Soil and water conservation monitoring method and system for construction projects based on drone aerial photography

Abstract

The present invention relates to the technical field of soil and water conservation monitoring, and more specifically to a soil and water conservation monitoring method and system for construction projects based on drone aerial photography. The method comprises: step S1: photographing and correcting a first image, a first side image, and a second side image; step S2: performing texture mapping based on the first image, the first side image, and the second side image corresponding to the sub-region, and constructing a construction model; step S3: obtaining difference features based on the construction model and the landform model, and dividing multiple target object areas according to the difference features; step S4: calculating the parameter changes corresponding to the soil and water conservation factors according to the multiple target object areas, weighting them, and calculating the soil and water conservation data. The present invention solves the problem of insufficient accuracy of soil and water conservation data and improves the calculation accuracy of soil and water conservation data.

Description

Technical Field

[0001] The present invention relates to the technical field of soil and water conservation monitoring, and more specifically to a soil and water conservation monitoring method and system for construction projects based on unmanned aerial vehicle aerial photography. Background Art

[0002] With the modernization of society, especially the man-made soil erosion caused by road construction is inevitable. Therefore, how to minimize man-made soil erosion during road construction, or to accurately predict soil erosion and take measures to control soil erosion in advance, is a pain point problem that needs to be solved urgently in this field. Similar prior art includes a Chinese patent application with publication number CN115146209A, which proposes a method, system, storage medium and electronic device for monitoring soil and water conservation conditions, including: collecting multiple local images of a preset area through a drone; generating a DOM and a DEM of a preset area according to all local images; calculating the slope factor of the preset area according to the DEM of the preset area, and calculating the vegetation coverage of the preset area according to the DOM of the preset area; determining multiple soil and water conservation factors used to characterize the soil and water conservation conditions of the preset area according to the slope factor of the preset area and the vegetation coverage of the preset area, and using drones to meet the monitoring and evaluation of soil and water conservation conditions in a large area, providing a convenient, fast and efficient means of monitoring and evaluating regional ecological assessment indicators. Similar prior art also includes a Taiwan patent application with publication number TW202217492A, which discloses a method, system and application of drone slope soil-rock watershed image recognition, which includes an unmanned aircraft and an information processing unit. The unmanned aircraft is provided with an image capture device and a spraying device. The unmanned aircraft can fly to the slope and use an image acquisition device to acquire images of the slope to form an image of the slope. The information processing unit converts the slope image into a grayscale image, and performs a watershed algorithm on the grayscale image to cut it into multiple block images, and obtains the undulation distribution information of each block image of the slope, and then plans the flight path according to each undulation distribution information, controls the unmanned aircraft to fly over each block image in sequence according to the flight path, and determines the spraying amount of the vegetation substrate in each area of ​​the corresponding slope according to each undulation distribution information, and can perform image recognition of the watershed algorithm for slopes such as mountain fragmentation and collapse to determine the spraying thickness of the solidified colloid and the planted grass seeds. Although the above two patents have solved the problem of monitoring the soil and water conservation status, the monitoring accuracy is not high and cannot accurately reflect the status of soil and water conservation. Summary of the invention

[0003] In order to better solve the above problems, the present invention provides a soil and water conservation monitoring method for a construction project based on drone aerial photography, the method comprising the following steps:

[0004] Step S1: periodically acquiring a first image corresponding to each sub-area in the construction area and a first side image and a second side image of the sub-area at a set position by an aerial photography unit, and correcting the first image, the first side image and the second side image by a correction unit;

[0005] Step S2: performing texture mapping based on the first image, the first side image and the second side image corresponding to each of the sub-areas to construct a construction model of the construction area;

[0006] Step S3: obtaining difference features based on the construction model and the topographic model of the construction area before construction stored in the storage unit, and dividing the construction model into a plurality of target object areas corresponding to factors affecting soil and water conservation according to the difference features, wherein the factors affecting soil and water conservation include soil disturbance, vegetation change and soil pile volume;

[0007] Step S4: Calculate the parameter changes corresponding to the soil and water conservation factors according to the plurality of target object areas, perform weighting, and calculate the soil and water conservation data.

[0008] As a preferred technical solution, step S1 includes:

[0009] Step S11: dividing the construction area into a plurality of different sub-areas based on the plane layout, and photographing a plurality of first images of the construction area at a preset height at the center of the construction area, rotating periodically at a set angle, wherein two adjacent first images overlap;

[0010] Step S12: judging whether the sub-region needs to be photographed from the side according to the regional features of each sub-region, when the height difference between the highest position and the lowest position of the sub-region is greater than a preset difference, the sub-region needs to be photographed from the side, and obtaining a first side image and a second side image of the sub-region that needs to be photographed from the side, wherein the first side image and the second side image are relative side images;

[0011] Step S13: The correction unit performs correction based on the pixel grayscale information of the first image, the first side image, and the second side image.

[0012] As a preferred technical solution, step S13 includes:

[0013] Step S130: taking the first image, the first side image and the second side image as images to be corrected, and selecting a target object area from the images to be corrected, wherein the target object area is a terrain feature area or an area related to soil and water conservation;

[0014] Step S131: taking an area at a preset distance from the edge of the target object area as a transition area, and taking an area of ​​the transition area away from the target object area as an adjacent area, respectively obtaining a first grayscale distribution map of the target object area, a second grayscale distribution map of the transition area, and a third grayscale distribution map of the adjacent area, wherein the first grayscale distribution map, the second grayscale distribution map, and the third grayscale distribution map are respectively proportion maps of pixels with different grayscale values ​​in corresponding areas, wherein the third grayscale distribution map is a grayscale distribution map of each annular pixel band in the transition area that is preset and equidistant from the target object area;

