land use types of construction land, terraced fields, thatched roofs, and vegetation in an embodiment of the present invention;

[0039] Figure 5 is a schematic diagram of the slope range of the area where the soil and water conservation measures are located using the slope tool in an embodiment of the present invention;

[0040] Figure 6 is a schematic diagram of calculating slope length in an embodiment of the present invention;

[0041] Figure 7 It is a schematic diagram of creating a new area field in the attribute table to measure the area of ​​the interpreted drainage ditch measures in an embodiment of the present invention;

[0042] Figure 8 is a schematic diagram of calculating the length and width of a drainage ditch in an embodiment of the present invention;

[0043] Fig. 9 is a schematic diagram of calculating the thatch covering area in an embodiment of the present invention;

[0044] Fig.10 It is an orthophoto map produced for afforestation and grass planting measures and slope-to-terracing soil and water conservation measures in an embodiment of the present invention;

[0045] Fig.11 It is a schematic diagram of using a grid calculator to calculate NDVI for the R band and NIR band of multispectral DOM in an embodiment of the present invention;

[0046] Fig.12 Schematic diagram of calculating vegetation coverage in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention.

[0048] See also Figure 1 One embodiment of the method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager of the present invention comprises the following steps:

[0049] Step 1: Use a multispectral imager on a drone to take aerial photos of the soil and water conservation work area to obtain multispectral information on soil and water conservation measures. Specifically, the route and altitude of the drone aerial photography are pre-set according to the soil and water conservation monitoring range, required accuracy and scale, and the multispectral imager on the drone is used to take aerial photos of the soil and water conservation monitoring area. Before taking off, the flight parameters need to meet the requirements of a heading overlap greater than 60% and a lateral overlap greater than 30%.

[0050] Step 2: Use the drone image processing software Agisoft Photoscan Professional to process the drone images, and generate a digital surface model (DSM) by performing aerial triangulation, terrain modeling, and splicing on the drone images. Figure 2 As shown in the figure), geometric correction, spectral correction, image registration and image fusion are performed to generate a multispectral orthophoto map DOM (as shown in the figure). Figure 3 shown).

[0051] Step three: Import the multispectral orthophoto DOM obtained in step two into the eCognition software, set the optimal scale parameter (Scale parameter), heterogeneity (Shape), and compactness (Compactness) multi-scale segmentation parameters for segmentation. The scale parameter of this embodiment is 50, the heterogeneity is 0.4, and the compactness is 0.5. For the segmented multispectral image spots, calculate the spectral characteristics (R band, G band, B band, NIR band, normalized vegetation index NDVI), shape characteristics (aspect ratio), and perform nearest neighbor supervised classification according to construction land, terraces, thatching, and vegetation land use types, and extract soil and water conservation measures (thatching, drainage ditches, slope conversion to terraces, afforestation and grass planting, etc., such as Figure 4 shown).

[0052] Step 4: Import the soil and water conservation measures obtained in step 3 and the digital surface model DSM obtained in step 2 into ArcGIS 10.6, and extract the quantity, specifications and parameters of different types of soil and water conservation measures. Obtain the elevation information of the digital surface model DSM in ArcGIS software. The elevation range of the monitoring area in this example is 390.7 to 567.1 meters (such as Figure 1 ); the slope tool in the 3D Analyst module is used to calculate the slope of the area where the soil and water conservation measures are located. The slope range of the monitoring area in this example is 0 to 87.4° (as shown in Figure 5 Calculate the slope length using the grid calculator (as shown in Figure 6 For the interpreted drainage measures, create an area field in the attribute table to measure the area (such as Figure 7 ), and create new line vectors to draw along the drain and perpendicular to the drain, and create new length fields to use the calculation geometry tool to calculate the length and width of the drain (as shown in Figure 8 For the interpreted covering measures, create an area field in the attribute table and use the Calculate Geometry tool to calculate the covering area (as shown in Fig. 9 shown);

[0053] For afforestation and grass planting and soil and water conservation measures such as slope conversion to terraces, orthophoto maps are produced (such as Fig.10The NDVI was calculated for the R band and NIR band of the multispectral DOM using the raster calculator in the Spatial Analyst tool of ArcGIS (as shown in Fig.11 As shown), the formula is as follows.

