Surveying and mapping method based on remote sensing big data analysis

Abstract

The application discloses a surveying and mapping method based on remote sensing big data analysis, and relates to the technical field of remote sensing surveying and mapping, and comprises the following steps: acquiring a judgment mean difference based on a real-time main body contour; acquiring a fuzzy threshold based on a second quantity of database surveying and mapping image maps; judging whether a real-time surveying and mapping grayscale map is qualified based on the judgment mean difference and the fuzzy threshold, if not, repeatedly acquiring a new real-time surveying and mapping image map until it is qualified, and marking the qualified real-time surveying and mapping image map as a qualified surveying and mapping image map; acquiring the qualified surveying and mapping image map of each detection point, and splicing the qualified surveying and mapping image map of each detection point to obtain a surveying and mapping overall report map; the application is used for solving the problems that the existing remote sensing surveying and mapping technology does not perform fuzzy judgment when acquiring images by unmanned aerial vehicle remote sensing surveying and mapping, resulting in that the finally acquired remote sensing images are not clear and the workload is increased.

Description

Technical Field

[0001] This invention relates to the field of remote sensing mapping technology, specifically a mapping method based on remote sensing big data analysis. Background Technology

[0002] With the rapid development of space information technology and sensor technology, remote sensing technology has become one of the core means in the modern surveying and mapping field. There are various existing remote sensing-based surveying and mapping methods, such as UAV remote sensing surveying and satellite remote sensing surveying.

[0003] Existing UAV remote sensing mapping relies on unmanned aerial vehicles (UAVs) carrying sensors to acquire survey maps. However, UAVs are affected by environmental factors, such as wind, which can affect their stability and consequently the acquisition of images, resulting in blurry images. Typically, in such cases, the UAV needs to fly back to the acquisition point to retrieve the images, leading to unclear final remote sensing images and increased workload. For example, patent application CN118274805A discloses a mapping method based on remote sensing big data analysis. This method does not consider the blurring that can occur in UAV remote sensing mapping, resulting in unclear final remote sensing images and increased workload. In other words, existing remote sensing mapping technologies do not perform blur detection when UAVs acquire images, leading to unclear final remote sensing images and increased workload. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. It obtains the evaluation mean difference based on the real-time subject contour, obtains the fuzziness threshold based on a second number of database survey images, and determines whether the real-time survey grayscale image is qualified based on the evaluation mean difference and the fuzziness threshold. If it is not qualified, new real-time survey images are repeatedly obtained until it is qualified. The qualified real-time survey images are marked as qualified survey images. This solves the problem that the existing remote sensing survey technology does not perform fuzziness judgment when the image is acquired by UAV remote sensing survey, resulting in unclear remote sensing images and increased workload.

[0005] To achieve the above objectives, this application provides a mapping method based on remote sensing big data analysis, comprising the following steps:

[0006] Obtain the mapping points and mapping path of the UAV generated based on the mapping area;

[0007] As the drone travels along the mapping path, it acquires a real-time mapping image at each mapping point it reaches.

[0008] The real-time mapping image is converted to grayscale to obtain a real-time mapping grayscale image. A first subject threshold and a second subject threshold are obtained based on a first number of mapping subject images. The real-time mapping grayscale image is binarized based on the first subject threshold and the second subject threshold to obtain a real-time mapping binarized image. The real-time subject outline is obtained based on the real-time mapping binarized image.

[0009] The average error is judged based on the real-time subject outline;

[0010] The blur threshold is obtained based on the second number of database image maps;

[0011] The evaluation mean difference and fuzzy threshold are used to determine whether the real-time mapping grayscale image is qualified. If it is not qualified, new real-time mapping images are repeatedly acquired until it is qualified. The qualified real-time mapping images are marked as qualified mapping images.

[0012] Obtain qualified surveying images of each detection point, and stitch together the qualified surveying images of each detection point to obtain the overall surveying report.

[0013] Furthermore, the process of converting the real-time mapping image to grayscale to obtain a real-time mapping grayscale image includes the following sub-steps:

[0014] Obtain the RGB value of each pixel in the real-time mapping image and mark it as the real-time mapping RGB value;

[0015] The weighted average grayscale conversion formula is used to convert all real-time RGB values ​​into grayscale values ​​to obtain a real-time grayscale image.