[0015] Step S132: Based on the first grayscale distribution map, obtain the first maximum proportion grayscale value in the target object area, and based on the second grayscale distribution map, obtain the second maximum proportion grayscale value in the adjacent area, and calculate the third maximum proportion grayscale value in each annular pixel band in the transition area and the target object area. When the proportion of pixels corresponding to the first maximum proportion grayscale value is greater than or equal to a first threshold, convert the grayscale values ​​of the pixels in the target object area to the first maximum proportion grayscale value, and respectively obtain the first proportion corresponding to the first maximum proportion grayscale value and the second proportion corresponding to the second maximum proportion grayscale value in the third grayscale distribution map, and obtain the The third proportion corresponding to the first maximum proportion grayscale value in the first grayscale distribution map and the fourth proportion corresponding to the second maximum proportion grayscale value in the second grayscale distribution map are calculated, and the first difference between the first proportion and the third proportion and the second difference between the second proportion and the fourth proportion are calculated respectively. When the first difference is less than the second difference, and the first difference is less than the set threshold, the first maximum proportion grayscale value in the preset area in the transition area is retained, and the pixel grayscale values ​​in the preset area are replaced with the first maximum proportion grayscale value, and the target object area and the preset area in the transition area are used as new target object areas.

[0016] As a preferred technical solution, the step S132 further includes:

[0017] When the first maximum grayscale value proportion is less than the first threshold, multiple key grayscale values ​​are selected according to the characteristics of the target object area, and the reference proportion of each key grayscale value is obtained based on the first grayscale distribution map, the size of the preset area is adjusted and the measured proportions of multiple key grayscale values ​​in the preset area are calculated, and when the difference between the measured proportion of any key grayscale value and the corresponding reference proportion is less than or equal to the second threshold, the grayscale value of the pixel corresponding to the key grayscale value in the preset area is retained and the grayscale values ​​of other pixels other than the pixel corresponding to the key grayscale value are replaced with the minimum grayscale value in the second grayscale distribution map.

[0018] As a preferred technical solution, step S2 includes the following steps:

[0019] Step S21: establishing a coordinate system with the center position of the construction area as the origin, and mapping the first image, the first side image and the second side image corresponding to each sub-area to the coordinate system according to the position relationship;

[0020] Step S22: obtaining first overlapping areas of the first image corresponding to the second sub-area and the second overlapping areas of the third image corresponding to the adjacent first sub-area, respectively, and selecting a first reference pattern in the first target object area in the first overlapping areas and a second reference pattern in the second target object area in the second overlapping areas, respectively, and aligning the second sub-area with the first reference pattern in the first sub-area and the second reference pattern in the third sub-area, respectively;

[0021] Step S23: Select reference positions corresponding to the first lateral image and the second lateral image respectively from the same sub-region corresponding to the first lateral image and the corresponding second lateral image, and perform correction and alignment.

[0022] As a preferred technical solution, step S4 includes:

[0023] The parameter change corresponding to the soil and water conservation factor is calculated through the target object area, wherein the first target object area is obtained through the grayscale value of the image pixel corresponding to the construction model of the deep soil, and the soil disturbance amount is the area corresponding to the first target object area; the second target object area is also obtained through the grayscale value of the pixel corresponding to the construction model of the vegetation, and the difference between the pixel area in the second target object area and the pixel area of ​​the vegetation in the landform model is used as the vegetation loss amount, and the corresponding third target object area is obtained based on the unevenness of the pile soil and the grayscale of the image pixel corresponding to the deep soil, and the pile soil amount is obtained according to the pile soil volume corresponding to the third target object area, and the soil disturbance amount, the vegetation loss amount and the pile soil amount are normalized and weighted to obtain the soil and water conservation data.

[0024] As a preferred technical solution, the larger the soil and water conservation data is, the worse the soil and water conservation capacity is, and conversely, the smaller the soil and water conservation data is, the stronger the soil and water conservation capacity is.

[0025] As a preferred technical solution, the step S4 further includes a step S5:

[0026] The rainfall in the future time period of the construction area is obtained through the meteorological station, the third ratio of the rainfall data to the historical average rainfall data is calculated, and the product of the third ratio and the historical rainfall erosion factor stored in the storage unit is used as the current rainfall erosion factor. The slope length factor and the slope gradient factor are also obtained based on the terrain information in each of the sub-areas. The current vegetation coverage and management factor is obtained based on the second ratio of the vegetation pixel area before and after construction in each sub-area, and the second ratio is used to obtain the current vegetation coverage and management factor. The product of the soil and water conservation data and the historical soil and water conservation measures factor stored in the storage unit is used as the current soil and water conservation measures factor. The current rainfall erosion factor, the slope length factor, the slope factor, the current vegetation coverage and management factor, the current soil and water conservation measures factor and the soil erodible factor obtained through soil analysis corresponding to each of the sub-areas are input into the universal soil loss equation to obtain the soil loss amount of each of the sub-areas, and when the soil loss amount is greater than the set threshold, a warning message is sent to the management unit.

[0027] The present invention also provides a soil and water conservation monitoring system for construction projects based on drone aerial photography, the system is used to implement the above method, and the system includes:

[0028] an aerial photography unit, configured to periodically acquire a first image corresponding to each sub-area in the construction area and a first side image and a second side image of the sub-area at a set position, and correct the first image, the first side image and the second side image by a correction unit;

[0029] A model creation unit, configured to perform texture mapping based on the first image, the first side image, and the second side image corresponding to each of the sub-areas to construct a construction model of the construction area;

[0030] a comparison unit, configured to obtain difference features based on the construction model and the topographic model of the construction area before construction stored in the storage unit, and divide the construction model into a plurality of target object areas corresponding to factors affecting soil and water conservation according to the difference features, wherein the factors affecting soil and water conservation include soil disturbance, vegetation change and soil pile volume;

[0031] The calculation unit is used to calculate the parameter changes corresponding to the soil and water conservation factors according to the multiple target object areas, perform weighting, and calculate the soil and water conservation data.

