阅读说明：本技术 地形勘测中的地形勘测点布测优化方法 (Method for optimizing distribution of measuring points in topographic survey ) 是由 王英 彭令发 尹迎身 龚家国 冶运涛 于 2021-06-28 设计创作，主要内容包括：本发明公开了一种地形勘测中的地形勘测点布测优化方法,其包括镶嵌拼接得到研究区影像；对研究区影像进行边缘检测得到待研究区域内的地物轮廓线；获取由地物轮廓线围成的面矢量多边形和地物轮廓线构成的面矢量多边形,并结合分辨率为30m的DEM数据统计每个多边形的特征值；将特征值相似度大于预设阈值时对应的多边形合并为同一分区块；计算多边形之间特征值的相似度,并将相似度大于预设阈值时对应的多边形合并为同一分区块；采用碎部测量原则将每个分区块划分成若干图块,当图块中的特征值大于等于预设特征值时,根据数字地形测量规范进行地形勘测点布测；否则采用米字形进行地形勘测点布测,且米字形的边点位于图块的边界轮廓线上。(The invention discloses a topographic survey point distribution optimization method in topographic survey, which comprises the steps of inlaying and splicing to obtain an image of a research area; carrying out edge detection on the image of the research area to obtain a ground object contour line in the area to be researched; acquiring a surface vector polygon formed by the contour lines of the ground objects and a surface vector polygon formed by the contour lines of the ground objects, and counting the characteristic value of each polygon by combining DEM data with the resolution of 30 m; merging the corresponding polygons into the same partition block when the similarity of the characteristic values is greater than a preset threshold value; calculating the similarity of characteristic values among the polygons, and merging the corresponding polygons into the same partition block when the similarity is greater than a preset threshold value; dividing each block into a plurality of image blocks by adopting a fragmentary measurement principle, and performing topographic survey point distribution according to a digital topographic survey specification when the characteristic value in each image block is greater than or equal to a preset characteristic value; otherwise, the shape of the Chinese character mi is adopted to carry out the distribution of the topographic survey points, and the edge points of the shape of the Chinese character mi are positioned on the boundary contour line of the image block.)

1. A method for optimizing the distribution of survey points in terrain survey is characterized by comprising the following steps:

s1, acquiring a multispectral image and a panchromatic image of the area to be researched, preprocessing the multispectral image and the panchromatic image, and then mosaic-splicing the preprocessed images to obtain a research area image;

s2, performing edge detection on the image of the research area by using a Sobel operator to obtain a ground object contour line in the area to be researched;

s3, acquiring a surface vector polygon formed by the contour lines of the ground objects and a surface vector polygon formed by the contour lines of the ground objects, and counting the characteristic value of each polygon by combining DEM data with the resolution of 30 m;

s4, calculating the similarity of the characteristic values among the polygons, and merging the corresponding polygons into the same partition block when the similarity is greater than a preset threshold value;

s5, dividing each block into a plurality of blocks by adopting a fragmentary measurement principle, and performing topographic survey point distribution according to a digital topographic survey specification when the characteristic value in each block is greater than or equal to a preset characteristic value;

and when the characteristic value in the image block is smaller than the preset characteristic value, performing topographic survey point distribution by adopting the shape of a Chinese character 'mi', wherein the edge points of the shape of the Chinese character 'mi' are positioned on the boundary contour line of the image block.

2. The method for optimizing the layout of survey points in a terrain survey of claim 1, wherein said step S1 further comprises:

s11, acquiring a multispectral image and a panchromatic image of the region to be researched, and sequentially carrying out radiometric calibration, atmospheric correction and orthorectification on the multispectral image and carrying out radiometric calibration and orthorectification on the panchromatic image;

s12, carrying out image fusion on the multispectral image and the panchromatic image after the incidence correction, and carrying out geometric correction on the fused image by adopting a remote sensing image with the resolution higher than the resolution of the multispectral image and the panchromatic image;

and S13, carrying out mosaic splicing on the images after geometric correction to obtain a final image of the research area.

