Water conservancy project topographic surveying and mapping method based on unmanned aerial vehicle remote sensing technology

Abstract

The invention relates to the technical field of image data processing, in particular to a water conservancy project topographic mapping method based on an unmanned aerial vehicle remote sensing technology, and the method comprises the steps: collecting the image information of a dam at a plurality of positions and a plurality of angles through an unmanned aerial vehicle, so as to obtain a plurality of initial dam images; determining a first target corner point of a hot spot area in the initial dam image, and determining a second target corner point of a floating object area in the initial dam image; removing the first target corner point and the second target corner point to obtain a processed target dam image; and splicing the plurality of target dam images to obtain and output a water conservancy project topographic map. According to the water conservancy project topographic mapping method based on the unmanned aerial vehicle remote sensing technology, the angular points of the hot spot area and the floating object area are accurately removed, in the image matching process, matching errors are reduced, the image splicing quality and accuracy are remarkably improved, and a more accurate and clearer water conservancy project topographic map is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a method for topographic mapping of water conservancy projects based on unmanned aerial vehicle remote sensing technology. Background Art

[0002] Mapping the topography of water conservancy projects is not only the basis for ensuring the quality of project design and construction, but also provides support for water resource management, environmental protection and disaster prevention. It is an important prerequisite and guarantee for the implementation of effective water conservancy projects. Traditional topographic mapping of water conservancy projects usually relies on manual measurement or aerial photography, which is not only time-consuming and labor-intensive, but also costly. The topographic mapping of water conservancy projects using drone remote sensing technology is based on taking dam images from multiple angles and multiple positions, thereby obtaining a large amount of high-resolution image data, and splicing the captured dam images to obtain a topographic mapping map of the water conservancy project, which improves mapping efficiency and data accuracy.

[0003] Due to the presence of floating objects on the water surface and the refraction of the water surface, ripples and bright spots will appear in the water area of ​​the image. When stitching together several dam images, directly using the detected corner points for matching, the movement direction of the floating objects themselves is relatively random. During the matching process of the corner points identified in its area, the parallax of the corresponding pixel positions in the local area will be abnormal. In addition, the corner points in the bright spot area of ​​the water surface are easily matched as strong feature points, which leads to errors in the matching results, and ultimately makes the topographic survey map of the water conservancy project inaccurate. Summary of the invention

[0004] In order to solve the technical problem of how to determine the bright spot area and floating objects on the water surface and eliminate them, the purpose of the present invention is to provide a water conservancy project topographic mapping method based on UAV remote sensing technology. The technical scheme adopted is as follows: The present invention provides a method for surveying and mapping water conservancy project terrain based on unmanned aerial vehicle remote sensing technology, the method comprising: Using drones to collect image information of the dam at multiple locations and angles to obtain multiple initial dam images; Determine a first target corner point of a bright spot area in the initial dam image, and determine a second target corner point of a floating object area in the initial dam image; removing the first target corner point and the second target corner point to obtain a processed target dam image; A plurality of target dam images are spliced ​​together to obtain and output a water conservancy project topographic map.

[0005] Furthermore, the step of determining a first target corner point of the bright spot area in the initial dam image includes: Determine an initial corner point image corresponding to the initial dam image; Determine the bright spot area of ​​the water surface according to the corner point shape of the corner points in the initial corner point image; The corner point corresponding to the bright spot area is determined as the first target corner point.

[0006] Furthermore, the step of determining the bright spot area of ​​the water surface according to the corner point morphology of the corner points in the initial corner point image includes: Selecting an input corner point in the initial corner point image and determining an edge line where the input corner point is located; Determine the gradient concentration value of the pixel point at the position corresponding to the corner point on the edge line where the input corner point is located; The gradient concentration value is compared with a preset judgment threshold to determine the bright spot area of ​​the water surface.

[0007] Furthermore, the step of determining the gradient concentration value of the pixel point at the position corresponding to the corner point on the edge line where the input corner point is located includes: Determine the input gradient value of the pixel point at the position corresponding to the input corner point; Determine the gradient mean of the pixel gradients at positions corresponding to all corner points on the edge line where the input corner point is located; A gradient concentration value is calculated based on the input gradient value and the gradient mean.

[0008] Furthermore, the step of determining a second target corner point of the floating object area in the initial dam image comprises: Determine an initial corner point image corresponding to the initial dam image; Determine the floating object area according to the morphological change information and relative position information of the corner points in the initial corner point image; The corner point corresponding to the floating object area is determined as the second target corner point.

[0009] Furthermore, the step of determining the floating object area according to the morphological change information and relative position information of the corner points in the initial corner point image includes: Determine any edge line of the non-bright spot area in the initial corner point image; Determine, from the arbitrary edge lines, target edge lines with similar morphology between the plurality of initial corner point images, wherein the morphological similarity indicates that the morphological similarity between the edge lines is greater than a preset similarity threshold; Based on the preset path of the drone, determine the potential edge line and the actual edge line corresponding to the target edge line at the corresponding acquisition time; wherein the potential edge line and the actual edge line are used to characterize the morphological change information and relative position information of the corner point; The potential edge line is compared with the actual edge line to determine the floating object area.