[0032] The present invention also provides a computer storage medium, which stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute the above-mentioned soil and water conservation monitoring method for construction projects based on drone aerial photography.

[0033] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0034] The present invention divides the planar layout of the construction area corresponding to the road project into a plurality of different sub-areas, and at a preset height above the center position, uses an unmanned aerial photography unit to periodically rotate at a set angle to shoot a plurality of the first images of the construction area, and shoots the sub-area that needs to be photographed from the side according to the regional characteristics of each sub-area, so as to obtain a more comprehensive image of the construction area, and uses the first image, the first side image and the second side image corresponding to the sub-area as the images to be corrected, and selects the target object area from the images to be corrected, and judges whether the target object area is an area with a single gray value or an area with multiple gray values ​​according to the first maximum gray value in the target object area, and uses different The correction method not only obtains the accurate range of the above-mentioned target object area but also increases the contrast of the boundary area, thereby improving the accuracy of the soil and water conservation data, and also maps the corrected image to obtain the above-mentioned construction model, and obtains the difference characteristics based on the above-mentioned construction model and the landform model, and obtains the corresponding target object area based on the above-mentioned difference characteristics, and calculates the parameter change amount corresponding to the soil and water conservation factor based on the above-mentioned target object area for weighted acquisition of the above-mentioned soil and water conservation data, and also calculates the parameter change of the soil loss equation of the above-mentioned construction area after the construction of the construction area, and predicts the soil loss amount of each of the above-mentioned sub-areas in the future time period. Through the mutual cooperation between the above-mentioned technical solutions, the above-mentioned soil and water conservation data and soil loss amount can be accurately obtained, thereby guiding the user's soil and water conservation work. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of the soil and water conservation monitoring method for construction projects based on drone aerial photography of the present invention;

[0036] Figure 2 This is a structural diagram of the soil and water conservation monitoring system for construction projects based on drone aerial photography of the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] The present invention provides a soil and water conservation monitoring method for a construction project based on drone aerial photography, such as Figure 1 As shown, the method comprises the following steps:

[0039] Step S1: periodically acquiring a first image corresponding to each sub-area in the construction area and a second image at different angles at a set position by an aerial photography unit, and acquiring a corrected monitoring image by correcting the first image and the second image, wherein the first image and the second image are depth images;

[0040] Specifically, the construction area is divided into a plurality of different sub-areas with the center position of the construction area as the center of the circle, and a plurality of the first images of the construction area are photographed by a drone aerial photography unit at a preset height above the center position by periodically rotating at a set angle, thereby improving the shooting efficiency, and the images of adjacent sub-areas are partially overlapped. Whether the sub-area needs side shooting is also determined by the area characteristics of each sub-area. The area characteristics include whether the terrain of the sub-area is flat. When the terrain of the sub-area is relatively flat, side shooting is not required. When the terrain of the sub-area is uneven, in order to obtain more comprehensive terrain information, it is necessary to shoot the sub-area from the side, and obtain the first side image and the second side image, and the first side image and the second side image are opposite surfaces. The correction unit corrects the first image, the first side image and the second side image. Through the technical solution, a comprehensive terrain image of the construction area can be obtained, laying a foundation for obtaining accurate soil and water conservation data.

[0041] Step S2: performing texture mapping based on the positional relationship between a first monitoring image in the monitoring image and a second monitoring image overlapping with the first monitoring image, so as to construct a construction model of the construction area;

[0042] Specifically, a coordinate system is established with the center position of the above-mentioned construction area as the origin. Since each of the above-mentioned sub-areas is a sector with the center position of the above-mentioned construction area as the center, when establishing the above-mentioned construction model during construction in the above-mentioned construction area, it is necessary to map the above-mentioned first image, the above-mentioned first side image and the above-mentioned second side image corresponding to each of the above-mentioned sub-areas as texture images into the above-mentioned coordinate system, wherein the above-mentioned coordinate system is a three-dimensional coordinate system. Since there will be overlapping parts between adjacent above-mentioned texture images, in order to make the texture images corresponding to adjacent above-mentioned sub-areas more accurately fused, the first overlapping area between the third image of the above-mentioned second sub-area taken at the center position of the above-mentioned construction area and the second image corresponding to the adjacent first sub-area, and the first overlapping area with the adjacent above-mentioned third image are respectively obtained. The second overlapping area between the sub-areas corresponds to the fourth image. In order to make the third image of the above-mentioned second sub-area accurately merge with the third image of the above-mentioned first sub-area and the fourth image of the above-mentioned third sub-area, a first reference pattern is selected in the above-mentioned first overlapping area, and a second reference image is selected in the second overlapping area, and the above-mentioned second image is aligned with the above-mentioned third image and the fourth image respectively based on the above-mentioned first reference pattern and the above-mentioned second reference pattern. For the above-mentioned first side image and the above-mentioned second side image, at least three reference positions on the edge of the first image of the corresponding sub-area are selected for alignment, wherein the above-mentioned first image includes the above-mentioned second image, the above-mentioned third image and the above-mentioned fourth image. Through the above-mentioned technical solution, the above-mentioned construction model with precise mapping can be obtained.