3. The method for optimizing the layout of survey points in a terrain survey of claim 1, wherein said step S2 further comprises:

s21, converting the image of the research area into a jp2 format picture consisting of R, G, B three bands;

s22, converting the picture into a gray-scale image, and then carrying out normalization processing;

s23, calling a sobel operator to perform convolution on the normalized image in four directions of 0 degree, 45 degrees, 90 degrees and 135 degrees respectively to obtain gradient values of each pixel point in the image in the four directions:

wherein A is the pixel value of the pixel point after normalization; g_0°、G_45°、G_90°、G_135°The gradient values of the pixel points in the directions of 0 degree, 45 degrees, 90 degrees and 135 degrees are respectively;

s23, calculating the gradient value of each pixel after convolution according to the gradient values of the pixels in four directions:

wherein G is_xAnd G_yGradient values of the pixel points in the x direction and the y direction are respectively; g is the gradient value after the convolution of the pixel points;

s24, judging whether the gradient value of the pixel point is larger than a preset gradient threshold value or not; if yes, marking the pixel points as edge points, otherwise marking the pixel points as non-edge points;

and S25, adopting all edge points of the marks as contour lines of the ground objects in the area to be researched.

4. The method of optimizing the layout of survey points in a terrain survey of claim 3 wherein the predetermined gradient threshold is the average of the convolved gradient values of all pixel points of the area under study.

5. The method for optimizing the layout of survey points in a terrain survey of claim 1, wherein said step S3 further comprises:

s31, geographic registration is carried out by adopting the research area image obtained in S13, a geographic coordinate system is defined as GCS _ WGS _1984, and an edge image with spatial reference is obtained;

s32, assigning the pixel value of the ground object contour line in the edge image to be 1, assigning the pixel of the non-edge point to be 0, and converting the assigned edge image into a surface vector by a grid to obtain a surface vector polygon formed by a surface vector polygon surrounded by the ground object contour line and the ground object contour line;

s33, according to DEM data with the resolution of 30m, counting the mean value, variance, standard deviation, median, mode and the most value of the characteristic value of the internal elevation of each polygon:

X_min＝min(X₁，X₂...X_i)，X_max＝max(X₁，X₂...X_i)

wherein, X_minThe minimum value of the elevation values in the polygon is obtained; x_maxThe maximum value of the elevation value in the polygon; m_OIs the mode of the elevation values in the polygon;the average value of the elevation values in the polygon is obtained; x_iIs the ith minimum elevation value, M, within the polygon_dThe median of the elevation values in the polygon is shown, sigma is the standard deviation of the elevations in the polygon, and n is the number of the elevations in the polygon.

6. The method of optimizing the layout of survey points in a terrain survey of claim 5 wherein, when n is an odd number,when N is an even number, the number of bits in the bit line is,

7. the method for optimizing the layout of survey points in a terrain survey of claim 5, wherein said step S4 further comprises:

s41, forming a set by all polygons, and randomly selecting a polygon in the set;

s42, calculating the ratio of the selected polygon to the corresponding characteristic value of the rest polygons in the set, and taking 6 ratios of the two polygons as a group of data;

s43, judging whether all the 6 specific values in each group of data are larger than a preset threshold value, if so, marking the current polygon in the remaining polygons, otherwise, not marking;

s44, after each group of data is compared with a preset threshold value, judging whether a marked polygon exists, if so, entering a step S45, otherwise, entering a step S46;

s45, merging the marked polygon and the selected polygon into a polygon, adopting the non-merged polygon to update a set, and randomly selecting a polygon to enter the step S42;

s46, the set is updated by using the rest polygons in the set, and a polygon is randomly selected to enter the step S42.

Technical Field

The invention relates to a data acquisition path planning method, in particular to a method for optimizing the distribution and measurement of a topographic survey point in topographic survey.

Background

The DEM is a solid earth surface model which expresses the ground elevation in a group of ordered numerical value array forms and is widely applied to the aspects of surveying and mapping, hydrology, meteorology, engineering construction, communication, military and the like at the present stage. For hydraulic engineering, the existing simulations of some hydrological runoff, rainfall, flood control, disaster reduction and the like all need support of the DEM, and topographic data is the most basic and most core part in hydrological model simulation.