[0016] Furthermore, obtaining the first subject threshold and the second subject threshold based on the first number of surveyed subject images includes the following sub-steps:

[0017] Obtain the first number of images containing only the main body of the survey, and mark them as the main body of the survey;

[0018] The main survey map is converted to grayscale to obtain a grayscale image of the main survey map.

[0019] Obtain the grayscale value of each pixel in the grayscale image of the main subject and mark it as the grayscale value of the main subject.

[0020] The grayscale values ​​of the surveyed subjects are sorted from smallest to largest and labeled as Lym1 to Lym. i ;

[0021] The number of grayscale values ​​of the main subject obtained is marked as the grayscale number of the main subject.

[0022] The first position ratio is calculated as: Wzb1 = m1 * Hzt, where Wzb1 is the first position ratio, m1 is the first position coefficient, the range of m1 is (0, 0.5), and Hzt is the gray level of the subject being surveyed.

[0023] If the ratio of the first position is an integer, then Lym （Wzb1） Mark it as the first position value; if the first position ratio is not an integer, obtain the integers on the left and right sides of Wzb1, and mark them as Wzb1n and Wzb1x respectively, and calculate Lym. （Wzb1n） With Lym （Wzb1x） The mean value is marked as the first position value;

[0024] The second position ratio is obtained as: Wzb2 = m2 * Hzt, where Wzb2 is the second position ratio, m2 is the second position coefficient, and the range of m2 is [0.5, 1).

[0025] If the ratio of the second position is an integer, then Lym （Wzb2） Mark the value as the second position value; if the second position ratio is not an integer, obtain the integers on the left and right sides of Wzb2, and mark them as Wzb2n and Wzb2x respectively, and calculate Lym. （Wzb2n） With Lym （Wzb2x） The mean value is marked as the second position value.

[0026] Furthermore, obtaining the first subject threshold and the second subject threshold based on the first number of surveyed subject images also includes the following sub-steps:

[0027] grayscale of the main body of the survey Figure 1 Establish u*u pixel squares at each corner, marking them as average pixel squares; move the average pixel squares with a step size of N, starting the movement along the grayscale of the main subject being measured. Figure 1 When the object moves to the edge of the grayscale image of the main body, it moves one step along this edge, and then moves in the opposite direction of the initial movement. When the object moves to the edge of the grayscale image of the main body again, it moves one step along this edge. The above operation is repeated until the object can no longer move in the grayscale image of the main body.

[0028] The average grayscale value of the main subject within the average pixel square is obtained for each movement and marked as the average grayscale value of the square.

[0029] Obtain the number of average gray values ​​of the squares between the first position value and the second position value, and mark them as the number of intermediate distributions;

[0030] Get the number of squares with average gray values ​​less than the first position value, and mark them as the smaller distribution number;

[0031] Get the number of squares with an average gray value greater than the first position value, and mark them as the number of larger distributions;

[0032] The proportion of the first distribution is calculated as: B1 = Fn / Fz; where B1 is the proportion of the first distribution, Fn is the number of the smaller distribution, and Fz is the number of the middle distribution.

[0033] The proportion of the second distribution is calculated as: B2 = Fx / Fz; where B2 is the proportion of the second distribution and Fx is the number of the larger distribution.

[0034] Furthermore, obtaining the first subject threshold and the second subject threshold based on the first number of surveyed subject images also includes the following sub-steps:

[0035] The threshold for the first subject is calculated as: Lz1 = Wzz1 - B1 * (Wzz2 - Wzz1).

[0036] The second subject threshold is calculated as: Lz2 = Wzz2 + B2 * (Wzz2 - Wzz1); where Lz1 is the first subject threshold, Lz2 is the second subject threshold, Wzz1 is the first position value, and Wzz2 is the second position value.

[0037] Furthermore, the process of binarizing the real-time mapping grayscale image based on the first subject threshold and the second subject threshold to obtain a real-time mapping binarized image includes the following sub-steps:

[0038] Set the grayscale values ​​of the surveying subjects in the real-time surveying grayscale image that are greater than or equal to the first subject threshold and less than or equal to the second subject threshold to 0, and set the grayscale values ​​of the surveying subjects in the real-time surveying grayscale image that are less than the first subject threshold or greater than the second subject threshold to 255 to obtain a real-time surveying binarized image.

[0039] Furthermore, obtaining the real-time subject outline based on the real-time mapping binarized image includes the following sub-steps:

[0040] In the real-time mapping binarized image, pixels with a grayscale value of 0 are marked as main pixels. The largest continuous region composed of main pixels is marked as the main region. The part adjacent to the region with a grayscale value of 255 pixels in the real-time mapping binarized image is marked as the real-time main outline.