[0043] Step S3: obtaining difference features based on the construction model and the topographic model of the construction area before construction stored in the storage unit, and dividing the construction model into a plurality of target object areas corresponding to factors affecting soil and water conservation according to the difference features, wherein the factors affecting soil and water conservation include soil disturbance, vegetation change and soil pile volume;

[0044] Specifically, in order to obtain the degree of decline in soil and water conservation capacity caused during the construction of the project, it is necessary to calculate the changes in the above-mentioned soil and water conservation related factors before and after construction. Therefore, the construction model during construction in the above-mentioned construction area is compared with the above-mentioned landform model before construction to obtain difference characteristics and the correspondence between the above-mentioned difference characteristics and the above-mentioned soil and water conservation related factors. Then, based on the characteristics of the above-mentioned soil and water conservation related factors, such as pixel gray value characteristics and flatness characteristics, the above-mentioned target object area corresponding to each of the above-mentioned soil and water conservation related factors is obtained, and finally the soil and water conservation data is accurately calculated through the above-mentioned soil and water conservation related factors.

[0045] Step S4: Calculate the parameter changes corresponding to the soil and water conservation factors according to the plurality of target object areas, perform weighting, and calculate the soil and water conservation data.

[0046] Specifically, due to the construction of construction projects, especially the highway construction process, soil disturbance, vegetation destruction and soil volume mainly affect soil erosion, that is, the parameters corresponding to the above-mentioned soil and water conservation factors, because the deep soil is exposed after soil disturbance, so that the pixel grayscale corresponding to the deep soil is displayed in the first target object area in the above-mentioned construction model, therefore, the pixel area of ​​the above-mentioned first target object area can be calculated, and the pixel area can be used as the soil disturbance amount, and the pixel area of ​​the above-mentioned second target object area can be calculated and used as the vegetation grayscale value in the above-mentioned construction model. The difference between the area of ​​vegetation pixels in the above-mentioned landform model and the area of ​​vegetation pixels in the above-mentioned landform model is taken as the above-mentioned vegetation loss amount. Since the soil pile is the soil turned up during construction, the soil color is the color of the deep soil. Due to the unevenness of the soil pile, the above-mentioned third target object area is obtained, and the above-mentioned soil pile amount is obtained through the volume of the above-mentioned third target object area. Through the above-mentioned technical scheme, the above-mentioned soil disturbance amount, the above-mentioned vegetation loss amount and the above-mentioned soil pile amount can be obtained based on the above-mentioned construction model, and the above-mentioned soil disturbance amount, the above-mentioned vegetation loss amount and the above-mentioned soil pile amount can be normalized and weighted to obtain accurate the above-mentioned soil and water conservation data.

[0047] Furthermore, the step S1 comprises:

[0048] Step S11: dividing the construction area into a plurality of different sub-areas based on the plane layout, and photographing a plurality of first images of the construction area at a preset height at the center of the construction area, rotating periodically at a set angle, wherein two adjacent first images overlap;

[0049] Specifically, the construction area is divided into a plurality of different sub-areas according to the plane layout, and at a preset height above the center position, a plurality of first images of the construction area are photographed by a drone aerial photography unit that periodically rotates at a set angle, thereby improving the shooting efficiency and partially overlapping adjacent sub-area images. The technical solution lays the foundation for constructing a construction model of the construction area using the first images.

[0050] Step S12: judging whether the sub-region needs to be photographed from the side according to the regional features of each sub-region, when the height difference between the highest position and the lowest position of the sub-region is greater than a preset difference, the sub-region needs to be photographed from the side, and obtaining a first side image and a second side image of the sub-region that needs to be photographed from the side, wherein the first side image and the second side image are relative side images;

[0051] Step S13: The correction unit performs correction based on the pixel grayscale information of the first image, the first side image, and the second side image.

[0052] Specifically, whether the sub-area needs to be photographed from the side is determined by the area features of each of the sub-areas, and the area features include whether the terrain of the sub-area is flat. When the terrain of the sub-area is relatively flat, side photography is not required. When the terrain of the sub-area is uneven, in order to obtain more comprehensive terrain information, it is necessary to photograph the sub-area from the side, and obtain the first side image and the second side image, and the first side image and the second side image are opposite surfaces, and the first image, the first side image and the second side image are corrected by the correction unit. Through the above technical solution, a comprehensive terrain image of the construction area can be obtained.

[0053] Furthermore, the step S13 includes:

[0054] Step S130: taking the first image, the first side image and the second side image as images to be corrected, and selecting a target object area from the images to be corrected, wherein the target object area is a terrain feature area or an area related to soil and water conservation;

[0055] Step S131: taking an area at a preset distance from the edge of the target object area as a transition area, and taking an area of ​​the transition area away from the target object area as an adjacent area, respectively obtaining a first grayscale distribution map of the target object area, a second grayscale distribution map of the transition area, and a third grayscale distribution map of the adjacent area, wherein the first grayscale distribution map, the second grayscale distribution map, and the third grayscale distribution map are respectively proportion maps of pixels with different grayscale values ​​in corresponding areas, wherein the third grayscale distribution map is a grayscale distribution map of each annular pixel band in the transition area that is preset and equidistant from the target object area;

[0056] Specifically, the first image, the first side image and the second side image corresponding to the sub-area are used as the images to be corrected, and the target object area is selected from the images to be corrected, and the target object area is a soil disturbance area, an earth pile area or a vegetation distribution area. When obtaining the target object area, the approximate range of the soil disturbance area, the earth pile area and the vegetation distribution area is obtained according to the grayscale value of the surface soil of the construction area, the grayscale value of the deep soil and the grayscale value distribution of the vegetation coverage area stored in the storage unit. However, for the junction of the target object area and other areas, the boundaries of each of the target object areas cannot be accurately obtained, and since there is a lot of dust in the construction area during construction, the image quality will be deteriorated. Therefore, by obtaining the first grayscale distribution map of each of the target object areas, the second grayscale distribution map of the adjacent area and the third grayscale distribution map of the transition area, not only the accurate boundaries of the target object areas can be further obtained, but also a higher-definition corrected image can be obtained, thereby obtaining accurate data on factors related to soil and water conservation.