[0010] Furthermore, the step of determining target edge lines having similar shapes between the plurality of initial corner point images comprises: Determine a first gradient of a pixel point at a position corresponding to a first corner point of a first edge line in one of the initial corner point images, and determine a second gradient of a pixel point at a position corresponding to a second corner point of a second edge line in the other initial corner point image; Based on the first gradient, the second gradient, the first number of corner points, and the second number of corner points, target edge lines having similar shapes between the plurality of initial corner point images are determined.

[0011] Further, the step of comparing the potential edge line with the actual edge line to determine the floating object area includes: Determine the relative distance between the corresponding pixel points at the potential corner points on the potential edge line and the corresponding pixel points at the actual corner points on the actual edge line corresponding to the potential corner points; Determine the maximum value and the minimum value of the distance between all corner points on the potential edge line and the pixel points at corresponding positions on the actual edge line; Based on the relative distance value, the maximum distance value, the minimum distance value and the number of corner points of the potential edge line, an estimated value of the actual edge line belonging to the edge of the floating object is calculated; The estimated value is compared with a preset edge threshold to determine the floating object area.

[0012] Furthermore, the step of removing the first target corner point and the second target corner point to obtain a processed target dam image includes: By using a bilinear interpolation algorithm, the colors and textures of the bright spot area and the floating object area corresponding to the first target corner point and the second target corner point are smoothly transitioned with the surrounding water surface area, so as to remove the corner points of the bright spot area and the floating object area on the water surface in the initial dam image and obtain the processed target dam image.

[0013] Furthermore, the step of splicing a plurality of the target dam images to obtain and output a water conservancy project topographic map comprises: The target dam image is taken as input, the SIFT algorithm is used to match the corner points in the target dam image, and the homography matrix is ​​used to align and stitch the images to output the water conservancy project topographic map.

[0014] The present invention has the following beneficial effects: The method comprises the following steps: using a drone to collect image information of the dam at multiple positions and angles to obtain multiple initial dam images; determining the first target corner point of the bright spot area in the initial dam image, and determining the second target corner point of the floating object area in the initial dam image; removing the first target corner point and the second target corner point to obtain a processed target dam image; splicing multiple target dam images to obtain and output a water conservancy project topographic map. The present invention determines the bright spot area and the floating object area on the water surface and eliminates them, thereby achieving the purpose of accurately mapping the topographic map of the water conservancy project.

[0015] Specifically, the advantage of the embodiments of the present invention over the prior art is that in the process of dam image stitching, bright spots and floating areas on the water surface usually have high local contrast and instability, and may be mistaken for valid feature points, resulting in corner point matching errors. The bright spots on the water surface and the corner points of the floating area are accurately removed through the above steps, so that in the process of image matching, matching errors are reduced, the quality and accuracy of image stitching are significantly improved, and ultimately a more accurate and clear topographic map of water conservancy projects is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flowchart of the steps of a method for topographic mapping of water conservancy projects based on UAV remote sensing technology provided by one embodiment of the present invention; Figure 2 A schematic diagram of edge detection obtained by performing edge detection on a dam image provided by an embodiment of the present invention; Figure 3 A detailed flow chart of step S2 in a method for topographic mapping of water conservancy projects based on UAV remote sensing technology provided by one embodiment of the present invention; Figure 4 Another detailed flow chart of step S2 in a method for topographic mapping of water conservancy projects based on UAV remote sensing technology provided by one embodiment of the present invention; Figure 5 A schematic diagram of potential edge line prediction in a method for water conservancy project terrain mapping based on UAV remote sensing technology provided by an embodiment of the present invention; Figure 6It is a schematic diagram of the structure of the hardware operating environment of the water conservancy project topographic surveying and mapping equipment involved in the embodiment of the present invention; Figure 7 It is a schematic diagram of the framework structure of the water conservancy project topographic surveying and mapping device involved in the embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a water conservancy project topographic mapping method based on UAV remote sensing technology proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0020] The following is a detailed description of a method for topographic surveying and mapping a water conservancy project based on UAV remote sensing technology provided by the present invention in conjunction with the accompanying drawings.

[0021] Embodiment 1: For the water conservancy project topographic mapping method based on UAV remote sensing technology provided by the present invention, please refer to Figure 1 , which shows a flowchart of the steps of a method for topographic surveying and mapping of water conservancy projects based on UAV remote sensing technology provided by an embodiment of the present invention.

[0022] The method comprises: Step S1, using a drone to collect image information of the dam at multiple positions and multiple angles to obtain multiple initial dam images; In this embodiment, the area to be covered can be determined according to the size and characteristics of the dam, and the flight route of the drone can be set. According to the preset flight route of the drone, a high-resolution camera equipped with the drone can be used to obtain a number of initial dam images at different positions and angles at preset acquisition periods. The preset acquisition period here can be set and adjusted as needed, for example, 1 second. The high resolution can be 1080p, 2K, 4K, etc., which can be freely selected according to different needs and cost budgets.