[0041] Furthermore, obtaining the evaluation mean difference based on the real-time subject contour includes the following sub-steps:

[0042] Establish a Cartesian coordinate system, labeled as the first analysis coordinate system. Place the real-time mapping binarized image in the first quadrant of the first analysis coordinate system, while ensuring that the two sides of the real-time mapping binarized image coincide with the horizontal and vertical axes of the first analysis coordinate system, respectively, to obtain the coordinates of the real-time main outline.

[0043] Establish a Cartesian coordinate system, labeled as the second analysis coordinate system, and place the real-time grayscale image in the first quadrant, while ensuring that the two sides of the real-time grayscale image coincide with the horizontal and vertical axes of the second analysis coordinate system, respectively; draw the real-time subject outline on the real-time grayscale image based on the coordinates of the real-time subject outline.

[0044] Obtain the pixels on the real-time subject outline in the real-time grayscale image and mark them as subject outline pixels;

[0045] In the real-time grayscale image, the main outline pixels and the pixels contained in the main outline pixels are marked as real-time main pixels.

[0046] A 3x3 pixel matrix is ​​established centered on each main outline pixel, labeled as the dilation matrix. Pixels in the dilation matrix that are not real-time main pixels are identified and labeled as first-dilated pixels. The average grayscale value of the first-dilated pixels is obtained and labeled as Pz1. A dilation matrix is ​​then established centered on each first-dilated pixel. Pixels in the dilation matrix that are not real-time main pixels and the first-dilated pixels are identified and labeled as second-dilated pixels. The average grayscale value of the second-dilated pixels is obtained and labeled as Pz2, and so on, until Pz1 is obtained. t ;

[0047] Transfer Pz1 to Pz t Marked as Pz j , where j is an integer from 1 to t;

[0048] The mean difference of the evaluation is calculated as follows: C1 represents the average difference in evaluation.

[0049] Furthermore, obtaining the blur threshold based on the second number of database mapping images includes the following sub-steps:

[0050] Acquire a second number of images of the same type as the main body of the survey and clearly marked survey points, and mark them as database survey image maps;

[0051] The step of obtaining the evaluation mean difference based on real-time surveying and mapping images is to obtain the evaluation mean difference of the database surveying and mapping images and mark it as the database evaluation mean difference;

[0052] Find the minimum average difference among all database evaluations and mark it as the fuzzy threshold.

[0053] Furthermore, based on the evaluation mean difference and fuzzy threshold, the real-time mapping grayscale image is judged to be qualified. If it is not qualified, a new real-time mapping image is repeatedly acquired until it is qualified. Marking the qualified real-time mapping image as a qualified mapping image includes the following sub-steps:

[0054] Determine if the average difference of the evaluation is less than the fuzzy threshold. If it is less, mark the real-time grayscale image as unqualified. Repeat the acquisition of new real-time image images until the average difference of the evaluation of the new real-time image images is greater than or equal to the fuzzy threshold. Mark the real-time image images with an average difference of evaluation greater than or equal to the fuzzy threshold as qualified image images.

[0055] The beneficial effects of this invention are as follows: This invention obtains the evaluation mean difference based on the real-time subject contour, obtains the fuzziness threshold based on a second number of database mapping images, and judges whether the real-time mapping grayscale image is qualified based on the evaluation mean difference and the fuzziness threshold. If it is not qualified, new real-time mapping images are repeatedly obtained until it is qualified. The qualified real-time mapping images are marked as qualified mapping images. The advantage is that fuzziness judgment is performed when the UAV remote sensing mapping acquires images, making the final acquired remote sensing image clearer and reducing the workload of repeated UAV flights.

[0056] This invention obtains a real-time mapping binarized image by binarizing a real-time mapping grayscale image based on a first subject threshold and a second subject threshold. The advantage is that the binarization based on the first subject threshold and the second subject threshold can more accurately separate the subject part and reduce interference.