[0057] Step S132: Based on the first grayscale distribution map, obtain the first maximum proportion grayscale value in the target object area, and based on the second grayscale distribution map, obtain the second maximum proportion grayscale value in the adjacent area, and calculate the third maximum proportion grayscale value in each annular pixel band in the transition area and the target object area. When the proportion of pixels corresponding to the first maximum proportion grayscale value is greater than or equal to a first threshold, convert the grayscale values ​​of the pixels in the target object area to the first maximum proportion grayscale value, and respectively obtain the first proportion corresponding to the first maximum proportion grayscale value and the second proportion corresponding to the second maximum proportion grayscale value in the third grayscale distribution map, and obtain the The third proportion corresponding to the first maximum proportion grayscale value in the first grayscale distribution map and the fourth proportion corresponding to the second maximum proportion grayscale value in the second grayscale distribution map are calculated, and the first difference between the first proportion and the third proportion and the second difference between the second proportion and the fourth proportion are calculated respectively. When the first difference is less than the second difference, and the first difference is less than the set threshold, the first maximum proportion grayscale value in the preset area in the transition area is retained, and the pixel grayscale values ​​in the preset area are replaced with the first maximum proportion grayscale value, and the target object area and the preset area in the transition area are used as new target object areas.

[0058] Specifically, according to the first maximum proportion gray value in the target object area, it is judged whether the target object area is an area with a single gray value or an area with multiple gray values. When the proportion of pixels corresponding to the first maximum proportion gray value is greater than or equal to the first threshold, that is, the number of pixels corresponding to the first maximum proportion gray value is large, for example, the proportion exceeds 95%. At this time, it is considered that the target object area is an area with a single gray value. Therefore, by converting all other gray values ​​in the target object area into the first maximum proportion gray value, and obtaining the first proportion and the second proportion of the first maximum proportion gray value and the second maximum proportion gray value in the transition area, and by comparing the first proportion with the above The first difference of the third proportion and the second difference between the second proportion and the fourth proportion are used to determine whether the preset area in the above-mentioned transition area belongs to the above-mentioned target object area. When the above-mentioned first difference is small and less than the above-mentioned set threshold, it means that the above-mentioned preset area is relatively close to the above-mentioned target object area. Therefore, the pixel grayscale values ​​in the above-mentioned preset area are replaced with the above-mentioned first maximum proportion grayscale value, wherein the above-mentioned preset area is the largest area that meets the above-mentioned conditions. Through the above-mentioned technical solution, the boundary of the above-mentioned target object area can be accurately obtained when the above-mentioned target object area is a single grayscale value, so as to obtain an accurate construction model based on the above-mentioned corrected image, thereby improving the accuracy of soil and water conservation related data.

[0059] Furthermore, the step S132 further includes:

[0060] When the first maximum grayscale value proportion is less than the first threshold, multiple key grayscale values ​​are selected according to the characteristics of the target object area, and the reference proportion of each key grayscale value is obtained based on the first grayscale distribution map, the size of the preset area is adjusted and the measured proportions of multiple key grayscale values ​​in the preset area are calculated, and in the maximum preset area where the difference between the measured proportion of any key grayscale value and the corresponding reference proportion is less than or equal to the second threshold, the grayscale values ​​of the pixels corresponding to the key grayscale values ​​in the preset area are retained and the grayscale values ​​of other pixels other than the pixels corresponding to the key grayscale values ​​are replaced with the minimum grayscale value in the second grayscale distribution map.

[0061] Specifically, when the proportion of the above-mentioned first maximum proportion grayscale value is less than the above-mentioned first threshold, it means that the above-mentioned target object area is not a single grayscale area, and therefore multiple key grayscale values ​​in the above-mentioned target object are obtained. When the above-mentioned target object is a vegetation coverage area, the above-mentioned multiple key grayscale values ​​can be grayscale values ​​corresponding to multiple vegetation colors at different growth and development stages. The size of the above-mentioned preset area is adjusted, and the proportion of each of the above-mentioned key grayscale values ​​in the above-mentioned preset area, that is, the above-mentioned measured proportion, is calculated, and the above-mentioned measured proportion is compared with the reference proportion corresponding to the above-mentioned key grayscale value in the above-mentioned target object area. When the difference between the reference proportion corresponding to any of the above-mentioned key grayscale values ​​and the measured proportion is less than or equal to the above-mentioned second threshold, it is considered that the above-mentioned preset area belongs to the above-mentioned target object area, wherein the above-mentioned preset area is the largest area in the above-mentioned transition area that meets the above-mentioned conditions. The pixels corresponding to each of the above-mentioned key grayscale values ​​are retained, and the remaining other pixels are replaced with the minimum grayscale value in the above-mentioned second grayscale distribution map. Through the above-mentioned technical solution, not only the boundary of the above-mentioned target object area with multiple grayscale values ​​is accurately obtained, but also the contrast between adjacent areas at the boundary of the above-mentioned target object area is increased.

[0062] Furthermore, the step S2 comprises the following steps:

[0063] Step S21: establishing a coordinate system with the center position of the construction area as the origin, and mapping the first image, the first side image and the second side image corresponding to each sub-area to the coordinate system according to the position relationship;

[0064] Step S22: obtaining first overlapping areas of the third image corresponding to the second sub-area and the second overlapping areas of the fourth image corresponding to the adjacent first sub-area, respectively, and selecting a first reference pattern in the first target object area in the first overlapping areas and a second reference pattern in the second target object area in the second overlapping areas, respectively, and aligning the second sub-area with the first reference pattern in the first sub-area and the second reference pattern in the third sub-area, respectively, wherein the first reference pattern is a triangle and the second reference pattern is a square;

[0065] Step S23: Select reference positions corresponding to at least three points on the edges of the first lateral image and the second lateral image from the same sub-area corresponding to the first lateral image and the corresponding second lateral image, and perform correction and alignment.