[0023] It should be further explained that the multiple positions and multiple angles here may refer to different angles of the same position and different positions of the same angle.

[0024] Step S2, determining a first target corner point of a bright spot area in the initial dam image, and determining a second target corner point of a floating object area in the initial dam image; Please refer to Figure 2 .

[0025] After obtaining multiple initial dam images, Harris corner point detection can be performed on the collected initial dam images to obtain all corner points in the dam images. The edge detection image obtained by edge detection on the initial dam image is as follows: Figure 2 shown.

[0026] From all the corner points, first target corner points in the bright spot area and second target corner points in the floating object area can be distinguished based on different features of the corner points.

[0027] In a preferred embodiment, please refer to Figure 3 The step of determining a first target corner point of the bright spot area in the initial dam image comprises: Step S201, determining an initial corner point image corresponding to the initial dam image; Step S202, determining the bright spot area of ​​the water surface according to the corner point morphology of the corner points in the initial corner point image; Step S203: determine the corner point corresponding to the bright spot area as the first target corner point.

[0028] Harris corner point detection is performed on the initial dam image to obtain all corner points in the dam image and the corresponding initial corner point image.

[0029] In the process of mapping the topography of water conservancy projects based on UAV remote sensing technology, due to the reflection and refraction of the water surface, when the water surface is blown by a breeze or there are other external interferences, ripples are likely to appear on the water surface. These ripples cause the direction and reflection angle of the incident light to change, and an area with concentrated reflected light and high brightness appears at the crest of the wave, which causes bright spots in the water area. The bright spots have stronger brightness and edge features than other areas in the image, and with the change of the position of the light source, the position of the bright spots in several initial dam images taken is also prone to change. Therefore, when the dam area is obtained by splicing the dam images taken, it is easy to be matched as a strong feature point, resulting in more errors in the matching results. In this regard, the bright spot area of ​​the water surface can be determined based on the angle where the edge features of the bright spots are more obvious, and according to the shape of the corner points. At the same time, these corner points are distinguished and identified as the first target corner points corresponding to the bright spot area.

[0030] Through this embodiment, the corner point features of the bright spot area are analyzed and clarified, and based on these unique corner point features, the bright spot area of ​​the water surface can be identified and determined relatively efficiently and accurately.

[0031] Based on the above embodiment, more specifically, the step of determining the bright spot area of ​​the water surface according to the corner point morphology of the corner points in the initial corner point image includes: Selecting an input corner point in the initial corner point image and determining an edge line where the input corner point is located; Determine the gradient concentration value of the pixel point at the position corresponding to the corner point on the edge line where the input corner point is located; The gradient concentration value is compared with a preset judgment threshold to determine the bright spot area of ​​the water surface.

[0032] Furthermore, the ripple shape of the water surface is irregular and undulating, so the gradient around the corner points of the bright spot area varies greatly, and the gradient distribution is more dispersed; since the edges of the dam body and floating objects on the water surface are relatively regular, the gradient around the corner points of the dam body and floating objects on the water surface varies less, and the gradient distribution is more concentrated.

[0033] In the initial corner point image, any corner point can be selected as the input corner point for calculation, and the edge line where the corner point is located is obtained. The gradient concentration value of the pixel point at the corresponding position of the corner point on the edge line where the corner point is located is calculated, and compared with the preset judgment threshold (the preset judgment threshold can be set as needed). When it is less than the preset judgment threshold, it means that the gradient around the corner point changes greatly and the gradient distribution is more dispersed. The edge line where the corner point is located belongs to the bright spot area of ​​the water surface. Otherwise, it does not belong to the bright spot area, that is, the non-bright spot area, so as to accurately judge whether the area to which the corner point belongs to the bright spot area.

[0034] Based on the above embodiment, more specifically, the step of determining the gradient concentration value of the pixel point at the position corresponding to the corner point on the edge line where the input corner point is located includes: Determine the input gradient value of the pixel point at the position corresponding to the input corner point; Determine the gradient mean of the pixel gradients at positions corresponding to all corner points on the edge line where the input corner point is located; A gradient concentration value is calculated based on the input gradient value and the gradient mean.

[0035] First Corner points are represented as input corner points. The above steps of this embodiment can be specifically expressed as the following process: No. The calculation formula for the gradient concentration value of the pixel point corresponding to the corner point on the edge line where the corner point is located is: in, Indicates The gradient concentration value of the pixel point at the corresponding position of the corner point on the edge line where the corner point is located; Represents the mean value of the gradient of the pixel points corresponding to all corner points on the edge line; Indicates the edge line The input gradient value of the pixel point corresponding to the corner point; Indicates The number of corner points on the edge line where the corner point is located.