[0057] This invention determines whether a real-time grayscale image is qualified based on the evaluation mean difference and a fuzzy threshold. Its advantage lies in obtaining the evaluation mean difference based on the dilation matrix. The evaluation mean difference is used as the change in the grayscale value of the pixel points on the edge of the object in the real-time grayscale image to determine whether the real-time grayscale image is fuzzy. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the system of the present invention;

[0059] Figure 2 This is a schematic diagram of the expansion matrix of the present invention;

[0060] Figure 3 This is a schematic diagram of the second expanded pixel of the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Example 1, please refer to Figure 1 As shown, this application provides a mapping method based on remote sensing big data analysis, including the following steps:

[0063] Step S1: Obtain the survey points and survey path of the UAV based on the survey area; the survey points and survey path can be manually set and input to ensure that the entire survey area is covered.

[0064] Step S2: As the UAV travels along the mapping path, it acquires a real-time mapping image every time it reaches a mapping point.

[0065] Step S3 involves converting the real-time mapping image to grayscale to obtain a real-time mapping grayscale image, obtaining a first subject threshold and a second subject threshold based on a first number of mapping subject images, performing binarization processing on the real-time mapping grayscale image based on the first subject threshold and the second subject threshold to obtain a real-time mapping binarized image, and obtaining the real-time subject outline based on the real-time mapping binarized image; Step S3 also includes the following sub-steps:

[0066] Step S301: Obtain the RGB value of each pixel in the real-time mapping image and mark it as the real-time mapping RGB value;

[0067] Step S302: Use the weighted average grayscale conversion formula to convert all real-time measured RGB values ​​into grayscale values ​​to obtain a real-time measured grayscale image; where the weighted average grayscale conversion formula is Hdz=0.299*R+0.587*G+0.114*B, Hdz is the grayscale value, and R, G and B are the three channel values ​​of the real-time measured RGB values ​​respectively.

[0068] Step S303: Obtain a first number of images containing only the main part of the survey, and mark them as the main part of the survey; the main part of the survey, for example, when surveying a building, the building is the main part of the survey, and the area around the building is not the main part of the survey; or, for example, when surveying a road, the road is the main part of the survey, and the area around the road is not the main part of the survey, and select one survey area;

[0069] Step S304: Perform grayscale processing on the main surveying image to obtain a grayscale image of the main surveying image;

[0070] Step S305: Obtain the grayscale value of each pixel in the grayscale image of the surveying subject and mark it as the grayscale value of the surveying subject;

[0071] Step S306: Sort the grayscale values ​​of the surveyed subject in ascending order and label them as Lym1 to Lym. i ;

[0072] Step S307: The number of grayscale values ​​of the surveying subject is marked as the grayscale number of the surveying subject;

[0073] Step S308, calculate the first position ratio as: Wzb1 = m1 * Hzt, where Wzb1 is the first position ratio, m1 is the first position coefficient, the range of m1 is (0, 0.5), and Hzt is the grayscale value of the subject being surveyed; if the first position ratio is an integer, then Lym... （Wzb1） Mark it as the first position value; if the first position ratio is not an integer, obtain the integers on the left and right sides of Wzb1, and mark them as Wzb1n and Wzb1x respectively, and calculate Lym. （Wzb1n） With Lym （Wzb1x）The mean value is marked as the first position value; m1 is usually set to 0.25. The first position value is selected from the values ​​distributed in the smaller half of the range. Therefore, m1 is set to less than 0.5. Within the range of 0 to 0.5, an intermediate value can be selected to ensure that the selected first judgment value is representative of the smaller half of the range. 0.25 is preferred.

[0074] Step S309, calculate the second position ratio as: Wzb2 = m2 * Hzt, where Wzb2 is the second position ratio, m2 is the second position coefficient, and the range of m2 is [0.5, 1), Hzt is the grayscale value of the surveyed subject; if the second position ratio is an integer, then Lym （Wzb2） Mark the value as the second position value; if the second position ratio is not an integer, obtain the integers on the left and right sides of Wzb2, and mark them as Wzb2n and Wzb2x respectively, and calculate Lym. （Wzb2n） With Lym （Wzb2x） The mean value is marked as the second position value; when m1 is 0.25, m2 is 0.75.