[0066] Specifically, a coordinate system is established with the center position of the above-mentioned construction area as the origin. Since each of the above-mentioned sub-areas is a sector with the center position of the above-mentioned construction area as the center, when establishing the above-mentioned construction model during construction in the above-mentioned construction area, it is necessary to map the above-mentioned first image, the above-mentioned first side image and the above-mentioned second side image corresponding to each of the above-mentioned sub-areas as texture images into the above-mentioned coordinate system, wherein the above-mentioned coordinate system is a three-dimensional coordinate system. Since there will be overlapping parts between adjacent above-mentioned texture images, in order to make the texture images corresponding to adjacent above-mentioned sub-areas more accurately fused, the first overlapping area between the third image of the above-mentioned second sub-area taken at the center position of the above-mentioned construction area and the second image corresponding to the adjacent first sub-area, and the first overlapping area with the adjacent above-mentioned third image are respectively obtained. The second overlapping area between the sub-areas corresponds to the fourth image. In order to make the third image of the above-mentioned second sub-area accurately merge with the third image of the above-mentioned first sub-area and the fourth image of the above-mentioned third sub-area, a first reference pattern is selected in the above-mentioned first overlapping area, and a second reference image is selected in the second overlapping area, and the above-mentioned second image is aligned with the above-mentioned third image and the fourth image respectively based on the above-mentioned first reference pattern and the above-mentioned second reference pattern. For the above-mentioned first side image and the above-mentioned second side image, at least three reference positions on the edge of the first image of the corresponding sub-area are selected for alignment, wherein the above-mentioned first image includes the above-mentioned second image, the above-mentioned third image and the above-mentioned fourth image. Through the above-mentioned technical solution, the above-mentioned construction model with precise mapping can be obtained.

[0067] Furthermore, the step S4 comprises:

[0068] The parameter change corresponding to the soil and water conservation factor is calculated through the target object area, wherein the first target object area is obtained through the grayscale value of the image pixel corresponding to the construction model of the deep soil, and the soil disturbance amount is the area corresponding to the first target object area; the second target object area is also obtained through the grayscale value of the pixel corresponding to the construction model of the vegetation, and the difference between the pixel area in the second target object area and the pixel area of ​​the vegetation in the landform model is used as the vegetation loss amount, and the corresponding third target object area is obtained based on the unevenness of the pile soil and the grayscale of the image pixel corresponding to the deep soil, and the pile soil amount is obtained according to the pile soil volume corresponding to the third target object area, and the soil disturbance amount, the vegetation loss amount and the pile soil amount are normalized and weighted to obtain the soil and water conservation data.

[0069] Specifically, due to the construction of construction projects, especially highway construction, soil erosion is mainly affected by soil disturbance, vegetation destruction and soil pile volume, that is, the parameters corresponding to the above-mentioned soil and water conservation factors, because the deep soil is exposed after soil disturbance, so that the pixel grayscale corresponding to the deep soil is displayed in the first target object area corresponding to the above-mentioned construction model. Therefore, the pixel area of ​​the above-mentioned first target object area can be calculated, and the above-mentioned pixel area can be used as the above-mentioned soil disturbance amount. The difference between the pixel area of ​​the above-mentioned second target object area and the vegetation pixel area in the above-mentioned landform model can be calculated through the vegetation grayscale value in the above-mentioned construction model and used as the above-mentioned vegetation loss amount. Since soil pile is carried out during construction, The turned-over soil has the color of the deep soil layer, and due to the unevenness of the soil pile, the third target object area is obtained, and the amount of soil pile is obtained through the volume of the third target object area, wherein the usage model and the landform model are respectively consistent with the actual terrain proportions of the construction area, and through the technical scheme, the soil disturbance amount, the vegetation loss amount and the amount of soil pile can be obtained based on the construction model, and the soil disturbance amount, the vegetation loss amount and the amount of soil pile are normalized and weighted, so as to obtain accurate soil and water conservation data, wherein the weight of the vegetation loss amount is the largest, the weight of the soil disturbance amount is the second, and the weight of the amount of soil pile is the smallest.

[0070] Furthermore, the larger the soil and water conservation data is, the worse the soil and water conservation capacity is, and conversely, the smaller the soil and water conservation data is, the stronger the soil and water conservation capacity is.

[0071] Furthermore, the step S4 further includes a step S5:

[0072] The rainfall in the future time period of the construction area is obtained through the meteorological station, and the third ratio of the rainfall data to the historical average rainfall data is calculated, and the product of the third ratio and the historical rainfall erosion factor stored in the storage unit is used as the current rainfall erosion factor. The slope length factor and the slope gradient factor are also obtained based on the terrain information in each of the sub-areas. The second ratio of the vegetation pixel area before and after construction in each sub-area is also based on the second ratio, and the product of the second ratio and the historical vegetation coverage and management factor stored in the storage unit is used as the current vegetation coverage and management factor. The product of the soil and water conservation data and the historical soil and water conservation measure factor stored in the storage unit is used as the current soil and water conservation measure factor. The current rainfall erosion factor, the slope length factor, the slope factor, the current vegetation coverage and management factor, the current soil and water conservation measure factor and the soil erodible factor obtained through soil analysis corresponding to each of the sub-areas are input into the universal soil loss equation to obtain the soil loss amount of each of the sub-areas, and when the soil loss amount is greater than the set threshold, a warning message is sent to the management unit.

[0073] Specifically, in order to monitor the amount of soil loss in the construction area in real time, the changes in the corresponding parameters of the above soil loss equation after the construction of the construction area are calculated, and the changed parameters are re-entered into the above soil loss equation to obtain the predicted value of the soil loss amount of each of the above sub-areas in the future time period, and when the above soil loss amount is greater than the set threshold, an early warning message is sent to the management unit, so that corresponding measures can be taken in time to prevent further deterioration of soil erosion.