[0036] In the formula It represents the average value of the deviation between the gradient of the pixel points at the corresponding position of each corner point on the edge line and the gradient of the overall pixel points. The larger the average value, the closer the The more concentrated the gradient of the pixel point at the corresponding position of the corner point on the edge line where the corner point is located is, the less concentrated the gradient of the pixel point at the corresponding position of the corner point is. The larger the value, the The more concentrated the gradients of the pixels at the corresponding positions of the corner points on the edge line where the corner points are located are, the more concentrated the gradients of the pixels at the corresponding positions of the corner points are.

[0037] Since the edges of the dam body and floating objects on the water surface are relatively regular, the gradient changes around the corner points on the dam body and floating objects on the water surface are small, and the gradient distribution is relatively concentrated. Therefore, a threshold judgment is performed on the gradient concentration degree of the corner points on the edge line where each corner point is located obtained in the above steps.

[0038] The preset judgment threshold can be set as needed, for example, the preset judgment threshold , when the gradient concentration of the corner point on the edge line where the corner point is located Greater than or equal to the preset judgment threshold When the edge line where the corner point is located does not belong to the bright spot area on the water surface; when the gradient concentration degree of the corner point on the edge line where the corner point is located is Less than the preset judgment threshold , it indicates that the edge line where the corner point is located belongs to the bright spot area on the water surface.

[0039] Through the calculation method and steps provided in this embodiment, the bright spot area and the non-bright spot area of ​​the water surface can be identified and determined from all corner point ranges more accurately and efficiently at the pixel level.

[0040] In another preferred embodiment, please refer to Figure 4 The step of determining a second target corner point of the floating object area in the initial dam image comprises: Step S211, determining an initial corner point image corresponding to the initial dam image; Step S212, determining the floating object area according to the morphological change information and relative position information of the corner points in the initial corner point image; Step S213: determine the corner point corresponding to the floating object area as the second target corner point.

[0041] You can continue to refer to Figure 2 .

[0042] Similar to the above embodiment, a corresponding initial corner point image can be obtained based on the initial dam image.

[0043] Floating objects on the water surface are also easily matched as strong feature points. The floating objects on the water surface are similar in shape to the corner points on the dam. However, since the edge of the dam body is more regular and straight and the gradient change is smaller, the edge of the floating object is more angular and complex than the edge of the dam body, and the gradient change is relatively large. The shape of the floating objects in the several dam images taken will change significantly, but the change in the shape of the dam body is not obvious. In addition, due to the presence of wind blowing and causing the water surface to fluctuate, the position of the floating objects on the water surface changes, while the position of the dam body does not change due to external factors, so the relative position of the floating objects and the dam body in the several dam images taken will change. Therefore, the floating objects can be determined according to the relative position of the floating objects and the morphological similarity index, that is, the floating object area is determined according to the morphological change information and relative position information of the corner points in the initial corner point image, and then the floating object area on the water surface is eliminated. At the same time, these corner points are distinguished and identified as the second target corner points corresponding to the floating object area.

[0044] Through this embodiment, the corner point features and position features of the floating object area are analyzed and clarified. Based on these unique corner point features and position features, the floating object area on the water surface can be identified and determined relatively efficiently and accurately.

[0045] Based on the above embodiment, more specifically, the step of determining the floating object area according to the morphological change information and relative position information of the corner points in the initial corner point image includes: Determine any edge line of the non-bright spot area in the initial corner point image; Determine, from the arbitrary edge lines, target edge lines with similar morphology between the plurality of initial corner point images, wherein the morphological similarity indicates that the morphological similarity between the edge lines is greater than a preset similarity threshold; Based on the preset path of the drone, determine the potential edge line and the actual edge line corresponding to the target edge line at the corresponding acquisition time; wherein the potential edge line and the actual edge line are used to characterize the morphological change information and relative position information of the corner point; The potential edge line is compared with the actual edge line to determine the floating object area.

[0046] For this embodiment, after the bright spot area is identified and determined, the non-bright spot area in the initial corner point image can be determined. Specifically, this embodiment is to screen out edge lines with a larger gradient concentration corresponding to the non-bright spot area. Edge lines with a smaller gradient concentration belong to the bright spot area on the water surface, and edge lines with a larger gradient concentration may belong to the dam body or floating objects on the water surface, so further judgment is required. The movement direction of the floating object itself is relatively random, and the matching process of the corner points identified in its area will cause abnormalities in the parallax of the corresponding pixel position in the local area, resulting in errors in the matching results.

[0047] Taking any edge line in the non-bright spot area of ​​one of the initial corner point images among multiple images as an example, the edge line is compared with the edge lines in other initial corner point images. Generally, the comparison can be made in pairs, that is, two images that are adjacent to each other in acquisition time can be compared, and a certain similarity threshold is set as needed. Based on the morphological similarity of the two edge lines being greater than the preset similarity threshold, the target edge lines with similar morphology between the multiple initial corner point images can be found and determined, which can be considered as the presentation of the same edge line in different images.