[0075] In practical applications, for example, if we acquire 10 4000*3000 pixel survey images, with roads as the main subject, and the grayscale value of the main subject is 1200000000, then the first position ratio is calculated as: Wzb1 = 0.25 * 120000000 = 30000000, where 30000000 is an integer. Then we obtain Lym... （30000000） =113, mark 113 as the first position value, and calculate the second position ratio: Wzb2=0.75*120000000=90000000, then 90000000 is an integer, obtain Lym. （90000000） =138;

[0076] Mark 138 as the second position value;

[0077] Step S310, in the grayscale of the surveying subject Figure 1 Establish u*u pixel squares at each corner, marking them as average pixel squares; move the average pixel squares with a step size of N, starting the movement along the grayscale of the main subject being measured. Figure 1 When the image moves to the edge of the grayscale image of the main subject, it moves one step along this edge, and then moves in the opposite direction of the initial movement. When it moves to the edge of the grayscale image of the main subject again, it moves one step along this edge. This operation is repeated until it can no longer move in the grayscale image of the main subject. The value of u is set to 3 or 5. When u is 3, the movement step is 3, so that the average pixel square moves through the maximum coverage of the image.

[0078] Step S311: The average gray value of all pixels in the average pixel square of the surveying subject is obtained at each movement and marked as the average gray value of the square; this method is used to smooth individual noise pixels with large or small gray values ​​of the surveying subject.

[0079] Step S312: Obtain the number of average gray values ​​of the squares between the first position value and the second position value, and mark them as the number of intermediate distributions;

[0080] Step S313: Obtain the number of squares with average gray values ​​less than the first position value and mark them as the number of smaller distributions;

[0081] Step S314: Obtain the number of squares with average gray values ​​greater than the first position value and mark them as the number of larger distributions;

[0082] Step S315, calculate the proportion of the first distribution as: B1 = Fn / Fz; where B1 is the proportion of the first distribution, Fn is the number of smaller distributions, and Fz is the number of intermediate distributions;

[0083] Step S316, calculate the proportion of the second distribution as: B2=Fx / Fz; where B2 is the proportion of the second distribution and Fx is the number of the larger distribution; the proportion of the first distribution is the proportion after smoothing the gray value of the main subject and dividing it into the first position value and the second position value as the real-time proportion.

[0084] Step S317, calculate the first subject threshold as: Lz1=Wzz1-B1*(Wzz2-Wzz1).

[0085] Step S318, calculate the second subject threshold as: Lz2=Wzz2+B2*(Wzz2-Wzz1); where Lz1 is the first subject threshold, Lz2 is the second subject threshold, Wzz1 is the first position value, and Wzz2 is the second position value; so that the gray values ​​of the subject part are between the first subject threshold and the second subject threshold.

[0086] Step S319: Set the gray values ​​of the surveying subjects in the real-time surveying grayscale image that are greater than or equal to the first subject threshold and less than or equal to the second subject threshold to 0, and set the gray values ​​of the surveying subjects in the real-time surveying grayscale image that are less than the first subject threshold or greater than the second subject threshold to 255, thereby obtaining a real-time surveying binarized image.

[0087] In practical applications, the number of intermediate distributions is 7,131,202, the number of smaller distributions is 2,343,128, and the number of larger distributions is 2,323,126. The proportion of the first distribution is calculated as: 2,343,128 / 7,131,202 = 0.33, and the proportion of the second distribution is calculated as: 2,323,126 / 7,131,202 = 0.33. The calculation result is rounded to two decimal places. Therefore, the threshold for the first subject is: Lz1 = 113 - 0.33 * (138) -113) = 105, the second subject threshold is calculated as: 138 + 0.33 * (138 - 113) = 144. The calculation results of the first subject threshold and the second subject threshold are rounded to integers. The gray values ​​of the surveying subjects in the real-time surveying grayscale image that are greater than or equal to 105 and less than or equal to 144 are set to 0. The gray values ​​of the surveying subjects in the real-time surveying grayscale image that are less than or equal to 105 or greater than or equal to 144 are set to 255. The real-time surveying binarized image is obtained.

[0088] Step S320: Mark the pixels with a gray value of 0 in the real-time mapping binarized image as main pixels, obtain the largest continuous area formed by the main pixels and mark it as the main area, and obtain the part adjacent to the area with a gray value of 255 pixels in the real-time mapping binarized image and mark it as the real-time main outline.

[0089] Step S4: Obtain the evaluation mean difference based on the real-time subject contour; Step S4 also includes the following sub-steps:

[0090] Step S401: Establish a Cartesian coordinate system, labeled as the first analysis coordinate system, and place the real-time mapping binarized image in the first quadrant of the first analysis coordinate system. At the same time, ensure that the two sides of the real-time mapping binarized image coincide with the horizontal and vertical axes of the first analysis coordinate system, and obtain the coordinates of the real-time main outline.