[0054]

[0055] The calculated NDVI is used to calculate vegetation coverage in ArcGIS Python interface according to the following formula (such as Fig.12 f is the vegetation coverage.

[0056]

[0057] In the formula, NDVI soil is the NDVI value of pure bare soil pixel, NDVI max It is the NDVI value under the condition of 100% vegetation coverage.

[0058] The characteristics of the soil and water conservation measures extracted and calculated in the above steps are shown in the following table:

[0059] Table 1

[0060]

[0061]

[0062] Step 5: Based on the number and specification parameters of various soil and water conservation measures extracted and calculated in step 4, and compared with the soil and water conservation plan and design documents, quantitative and qualitative evaluations are conducted on the soil and water conservation measures actually implemented in production and construction projects and soil and water loss control projects based on the Standard for Soil and Water Loss Prevention and Control of Production and Construction Projects (GB / T50434-2018). Specific examples are as follows:

[0063] (1) Evaluate the implementation of the slope-to-terrace system by comparing the terrace area and vegetation coverage with the implementation plan;

[0064] (2) The topsoil protection rate is evaluated by dividing the volume of topsoil protected by the temporary covering area and topography by the volume of topsoil stripped and stored.

[0065] Topsoil protection rate = volume of protected topsoil / volume of stripped and stored topsoil

[0066] (3) The forest and grass vegetation recovery rate and forest and grass coverage rate indicators are evaluated by the number, area, vegetation survival rate, preservation rate and coverage of afforestation and grass planting measures.

[0067] Forest and grass vegetation recovery rate = current afforestation and grass vegetation area / designed afforestation and grass vegetation area

[0068] Forest and grass coverage rate = current afforestation and grass planting area / planned afforestation and grass planting area

[0069] (4) The interception rate index is evaluated by dividing the actual amount of permanent waste and temporary soil piles obtained by the length of the retaining wall and the area of ​​the retaining wall by the total amount of permanent waste and temporary soil piles.

[0070] Slag interception rate = actual intercepted slag (earth pile) / total slag (earth pile) of the project

[0071] (5) The ratio of the sum of the engineering measures area, plant measures area and hardened area in Table 1 to the disturbed land area was obtained through drone photography, and the disturbed land remediation rate was calculated.

[0072] Disturbed land remediation rate = (engineering measure area + plant measure area + hardening area) / disturbed land area

[0073] The measured values ​​of the specification parameters and the quantitative evaluation results of the soil and water conservation measures in the examples of the present invention are shown in Table 2, and the qualitative evaluation results of the soil and water conservation measures are shown in Table 3:

[0074] Table 2

[0075]

[0076] Table 3

[0077] Soil and Water Conservation Measures Qualitative evaluation Topsoil protection rate Not up to standard Disturbance Land Remediation Rate Meet the Standard

[0078] The final evaluation results are met or not met. The calculated results can be compared with the required (design) results. If the deviation is within 30%, it is qualified, and if it exceeds, it is unqualified.

[0079] In the soil and water conservation measures in this example, the topsoil protection rate is the measured value of temporary thatching, which is 928.73, and the deviation is far greater than 30%, which does not meet the standard; the disturbed land remediation rate has a deviation of less than 30%, so it meets the standard.