[0048] UAVs generally fly along a pre-set path. Therefore, the potential edge line of the target edge line in the dam image corresponding to the current moment can be predicted and obtained based on the flight distance between the UAVs at different moments. To understand this process, please refer to Figure 5 , Figure 5A schematic diagram of potential edge line prediction in a method for topographic mapping of water conservancy projects based on UAV remote sensing technology provided by an embodiment of the present invention; as shown in the figure, the current edge line in the image collected by the UAV at the current moment can be seen. Based on a specific flight route and distance, the position of the potential edge line in the image at the next moment can be predicted. Here, the edge line is predicted to change from the lower side of the image to the upper side. In fact, it can be understood that the change of the physical position of the UAV is reflected and predicted in the change of the edge line in the image; since the morphology of the floating objects and the dam body in the image taken at intervals of one acquisition cycle will not change much, the potential edge line and the actual edge line can be compared and calculated, and the area to which it belongs can be determined according to the relative distance between the potential edge line and the actual edge line actually photographed in the image at the next moment and the length of the edge line. Specifically, the uniformity of the distance between the pixel points at the same corner point corresponding to the potential and actual edge lines can be used. The larger the value, the worse the uniformity of the distance between the pixel points at the corner point corresponding to the potential edge line and the pixel points at the corner point corresponding to the edge line actually photographed in the image, and the greater the possibility that the edge line belongs to the edge of the floating object. The smaller the value of the edge line length is, the smaller the length of the edge line is, and the greater the possibility that the edge line belongs to the edge of the floating object is. Thus, the possibility that the target edge line belongs to the edge of the floating object is comprehensively judged, and then the floating object area is determined.

[0049] Through this embodiment, based on the idea and principle that the morphological change information and relative position information of floating objects and dam bodies are significantly different, starting from the image parameter comparison level of potential edge lines and actual edge lines of the same edge line, the floating object area on the water surface can be identified and determined more efficiently and accurately.

[0050] Based on the above embodiment, more specifically, the step of determining target edge lines with similar shapes between a plurality of the initial corner point images includes: Determine a first gradient of a pixel point at a position corresponding to a first corner point of a first edge line in one of the initial corner point images, and determine a second gradient of a pixel point at a position corresponding to a second corner point of a second edge line in the other initial corner point image; Based on the first gradient, the second gradient, the first number of corner points, and the second number of corner points, target edge lines having similar shapes between the plurality of initial corner point images are determined.

[0051] In each of the initial dam images and the corresponding corner point images, based on the edge lines with large gradient concentration in each of the dam images and the corner point images, the edge lines with similar morphological features in each image are taken as the same edge line. The morphological similarity of each edge line in any two different images is calculated. Here, any two images with a time relationship of image acquisition can be compared.

[0052] The first The first edge line (corresponding to the first edge line mentioned above for ease of description and distinction) and the The calculation formula for the morphological similarity of the edge line (corresponding to the second edge line mentioned above for ease of description and distinction) is: in, Indicates the first The edge line and The morphological similarity of the edge lines (degree); Indicates that one of the images On the edge line The first gradient of the pixel point at the position corresponding to the first corner point (corresponding to the first corner point mentioned above for ease of description and distinction); Indicates the first On the edge line The second gradient of the pixel point at the position corresponding to the corner point (corresponding to the second corner point for ease of description and distinction); Indicates The number of first corner points of the corner points on the edge line; Indicates The number of second corner points of the corner points on the edge line; represents an exponential function with a natural constant as base; represents the absolute value function.

[0053] In the formula Indicates that one of the images The mean value of the gradient of the pixel points corresponding to all corner points on the edge line. Indicates the first The smaller the difference between the two means, the more similar the shapes of the two edge lines are. The larger the value of The edge line and The more similar the shapes of the edge lines are.

[0054] Here you can set a preset similarity threshold , when the first The edge line and The morphological similarity of the edge lines is greater than the similarity threshold When the first The edge line and edge lines are recorded as the same edge line; when the first edge line in two different images The edge line and The morphological similarity of the edge lines is less than or equal to the similarity threshold When the first The edge line and The edge lines are recorded as different edge lines.

[0055] Based on the above embodiment, more specifically, the step of comparing the potential edge line with the actual edge line to determine the floating object area includes: Determine the relative distance between the corresponding pixel points at the potential corner points on the potential edge line and the corresponding pixel points at the actual corner points on the actual edge line corresponding to the potential corner points; Determine the maximum value and the minimum value of the distance between all corner points on the potential edge line and the pixel points at corresponding positions on the actual edge line; Based on the relative distance value, the maximum distance value, the minimum distance value and the number of corner points of the potential edge line, an estimated value of the actual edge line belonging to the edge of the floating object is calculated; The estimated value is compared with a preset edge threshold to determine the floating object area.