At present, when DEM data is manufactured, a common method is to determine some topographic survey points, acquire elevation data of the topographic survey points and generate final DEM data in an interpolation mode. At present, when a topographic survey point is selected, the topographic survey point is generally selected in an equidistant mode in a region to be researched, then, elevation data are acquired manually or by an unmanned aerial vehicle, and due to the fact that the topography of partial regions of the research region is similar, the acquired elevation data are basically similar, and when the partial similar data generate DEM data through interpolation, the accuracy of the DEM data is not greatly contributed.

And these data all adopt artifical or unmanned aerial vehicle's mode to gather back, this has increased the work load and the artifical cost of artifical data collection undoubtedly, or the lease cost when unmanned aerial vehicle gathered data.

Disclosure of Invention

Aiming at the defects in the prior art, the method for optimizing the distribution of the survey points in the terrain survey solves the problem of cost waste caused by unreasonable layout of the existing survey points in the terrain survey.

In order to achieve the purpose of the invention, the invention adopts the technical scheme that:

a method for optimizing the layout of survey points in a terrain survey is provided, which comprises the following steps:

s1, acquiring a multispectral image and a panchromatic image of the area to be researched, preprocessing the multispectral image and the panchromatic image, and then mosaic-splicing the preprocessed images to obtain a research area image;

s2, performing edge detection on the image of the research area by using a Sobel operator to obtain a ground object contour line in the area to be researched;

s3, acquiring a surface vector polygon formed by the contour lines of the ground objects and a surface vector polygon formed by the contour lines of the ground objects, and counting the characteristic value of each polygon by combining DEM data with the resolution of 30 m;

s4, calculating the similarity of the characteristic values among the polygons, and merging the corresponding polygons into the same partition block when the similarity is greater than a preset threshold value;

s5, dividing each block into a plurality of blocks by adopting a fragmentary measurement principle, and performing topographic survey point distribution according to a digital topographic survey specification when the characteristic value in each block is greater than or equal to a preset characteristic value;

and when the characteristic value in the image block is smaller than the preset characteristic value, performing topographic survey point distribution by adopting the shape of a Chinese character 'mi', wherein the edge points of the shape of the Chinese character 'mi' are positioned on the boundary contour line of the image block.

The invention has the beneficial effects that: according to the scheme, the remote sensing technology is used for carrying out ground object interpretation so as to realize reasonable partitioning, and then characteristic value comparison is carried out by combining with a public DEM (digital elevation model) with 30m spatial resolution, so that reasonable secondary verification is carried out on partitioned data, so that similar landforms are partitioned to the same area, and a large amount of invalid terrain survey is discharged;

and because similar landforms are divided into the same region, the region is mainly of similar landforms, a small number of landform survey points can represent the whole region, and although the number of the landform survey points is greatly reduced, the accuracy of the DEM generated in the subsequent interpolation process cannot be influenced.

After this scheme carries out the reasonable layout of topography reconnaissance point, if adopt the manual work to carry out elevation data acquisition, can reduce artifical data acquisition's work load and artifical this, if adopt unmanned aerial vehicle to carry out data acquisition, can optimize unmanned aerial vehicle's aircraft route, shorten the length of time that unmanned aerial vehicle gathered data, and then reduce unmanned aerial vehicle gathered data's expensive lease cost.

Drawings

FIG. 1 is a flow chart of a method of optimizing the layout of points in a terrain survey.

Detailed Description

The following description of the embodiments of the present invention is provided to facilitate the understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and it will be apparent to those skilled in the art that various changes may be made without departing from the spirit and scope of the invention as defined and defined in the appended claims, and all matters produced by the invention using the inventive concept are protected.

Referring to FIG. 1, FIG. 1 illustrates a method of point placement optimization in a terrain survey; as shown in fig. 1, the method includes steps S1 through S5.

In step S1, a multispectral image and a panchromatic image of the region to be studied are obtained, the multispectral image and the panchromatic image are preprocessed, and then the preprocessed images are mosaiced to obtain a study region image.