[0091] Step S402: Establish a Cartesian coordinate system, labeled as the second analysis coordinate system, and place the real-time grayscale image in the first quadrant, while ensuring that the two sides of the real-time grayscale image coincide with the horizontal and vertical axes of the second analysis coordinate system, respectively; draw the real-time subject outline on the real-time grayscale image based on the coordinates of the real-time subject outline; because although the image has been processed, the size of the real-time grayscale image and the real-time binary image remains unchanged, so after the same image placement, the relative positions of the coordinate points obtained on the real-time binary image and the data edges drawn on the real-time grayscale image remain unchanged.

[0092] Step S403: Obtain the pixels on the real-time main body outline in the real-time grayscale image and mark them as the main body outline pixels;

[0093] Step S404: In the real-time grayscale image, mark the main outline pixels and the pixels contained in the main outline pixels as real-time main pixels.

[0094] Step S405: Establish a 3*3 pixel matrix centered on each main outline pixel, labeled as the dilation matrix; obtain pixels in the dilation matrix that are not real-time main pixels, labeled as first dilated pixels, and obtain the average grayscale value of the first dilated pixels, labeled as Pz1; establish a dilation matrix centered on each first dilated pixel, obtain pixels in the dilation matrix that are not real-time main pixels and pixels of the first dilated pixels, labeled as second dilated pixels, and obtain the average grayscale value of the second dilated pixels, labeled as Pz2, and so on, until Pz1 is obtained. t Because the real-time subject outline may be irregular in shape, it is difficult to obtain the pixels around the outer edge of the real-time subject outline. The dilation matrix method can solve this problem. The size of t is related to the number of pixels in the real-time mapping image. For example, the size of t is the minimum number of pixels required for a subject outline pixel to completely reach the non-subject area, which is when t=3.

[0095] Step S406, transfer Pz1 to Pz t Marked as Pz j , where j is an integer from 1 to t;

[0096] Step S407, calculate the average difference in evaluation: Where C1 is the average difference in evaluation;

[0097] For practical applications, please refer to Figure 2 and Figure 3 As shown, for example, Pz1 to Pz3 are obtained as 105, 96 and 85 respectively, and the average difference of evaluation is calculated as C1=10.

[0098] Step S5: Obtain the blur threshold based on the second number of database mapping images; Step S5 also includes the following sub-steps:

[0099] Step S501: Obtain a second number of images of survey points that are of the same type as the main surveying body and are clearly marked, and mark them as database surveying image maps; the database surveying image maps are images of survey points that have already been clearly marked in the existing database.

[0100] Step S502: Based on the evaluation mean difference obtained from the real-time surveying image map, obtain the evaluation mean difference of the database surveying image map and mark it as the database evaluation mean difference; Since the evaluation mean difference of the database surveying image map is the same as the evaluation mean difference obtained from the real-time surveying image map, it will not be described again here.

[0101] Step S503: Obtain the minimum value of the average difference of all database evaluations and mark it as the fuzzy threshold;

[0102] In practical applications, for example, if the average difference of the database evaluation is obtained as 9, 8, 9, 10 and 12 respectively, and the minimum value of the average difference of all database evaluations is 8, then 8 is the fuzzy threshold.

[0103] Step S6: Based on the evaluation mean difference and fuzzy threshold, determine whether the real-time mapping grayscale image is qualified. If it is not qualified, repeatedly acquire new real-time mapping images until it is qualified, and mark the qualified real-time mapping images as qualified mapping images. Step S6 also includes the following sub-steps:

[0104] Step S601: Determine whether the average difference of the evaluation is less than the fuzzy threshold. If it is less, mark the real-time mapping grayscale image as unqualified. Repeat the acquisition of new real-time mapping images until the average difference of the evaluation of the new real-time mapping images is greater than or equal to the fuzzy threshold. Mark the real-time mapping images with an average difference of evaluation greater than or equal to the fuzzy threshold as qualified mapping images.

[0105] In practical applications, if the average difference of the evaluation is greater than the fuzziness threshold of 8, the real-time surveying image is marked as a qualified surveying image. This method is used to dilate the data to obtain the changes in the gray values ​​of the pixels along the main outline. In a clear real-time surveying image, the gray value changes of the pixels along the main outline are obvious. In a fuzzy real-time surveying image, there is pixel confusion between the main part and the non-main part, resulting in no obvious changes in the gray value of the pixels along the main outline. Therefore, it is possible to determine whether an image is fuzzy based on the average difference of the evaluation.