[0056] No. The edge line represents the target edge line, and the calculation formula for the possibility of belonging to the edge of the floating object is: in, Indicates The probability that the edge line belongs to the edge of the floating object (estimated value); Indicates The potential edge line The pixel point corresponding to the corner point is the same as the pixel point on the actual edge line. The relative distance value of the pixel points at the corresponding positions of the corner points; Indicates The maximum distance between all corner points on the potential edge line and the corresponding pixel points on the actual edge line; Indicates The minimum distance between all corner points on the potential edge line and the corresponding pixel points on the actual edge line; Indicates The number of corner points on the potential edge line and the actual edge line (the number of corner points on the two is the same, so take any one of the edge lines); Represents a linear normalization function that normalizes the data value to Within the range.

[0057] in the formula Indicates the uniformity of the distance between the pixel points at the corresponding positions of the corner points on the potential edge line and the pixel points at the corresponding positions of the corner points on the edge line actually captured in the image. The larger the value, the worse the uniformity of the distance between the pixel points at the corresponding positions of the corner points on the potential edge line and the pixel points at the corresponding positions of the corner points on the edge line actually captured in the image, and the greater the possibility that the edge line belongs to the edge of a floating object. 1 / The larger the value is, the smaller the length of the edge line is, and the greater the possibility that the edge line belongs to the edge of a floating object.

[0058] Here the preset edge threshold can be set as needed. You can set the preset edge threshold , when The probability that the edge line belongs to the edge of the floating object is greater than or equal to the edge threshold When The first edge line is recorded as the edge of the float; The probability that the edge line belongs to the edge of the floating object is less than the edge threshold When The edge line is recorded as the edge of the dam body.

[0059] According to the above steps, the edge of floating objects in each image is obtained, and the floating object area on the water surface is obtained using the contour detection algorithm.

[0060] Step S3, removing the first target corner point and the second target corner point to obtain a processed target dam image; Based on the initial dam image and the corresponding corner point image, the first target corner point and the second target corner point are removed to obtain processed target dam images.

[0061] In a preferred embodiment, step S3 specifically includes: By using a bilinear interpolation algorithm, the colors and textures of the bright spot area and the floating object area corresponding to the first target corner point and the second target corner point are smoothly transitioned with the surrounding water surface area, so as to remove the corner points of the bright spot area and the floating object area on the water surface in the initial dam image and obtain the processed target dam image.

[0062] That is, after obtaining the bright spots and floating object areas according to the above embodiment, a bilinear interpolation algorithm is applied to the corner point areas in the bright spots and floating object areas to smoothly transition the color and texture of the bright spots and floating object areas with the surrounding water surface areas, thereby relatively completely and effectively removing the corner points of the bright spots and floating object areas on the water surface.

[0063] Step S4, stitching the multiple target dam images to obtain and output a water conservancy project topographic map.

[0064] Based on the splicing of each target dam image, a complete water conservancy project topographic map is obtained and output.

[0065] In a preferred embodiment, step S4 specifically includes: The target dam image is taken as input, the SIFT algorithm is used to match the corner points in the target dam image, and the homography matrix is ​​used to align and stitch the images to output the water conservancy project topographic map.

[0066] That is, after removing bright spots on the water surface and corner points of floating objects in several acquired dam images, the processed dam images are used as input, and the SIFT (Scale-invariant feature transform) feature descriptor is used to match the corner points in the processed dam images, and the homography matrix is ​​calculated for image alignment. The images are then precisely spliced ​​to output an accurate and complete water conservancy project topographic map, thereby completing the surveying and mapping of the water conservancy project topography.

[0067] The advantage of the embodiments of the present invention over the prior art is that during the dam image stitching process, the bright spots and floating areas on the water surface usually have high local contrast and instability, which may be mistaken for valid feature points, resulting in corner point matching errors. By accurately removing the bright spots on the water surface and the corner points of the floating area, the matching errors are reduced during the image matching process, the quality and accuracy of image stitching are significantly improved, and ultimately a more accurate and clear topographic map of the water conservancy project is obtained.

[0068] Embodiment 2: The embodiment of the present invention also provides a water conservancy project topography mapping device based on UAV remote sensing technology. The water conservancy project topography mapping device can be a data processing terminal such as a UAV, a personal computer, a workstation, a server, etc., which is not limited here.

[0069] like Figure 6 As shown, Figure 6 It is a structural schematic diagram of the hardware operating environment of the water conservancy project topographic surveying and mapping equipment involved in the embodiment of the present invention.

[0070] like Figure 6As shown, the water conservancy project topographic mapping equipment may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display (Display), an input unit such as a control panel, and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WIFI interface). The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001. The memory 1005 as a computer storage medium may include a water conservancy project topographic mapping program.

[0071] Those skilled in the art will understand that Figure 6 The hardware structure shown in the figure does not constitute a limitation of the device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0072] Continue to refer to Figure 6 , Figure 6 The memory 1005 as a computer-readable storage medium may include an operating system, a user interface module, a network communication module, and a water conservancy project topographic surveying and mapping program.