In this embodiment, step S1 preferably further includes:

s11, acquiring a multispectral image and a panchromatic image of the region to be researched, and sequentially carrying out radiometric calibration, atmospheric correction and orthorectification on the multispectral image and carrying out radiometric calibration and orthorectification on the panchromatic image;

s12, carrying out image fusion on the multispectral image and the panchromatic image after the incidence correction to obtain a region image to be researched with higher precision, and then carrying out geometric correction on the fused image by adopting a remote sensing image with the resolution higher than the resolution of the multispectral image and the panchromatic image; and performing band fusion by adopting the multispectral image and the panchromatic band image, so as to improve the resolution of the remote sensing image.

And S13, carrying out mosaic splicing on the images after geometric correction to obtain a final image of the research area.

In step S2, a Sobel operator is used to perform edge detection on the image of the research area, so as to obtain a contour line of the ground feature in the area to be researched.

In an embodiment of the present invention, the step S2 further includes:

and S21, converting the image of the research area into a picture in a jp2 format consisting of R, G, B three bands, wherein the resolution of the converted picture can be kept unchanged.

S22, converting the picture into a gray-scale image, and then carrying out normalization processing;

s23, calling a sobel operator to perform convolution on the normalized image in four directions of 0 degree, 45 degrees, 90 degrees and 135 degrees respectively to obtain gradient values of each pixel point in the image in the four directions:

wherein A is the pixel value of the pixel point after normalization;

s23, calculating the gradient value of each pixel after convolution according to the gradient values of the pixels in four directions:

wherein G is_0°、G_45°、G_90°、G_135°The gradient values of the pixel points in the directions of 0 degree, 45 degrees, 90 degrees and 135 degrees are respectively; g_xAnd G_yGradient values of the pixel points in the x direction and the y direction are respectively; g is the gradient value after the convolution of the pixel points;

s24, judging whether the gradient value of the pixel point is larger than a preset gradient threshold value or not; if yes, marking the pixel points as edge points, otherwise marking the pixel points as non-edge points;

the preset gradient threshold value of the scheme is preferably the average value of the gradient values of all pixel points in the region to be researched after convolution.

In order to further improve the marking accuracy, the preset gradient threshold of the scheme can be optimized as follows:

s241, selecting areas with the ground object types larger than the preset number in the remote sensing image of the research area as test areas, and drawing ground object contour lines in an artificial marking mode to serve as verification samples;

s242, acquiring a ground object contour line extracted from a block corresponding to the test area in the area to be researched, and adopting a test sample formed by the extracted ground object contour line;

s243, respectively calculating coverage areas F1 and F2 of the area formed by the verification sample and the test sample, and judging whether the ratio of the coverage area F2 to the coverage area F1 is larger than or equal to an area threshold value;

s244, outputting a preset gradient threshold when the ratio is larger than or equal to the area threshold; when the ratio is smaller than the area threshold, the initial preset gradient threshold is set to 2 × initial preset gradient threshold, and then the process returns to step S243.

And S25, adopting all the marked edge points as the contour lines of the ground objects in the area to be researched, and setting the edge points as white points and the non-edge points as black points for facilitating subsequent checking.

In step S3, a surface vector polygon formed by the surface vector polygons surrounded by the feature lines and the feature lines is obtained, and feature values of each polygon are counted in combination with DEM data with a resolution of 30m (the data source of the DEM data with a resolution of 30m may refer to the website http:// www.gscloud.cn /).

In an embodiment of the present invention, the step S3 further includes:

s31, carrying out geographic registration by adopting the remote sensing image obtained by S13, and defining a geographic coordinate system GCS _ WGS _1984 to obtain an edge image with spatial reference;

s32, assigning the pixel value of the ground object contour line in the edge image to be 1, assigning the pixel of the non-edge point to be 0, and converting the assigned edge image into a surface vector by a grid to obtain a surface vector polygon formed by a surface vector polygon surrounded by the ground object contour line and the ground object contour line;

s33, according to DEM data with the resolution of 30m, counting the mean value, variance, standard deviation, median, mode and the most value of the characteristic value of the internal elevation of each polygon:

X_min＝min(X₁，X₂...X_i)，X_max＝max(X₁，X₂...X_i)

wherein, X_minThe minimum value of the elevation values in the polygon is obtained; x_maxThe maximum value of the elevation value in the polygon; m_OIs the mode of the elevation values in the polygon;the average value of the elevation values in the polygon is obtained; x_iIs the ith minimum elevation value, M, within the polygon_dThe median of the elevation values in the polygon is shown, sigma is the standard deviation of the elevation in the polygon, and n is the number of the elevation in the polygon;

when n is an odd number, the number of the transition metal atoms,when N is an even number, the number of bits in the bit line is,

in step S4, calculating the similarity of feature values between polygons, and merging corresponding polygons into the same partition when the similarity is greater than a preset threshold;

in an embodiment of the present invention, the step S4 further includes:

s41, forming a set by all polygons, and randomly selecting a polygon in the set;

s42, calculating the ratio of the selected polygon to the corresponding characteristic value of the rest polygons in the set:

Q_Ak2＝σ_A/σ_k，Q_Ak3＝M_{O_A}/M_{O_k}

Q_Ak4＝X_{min_A}/X_{min_k}、Q_Ak5＝X_{max_A}/X_{max_k}、Q_Ak6＝M_{d_A}/M_{d_k}

wherein the content of the first and second substances,σ_A、M_{O_A}、X_{min_A}、X_{max_A}and M_{d_A}Respectively the average value, the standard deviation, the mode, the maximum value, the minimum value and the median of the height value of the selected polygon A;σ_k、M_{O_k}、X_{min_k}、X_{max_k}and M_{d_k}Respectively the average value, standard deviation, mode, maximum value, minimum value and median of the kth polygon elevation value in the rest polygons; q_Ak1、Q_Ak2、Q_Ak3、Q_Ak4、Q_Ak5And Q_Ak6Respectively is the average value ratio, standard deviation ratio, mode ratio, maximum value ratio and minimum value ratio of the polygon A and the k-th polygon in the rest polygons;

then the 6 ratios (Q) of the two polygons_Ak1、Q_Ak2、Q_Ak3、Q_Ak4、Q_Ak5And Q_Ak6) As a set of data.

S43, judging whether all the 6 specific values in each group of data are larger than a preset threshold value, if so, marking the current polygon in the remaining polygons, otherwise, not marking;

s44, after each group of data is compared with a preset threshold value, judging whether a marked polygon exists, if so, entering a step S45, otherwise, entering a step S46;

s45, merging the marked polygon and the selected polygon into a polygon, adopting the non-merged polygon to update a set, and randomly selecting a polygon to enter the step S42;

s46, the set is updated by using the rest polygons in the set, and a polygon is randomly selected to enter the step S42.

According to the scheme, secondary verification is carried out on the subareas of the area to be researched through statistical characteristic value comparison; the landform in the same subarea range is relatively gentle, and the difference of the landform in elevation is relatively small. If the fluctuation degrees of the reeds are similar, after the reeds are divided into an area, when the reed terrain survey is carried out, a small amount of terrain survey can be selected due to the same landform, so that the economic cost required by the overall interval data acquisition is reduced.

In step S5, each block is divided into a plurality of blocks by using a fractional part measurement principle, and when the feature value in the block is greater than or equal to a preset feature value, it indicates that the terrain is relatively complicated, such as a depression, a mountain land, etc., at this time, the distribution of the terrain survey points is performed according to the digital terrain survey specification, wherein the terrain survey points are distributed on the boundary contour lines of the block as much as possible.

When the characteristic value in the image block is smaller than the preset characteristic value, the terrain of the image block is relatively flat, the terrain survey point distribution can be carried out by adopting the shape of a Chinese character mi, and the edge points of the shape of the Chinese character mi are positioned on the boundary contour line of the image block.

In conclusion, according to the method for optimizing the distribution of the topographic survey points, areas with similar terrains are divided into the same block, the quantity of the distribution of the topographic survey points in the block is reduced through reasonable distribution of the topographic survey points, the method for optimizing the distribution of the topographic survey points can ensure the accuracy of a DEM (digital elevation model) obtained through interpolation of elevation data acquired from the points subsequently while reducing the quantity of the topographic survey points, and meanwhile, the cost for acquiring the elevation data can be greatly reduced.