[0080] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for intelligent extraction and evaluation of soil and water conservation measures based on multispectral imaging, characterized in that The steps include: Step 1: Use a drone equipped with a multispectral imager to take aerial photos of the soil and water conservation work area to obtain multispectral information on soil and water conservation measures; Step 2: Use aerial or spacecraft image processing software to process the multispectral information of soil and water conservation measures obtained in step 1, generate a digital surface model (DSM) by analyzing the image through aerial photogrammetry, terrain modeling, and splicing, and perform geometric correction, spectral correction, image registration, and image fusion to generate a multispectral orthophoto map (DOM); Step 3: Import the multispectral orthophoto DOM obtained in step 2 into an image processing software with object-oriented functions, use an object-oriented interpretation algorithm, set the optimal multi-scale segmentation parameters for segmentation, calculate the band information and shape characteristics of the segmented multispectral image spots, perform nearest neighbor supervised classification according to land use type, and extract soil and water conservation measures; Step 4: Import the soil and water conservation measures extracted from the multispectral orthophoto DOM and the digital surface model DSM obtained in step 2 into the geographic information system or remote sensing image processing software, respectively extract the length, width, quantity, area, vegetation survival rate, preservation rate and coverage of different soil and water conservation measures obtained after supervised classification, obtain the elevation information of the digital surface model DSM in the geographic information system or remote sensing image processing software, calculate the slope, slope length, height, depth, thickness, and volume characteristics of the area where the soil and water conservation measures are located, and calculate the NDVI and vegetation coverage in the geographic information system or remote sensing image processing software; Step 5: Compare the characteristics of various soil and water conservation measures extracted and calculated in step 4 with the soil and water conservation plan and design documents, and conduct quantitative and qualitative evaluation of the effects of soil and water conservation measures based on the Standard for Soil and Water Loss Prevention and Control of Production and Construction Projects (GB / T 50434-2018).

2. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 1, characterized in that: The aviation or spacecraft image processing software used in step 2 is Agisoft Photoscan Professional software.

3. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 1, characterized in that: The image processing software with object-oriented function used in step 3 is eCognition software.

4. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 1, characterized in that: The geographic information system or remote sensing image processing software used in step 4 is ArcGIS.

5. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 1, characterized in that: In step 3, the band information of the multispectral image patch calculation includes R band, G band, B band, NIR band, normalized vegetation index NDVI, and vegetation coverage; the shape characteristics include aspect ratio; the land use types include paddy fields, dry land, woodland, grassland, residential buildings, roads, bare land, and water areas.

6. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 1, characterized in that: The soil and water conservation measures in step three include engineering measures, plant measures and temporary measures.

7. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 6, characterized in that: Engineering measures include drainage ditches, mortar-stone and grass-planted slope protection, mortar-stone retaining walls, land reclamation, topsoil stripping, topsoil recovery, and terrace construction; plant measures include trees, shrubs, and grass planting; temporary measures include temporary drainage ditches, temporary siltation ponds, temporary barriers, temporary thatch covers, and temporary vehicle washing pools.

8. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 7, characterized in that: The soil and water conservation measures extracted in step 4 and their corresponding characteristics are shown in the following table: Table 1 。 9. The method for intelligent extraction and evaluation of soil and water conservation measures based on a multispectral imager according to claim 8, characterized in that: Step 5: Quantitative evaluation of the effects of soil and water conservation measures, including: (1) Evaluate the implementation of the slope-to-terrace system by comparing the terrace area and vegetation coverage with the implementation plan; (2) The topsoil protection rate is evaluated by dividing the volume of topsoil protected by the temporary covering area and topography by the volume of topsoil stripped and stored; Topsoil protection rate = volume of protected topsoil / volume of stripped and stored topsoil (3) Evaluate the forest and grass vegetation recovery rate and forest and grass coverage index through the number, area, vegetation survival rate, vegetation preservation rate and vegetation coverage of afforestation and grass planting measures: Forest and grass vegetation recovery rate = current afforestation and grass vegetation area / designed afforestation and grass vegetation area Forest and grass coverage rate = current afforestation and grass planting area / planned afforestation and grass planting area (4) The interception rate index is evaluated by dividing the actual amount of permanent waste and temporary soil piles obtained by the length of the retaining wall and the area of ​​the retaining wall by the total amount of permanent waste and temporary soil piles: Slag interception rate = actual intercepted slag or soil pile / total slag or soil pile of the project (5) The ratio of the sum of the engineering measures area, plant measures area and hardened area in the above table to the disturbed land area is obtained through drone photography, and the disturbed land remediation rate is calculated: Disturbed land remediation rate = (engineering measures area + plant measures area + hardening area) / disturbed land area.

10. The method for intelligent extraction and evaluation of soil and water conservation measures based on multispectral imager according to claim 9, characterized in that: Step five is to conduct a qualitative evaluation of the effects of soil and water conservation measures, specifically including: comparing the results obtained from quantitative calculations with design requirements, and those within the preset threshold are considered to meet the standards, and those exceeding the preset threshold are considered to fail to meet the standards.

Images