[0073] exist Figure 6 In the embodiment, the network communication module is mainly used to connect to the server and can communicate data with the server; and the processor 1001 can call the water conservancy project topographic mapping program stored in the memory 1005 and execute the steps in the above embodiments.

[0074] The hardware structure of the water conservancy project topographic mapping equipment described above is used to implement various embodiments of the water conservancy project topographic mapping based on UAV remote sensing technology of the present invention.

[0075] In addition, the present invention also provides a water conservancy project topographic mapping device based on UAV remote sensing technology, please refer to Figure 7 , the water conservancy project topographic surveying and mapping device comprises: An image acquisition module A10 is used to acquire image information of the dam at multiple positions and multiple angles using a drone to obtain multiple initial dam images; A corner point recognition module A20 is used to determine a first target corner point of a bright spot area in the initial dam image, and to determine a second target corner point of a floating object area in the initial dam image; The image processing module A30 is used to remove the first target corner point and the second target corner point to obtain a processed target dam image; and to splice a plurality of the target dam images to obtain and output a water conservancy project topographic map.

[0076] Furthermore, the corner point recognition module A20 is further used for: Determine an initial corner point image corresponding to the initial dam image; Determine the bright spot area of ​​the water surface according to the corner point shape of the corner points in the initial corner point image; The corner point corresponding to the bright spot area is determined as the first target corner point.

[0077] Furthermore, the corner point recognition module A20 is further used for: Selecting an input corner point in the initial corner point image and determining an edge line where the input corner point is located; Determine the gradient concentration value of the pixel point at the position corresponding to the corner point on the edge line where the input corner point is located; The gradient concentration value is compared with a preset judgment threshold to determine the bright spot area of ​​the water surface.

[0078] Furthermore, the corner point recognition module A20 is further used for: Determine the input gradient value of the pixel point at the position corresponding to the input corner point; Determine the gradient mean of the pixel point gradients at positions corresponding to all corner points on the edge line where the input corner point is located; A gradient concentration value is calculated based on the input gradient value and the gradient mean.

[0079] Furthermore, the corner point recognition module A20 is further used for: Determine an initial corner point image corresponding to the initial dam image; Determine the floating object area according to the morphological change information and relative position information of the corner points in the initial corner point image; The corner point corresponding to the floating object area is determined as the second target corner point.

[0080] Furthermore, the corner point recognition module A20 is further used for: Determine any edge line of the non-bright spot area in the initial corner point image; Determine, from the arbitrary edge lines, target edge lines with similar morphology between the plurality of initial corner point images, wherein the morphological similarity indicates that the morphological similarity between the edge lines is greater than a preset similarity threshold; Based on the preset path of the drone, determine the potential edge line and the actual edge line corresponding to the target edge line at the corresponding acquisition time; wherein the potential edge line and the actual edge line are used to characterize the morphological change information and relative position information of the corner point; The potential edge line is compared with the actual edge line to determine the floating object area.

[0081] Furthermore, the corner point recognition module A20 is further used for: Determine a first gradient of a pixel point at a position corresponding to a first corner point of a first edge line in one of the initial corner point images, and determine a second gradient of a pixel point at a position corresponding to a second corner point of a second edge line in the other initial corner point image; Based on the first gradient, the second gradient, the first number of corner points, and the second number of corner points, target edge lines having similar shapes between the plurality of initial corner point images are determined.

[0082] Furthermore, the corner point recognition module A20 is further used for: Determine the relative distance between the corresponding pixel points at the potential corner points on the potential edge line and the corresponding pixel points at the actual corner points on the actual edge line corresponding to the potential corner points; Determine the maximum value and the minimum value of the distance between all corner points on the potential edge line and the pixel points at corresponding positions on the actual edge line; Based on the relative distance value, the maximum distance value, the minimum distance value and the number of corner points of the potential edge line, an estimated value of the actual edge line belonging to the edge of the floating object is calculated; The estimated value is compared with a preset edge threshold to determine the floating object area.

[0083] Furthermore, the image processing module A30 is also used for: By using a bilinear interpolation algorithm, the colors and textures of the bright spot area and the floating object area corresponding to the first target corner point and the second target corner point are smoothly transitioned with the surrounding water surface area, so as to remove the corner points of the bright spot area and the floating object area on the water surface in the initial dam image and obtain the processed target dam image.

[0084] Furthermore, the image processing module A30 is also used for: The target dam image is taken as input, the SIFT algorithm is used to match the corner points in the target dam image, and the homography matrix is ​​used to align and stitch the images to output the water conservancy project topographic map.

[0085] The specific implementation of the water conservancy project topography mapping device of the present invention is basically the same as the various embodiments of the water conservancy project topography mapping method described above, and will not be repeated here.

[0086] In addition, the present invention also provides a computer-readable storage medium. The computer-readable storage medium of the present invention stores a water conservancy project topography mapping program, wherein when the water conservancy project topography mapping program is executed by a processor, the steps of the water conservancy project topography mapping method as described above are implemented.