[0074] The present invention also provides a soil and water conservation monitoring system for construction projects based on drone aerial photography, the system is used to implement the above method, such as Figure 2 As shown, the system comprises:

[0075] an aerial photography unit, configured to periodically acquire a first image corresponding to each sub-area in the construction area and a first side image and a second side image of the sub-area at a set position, and correct the first image, the first side image and the second side image by a correction unit;

[0076] A model creation unit, configured to perform texture mapping based on the first image, the first side image, and the second side image corresponding to each of the sub-areas to construct a construction model of the construction area;

[0077] a comparison unit, configured to obtain difference features based on the construction model and the topographic model of the construction area before construction stored in the storage unit, and divide the construction model into a plurality of target object areas corresponding to factors affecting soil and water conservation according to the difference features, wherein the factors affecting soil and water conservation include soil disturbance, vegetation change and soil pile volume;

[0078] The calculation unit is used to calculate the parameter changes corresponding to the soil and water conservation factors according to the multiple target object areas, perform weighting, and calculate the soil and water conservation data.

[0079] The present invention also provides a computer storage medium, which stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute the above-mentioned soil and water conservation monitoring method for construction projects based on drone aerial photography.

[0080] In summary, the present invention divides the planar layout of the construction area corresponding to the road project into multiple different sub-areas, and at a preset height above the center position, the drone aerial photography unit periodically rotates at a set angle to shoot multiple first images of the construction area, and shoots the sub-area that needs to be photographed from the side according to the regional characteristics of each sub-area, so as to obtain a more comprehensive image of the construction area, and uses the first image, the first side image and the second side image corresponding to the sub-area as the images to be corrected, and selects the target object area from the images to be corrected, and judges whether the target object area is an area with a single grayscale value or an area with multiple grayscale values ​​according to the first maximum proportion grayscale value in the target object area, and adopts By using different correction methods, not only the accurate range of the above-mentioned target object area is obtained, but also the contrast of the boundary area is increased, thereby improving the accuracy of the soil and water conservation data. The corrected image is mapped to obtain the above-mentioned construction model, and the difference features are obtained based on the above-mentioned construction model and the landform model, and the corresponding target object area is obtained based on the above-mentioned difference features, and the parameter change amount corresponding to the soil and water conservation factor is calculated based on the above-mentioned target object area for weighted acquisition of the above-mentioned soil and water conservation data, and the soil loss equation parameter change of the above-mentioned construction area after the construction of the construction area is calculated, and the soil loss amount of each of the above-mentioned sub-areas in the future time period is predicted. Through the mutual cooperation between the above-mentioned technical solutions, the above-mentioned soil and water conservation data and soil loss amount can be accurately obtained, thereby guiding the user's soil and water conservation work.

[0081] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0083] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A soil and water conservation monitoring method for construction projects based on drone aerial photography, characterized in that: The method comprises the following steps: Step S1: periodically acquiring a first image corresponding to each sub-area in the construction area and a first side image and a second side image of the sub-area at a set position by an aerial photography unit, and correcting the first image, the first side image and the second side image by a correction unit; Step S2: performing texture mapping based on the first image, the first side image and the second side image corresponding to each of the sub-areas to construct a construction model of the construction area; Step S3: obtaining difference features based on the construction model and the topographic model of the construction area before construction stored in the storage unit, and dividing the construction model into a plurality of target object areas corresponding to factors affecting soil and water conservation according to the difference features, wherein the soil and water conservation factors include soil disturbance, vegetation change and soil pile volume; Step S4: Calculating the parameter changes corresponding to the soil and water conservation factors according to the plurality of target object areas, weighting them, and calculating the soil and water conservation data; Wherein, the step S1 comprises: Step S11: dividing the construction area into a plurality of different sub-areas based on the plane layout, and photographing a plurality of first images of the construction area at a preset height above the center of the construction area, rotating periodically at a set angle, wherein two adjacent first images overlap; Step S12: judging whether the sub-region needs to be photographed from the side according to the regional features of each sub-region, when the height difference between the highest position and the lowest position of the sub-region is greater than a preset difference, the sub-region needs to be photographed from the side, and obtaining a first side image and a second side image of the sub-region that needs to be photographed from the side, wherein the first side image and the second side image are relative side images; Step S13: correcting the first image, the first side image and the second side image according to the grayscale distribution diagrams of the first image, the first side image and the second side image; The step S13 comprises: Step S131: taking the first image, the first side image and the second side image as images to be corrected, and selecting a target object area from the images to be corrected, wherein the target object area is a terrain feature area or an area related to soil and water conservation; Step S132: taking an area at a preset distance from the edge of the target object area as a transition area, and taking an area of ​​the transition area away from the target object area as an adjacent area, respectively obtaining a first grayscale distribution map of the target object area, a second grayscale distribution map of the transition area, and a third grayscale distribution map of the adjacent area, wherein the first grayscale distribution map, the second grayscale distribution map, and the third grayscale distribution map are respectively proportion maps of pixels with different grayscale values ​​in corresponding areas, wherein the third grayscale distribution map is a grayscale distribution map of each annular pixel band in the transition area that is preset and equidistant from the target object area; Step S133: Based on the first grayscale distribution map, obtain the first maximum proportion grayscale value in the target object area, and based on the second grayscale distribution map, obtain the second maximum proportion grayscale value in the adjacent area, and calculate the third maximum proportion grayscale value in each annular pixel band in the transition area and the target object area. When the proportion of pixels corresponding to the first maximum proportion grayscale value is greater than or equal to a first threshold, convert the grayscale values ​​of the pixels in the target object area to the first maximum proportion grayscale value, and respectively obtain the first proportion corresponding to the first maximum proportion grayscale value and the second proportion corresponding to the second maximum proportion grayscale value in the third grayscale distribution map, and obtain the The third proportion corresponding to the first maximum proportion grayscale value in the first grayscale distribution map and the fourth proportion corresponding to the second maximum proportion grayscale value in the second grayscale distribution map are calculated, and a first difference between the first proportion and the third proportion and a second difference between the second proportion and the fourth proportion are calculated respectively. When the first difference is less than the second difference, and the first difference is less than a set threshold, the first maximum proportion grayscale value in the preset area in the transition area is retained, and the pixel grayscale values ​​in the preset area are replaced with the first maximum proportion grayscale value, and the target object area and the preset area in the transition area are used as the new target object area; The step S132 further includes: When the first maximum grayscale value proportion is less than the first threshold, multiple key grayscale values ​​are selected according to the characteristics of the target object area, and the reference proportion of each key grayscale value is obtained based on the first grayscale distribution map, the size of the preset area is adjusted and the measured proportions of multiple key grayscale values ​​in the preset area are calculated, and when the difference between the measured proportion of any key grayscale value and the corresponding reference proportion is less than or equal to the second threshold, the grayscale value of the pixel corresponding to the key grayscale value in the preset area is retained and the grayscale values ​​of other pixels other than the pixel corresponding to the key grayscale value are replaced with the minimum grayscale value in the second grayscale distribution map, wherein the key grayscale value is the grayscale value corresponding to a variety of vegetation colors at different growth and development stages.

2. The method according to claim 1, characterized in that The step S2 comprises the following steps: Step S21: establishing a coordinate system with the center position of the construction area as the origin, and mapping the first image, the first side image and the second side image corresponding to each sub-area to the coordinate system according to the position relationship; Step S22: obtaining first overlapping areas of the first image corresponding to the second sub-area and the second overlapping areas of the third image corresponding to the adjacent first sub-area, respectively, and selecting a first reference pattern in the first target object area in the first overlapping areas and a second reference pattern in the second target object area in the second overlapping areas, respectively, and aligning the second sub-area with the first reference pattern in the first sub-area and the second reference pattern in the third sub-area, respectively; Step S23: Select reference positions corresponding to the first lateral image and the second lateral image respectively from the same sub-region corresponding to the first lateral image and the corresponding second lateral image, and perform correction and alignment.

3. The method according to claim 1, characterized in that The step S4 comprises: The parameter change corresponding to the soil and water conservation factor is calculated through the target object area, wherein the first target object area is obtained through the image pixel grayscale value corresponding to the deep soil in the construction model, and the soil disturbance amount is the area corresponding to the first target object area; the second target object area is also obtained through the pixel grayscale value corresponding to the vegetation in the construction model, and the difference between the pixel area in the second target object area and the vegetation pixel area in the landform model is used as the vegetation loss amount, and the corresponding third target object area is obtained based on the unevenness of the pile soil and the image pixel grayscale corresponding to the deep soil, and the pile soil amount is obtained according to the pile soil volume corresponding to the third target object area, and the soil disturbance amount, the vegetation loss amount and the pile soil amount are normalized and weighted to obtain the soil and water conservation data.

4. The method according to claim 1, characterized in that: The step S4 further includes a step S5: The rainfall in the future time period of the construction area is obtained through the meteorological station, and the third ratio of the rainfall data to the historical average rainfall data is calculated, and the product of the third ratio and the historical rainfall erosion factor stored in the storage unit is used as the current rainfall erosion factor. The slope length factor and the slope gradient factor are also obtained based on the terrain information in each of the sub-areas. The second ratio of the vegetation pixel area before and after construction in each sub-area is also based on the second ratio, and the product of the second ratio and the historical vegetation coverage and management factor stored in the storage unit is used as the current vegetation coverage and management factor. The product of the soil and water conservation data and the historical soil and water conservation measures factor stored in the storage unit is used as the current soil and water conservation measures factor. The current rainfall erosion factor, the slope length factor, the slope factor, the current vegetation coverage and management factor, the current soil and water conservation measures factor and the soil erodible factor obtained through soil analysis corresponding to each of the sub-areas are input into the universal soil loss equation to obtain the soil loss amount of each of the sub-areas, and when the soil loss amount is greater than the set threshold, a warning message is sent to the management unit.

5. The method according to claim 1, characterized in that The larger the soil and water conservation data is, the worse the soil and water conservation capacity is; conversely, the smaller the soil and water conservation data is, the stronger the soil and water conservation capacity is.

6. A soil and water conservation monitoring system for construction projects based on drone aerial photography, the system being used to implement the method according to any one of claims 1 to 5, characterized in that: The system comprises: an aerial photography unit, configured to periodically acquire a first image corresponding to each sub-area in the construction area and a first side image and a second side image of the sub-area at a set position, and correct the first image, the first side image and the second side image by a correction unit; a model creation unit, configured to construct a construction model of the construction area based on texture mapping of the first image, the first side image, and the second side image corresponding to each of the sub-areas; A comparison unit, configured to obtain difference characteristics between the construction model and the topographic model of the construction area before construction stored in the storage unit, and divide the construction model into a plurality of target object areas corresponding to factors affecting soil and water conservation according to the difference characteristics, wherein the soil and water conservation factors include soil disturbance, vegetation change and soil pile volume; The calculation unit is used to calculate the parameter changes corresponding to the soil and water conservation factors according to the multiple target object areas, perform weighting, and calculate the soil and water conservation data.

7. A computer storage medium, characterized in that: The storage medium stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute the soil and water conservation monitoring method for construction projects based on drone aerial photography as described in any one of claims 1 to 5.

Images