[0087] Among them, the method implemented when the water conservancy project topography mapping program is executed can refer to the various embodiments of the water conservancy project topography mapping method of the present invention, and will not be repeated here.

[0088] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0090] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0091] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0092] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0093] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0094] The above description is only a preferred embodiment of the present invention, and does not limit the scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are included in the protection scope of the present invention.

Claims

1. A method for topographic mapping of water conservancy projects based on UAV remote sensing technology, characterized in that: The method comprises: Using drones to collect image information of the dam at multiple locations and angles to obtain multiple initial dam images; Determine a first target corner point of a bright spot area in the initial dam image, and determine a second target corner point of a floating object area in the initial dam image; removing the first target corner point and the second target corner point to obtain a processed target dam image; splicing a plurality of the target dam images to obtain and output a water conservancy project topographic map; The step of determining a first target corner point of the bright spot area in the initial dam image comprises: Determine an initial corner point image corresponding to the initial dam image; Determining the bright spot area of ​​the water surface according to the corner point morphology of the corner point in the initial corner point image, including: selecting an input corner point in the initial corner point image, determining the edge line where the input corner point is located; determining the gradient concentration value of the pixel point at the position corresponding to the corner point on the edge line where the input corner point is located; comparing the gradient concentration value with a preset judgment threshold to determine the bright spot area of ​​the water surface; Determine the corner point corresponding to the bright spot area as the first target corner point; The step of determining a second target corner point of the floating object area in the initial dam image comprises: Determining the floating object area according to the morphological change information and relative position information of the corner points in the initial corner point image, including: determining any edge line of the non-bright spot area in the initial corner point image; determining, from the arbitrary edge lines, a plurality of target edge lines with similar morphology between the initial corner point images, wherein the morphological similarity represents that the morphological similarity between the edge lines is greater than a preset similarity threshold; determining, based on the preset path of the drone, the potential edge line and the actual edge line corresponding to the target edge line at the corresponding acquisition time; wherein the potential edge line and the actual edge line are used to represent the morphological change information and relative position information of the corner points; and comparing the potential edge line and the actual edge line to determine the floating object area; The corner point corresponding to the floating object area is determined as the second target corner point.

2. The method for surveying and mapping water conservancy projects based on UAV remote sensing technology according to claim 1 is characterized in that: The step of determining the gradient concentration value of the pixel point at the position corresponding to the corner point on the edge line where the input corner point is located comprises: Determine the input gradient value of the pixel point at the position corresponding to the input corner point; Determine the gradient mean of the pixel gradients at positions corresponding to all corner points on the edge line where the input corner point is located; A gradient concentration value is calculated based on the input gradient value and the gradient mean.

3. The method for topographic mapping of water conservancy projects based on UAV remote sensing technology according to claim 1 is characterized in that: The step of determining target edge lines having similar shapes between a plurality of the initial corner point images comprises: Determine a first gradient of a pixel point at a position corresponding to a first corner point of a first edge line in one of the initial corner point images, and determine a second gradient of a pixel point at a position corresponding to a second corner point of a second edge line in the other initial corner point image; Based on the first gradient, the second gradient, the first number of corner points, and the second number of corner points, target edge lines having similar shapes between the plurality of initial corner point images are determined.

4. The method for surveying and mapping water conservancy projects based on UAV remote sensing technology according to claim 3 is characterized in that: The step of comparing the potential edge line with the actual edge line to determine the floating object area comprises: Determine the relative distance between the corresponding pixel points at the potential corner points on the potential edge line and the corresponding pixel points at the actual corner points on the actual edge line corresponding to the potential corner points; Determine the maximum value and the minimum value of the distance between all corner points on the potential edge line and the pixel points at corresponding positions on the actual edge line; Based on the relative distance value, the maximum distance value, the minimum distance value and the number of corner points of the potential edge line, an estimated value of the actual edge line belonging to the edge of the floating object is calculated; The estimated value is compared with a preset edge threshold to determine the floating object area.

5. The method for topographic surveying of water conservancy projects based on UAV remote sensing technology according to claim 1 is characterized in that: The step of removing the first target corner point and the second target corner point to obtain a processed target dam image comprises: By using a bilinear interpolation algorithm, the colors and textures of the bright spot area and the floating object area corresponding to the first target corner point and the second target corner point are smoothly transitioned with the surrounding water surface area, so as to remove the corner points of the bright spot area and the floating object area on the water surface in the initial dam image and obtain the processed target dam image.

6. The method for water conservancy project topography surveying and mapping based on UAV remote sensing technology according to claim 1 is characterized in that: The step of splicing a plurality of target dam images to obtain and output a water conservancy project topographic map comprises: The target dam image is taken as input, the SIFT algorithm is used to match the corner points in the target dam image, and the homography matrix is ​​used to align and stitch the images to output the water conservancy project topographic map.

Images