[0106] Step S7: Obtain qualified surveying and mapping images of each detection point, and stitch together the qualified surveying and mapping images of each detection point to obtain the overall surveying and mapping report. The overall surveying and mapping report obtained by stitching together the qualified surveying and mapping images is clearer.

[0107] Example 2: This application also provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, steps such as those in the mapping method based on remote sensing big data analysis are performed to achieve the following functions: acquiring mapping points and mapping paths of the UAV generated based on the mapping area; acquiring a real-time mapping image map each time the UAV reaches a mapping point while traveling along the mapping path; performing grayscale processing on the real-time mapping image map to obtain a real-time mapping grayscale map; acquiring a first subject threshold and a second subject threshold based on a first number of mapping subject images; and acquiring a first subject threshold and a second subject threshold based on the first subject threshold and the second subject threshold. The volume threshold is used to binarize the real-time surveying grayscale image to obtain a real-time surveying binarized image. The real-time subject outline is obtained based on the real-time surveying binarized image. The evaluation mean difference is obtained based on the real-time subject outline. The fuzzy threshold is obtained based on a second number of database surveying images. The quality of the real-time surveying grayscale image is judged based on the evaluation mean difference and the fuzzy threshold. If it is not qualified, a new real-time surveying image is repeatedly obtained until it is qualified. The qualified real-time surveying image is marked as a qualified surveying image. The qualified surveying image of each detection point is obtained. The qualified surveying images of each detection point are stitched together to obtain the overall surveying report image.

[0108] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0109] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the mapping method based on remote sensing big data analysis provided by the above methods. The method includes: acquiring mapping points and mapping paths of a UAV generated based on the mapping area; acquiring a real-time mapping image map each time the UAV reaches a mapping point while traveling along the mapping path; performing grayscale processing on the real-time mapping image map to obtain a real-time mapping grayscale map; and acquiring a first subject threshold based on a first number of mapping subject maps. Based on the first and second subject thresholds, the real-time grayscale image is binarized to obtain a real-time binary image. The real-time subject outline is obtained based on the real-time binary image. The evaluation mean difference is obtained based on the real-time subject outline. The fuzzy threshold is obtained based on the second number of database mapping images. The real-time grayscale image is judged to be qualified based on the evaluation mean difference and the fuzzy threshold. If it is not qualified, a new real-time mapping image is repeatedly obtained until it is qualified. The qualified real-time mapping image is marked as a qualified mapping image. The qualified mapping image of each detection point is obtained. The qualified mapping images of each detection point are stitched together to obtain the overall mapping report image.

[0110] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the above-mentioned mapping method based on remote sensing big data analysis to achieve the following functions: acquiring mapping points and mapping paths of a UAV generated based on the mapping area; acquiring a real-time mapping image map each time the UAV reaches a mapping point while traveling along the mapping path; performing grayscale processing on the real-time mapping image map to obtain a real-time mapping grayscale map; acquiring a first subject threshold and a second subject threshold based on a first number of mapping subject images; and based on the first subject... The real-time grayscale image is binarized using a threshold and a second subject threshold to obtain a real-time binary image. The real-time subject outline is then obtained based on the real-time binary image. The evaluation mean difference is obtained based on the real-time subject outline. A fuzzy threshold is obtained based on a second number of database mapping images. The real-time grayscale image is judged to be qualified based on the evaluation mean difference and the fuzzy threshold. If it is not qualified, a new real-time mapping image is repeatedly obtained until it is qualified. The qualified real-time mapping image is marked as a qualified mapping image. Qualified mapping images for each detection point are obtained, and the qualified mapping images for each detection point are stitched together to obtain the overall mapping report image.

[0111] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0112] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A mapping method based on remote sensing big data analysis, characterized in that, Includes the following steps: Obtain the mapping points and mapping path of the UAV generated based on the mapping area; As the drone travels along the mapping path, it acquires a real-time mapping image at each mapping point it reaches. The real-time mapping image is converted to grayscale to obtain a real-time mapping grayscale image. A first subject threshold and a second subject threshold are obtained based on a first number of mapping subject images. The real-time mapping grayscale image is binarized based on the first subject threshold and the second subject threshold to obtain a real-time mapping binarized image. The real-time subject outline is obtained based on the real-time mapping binarized image. The average error is judged based on the real-time subject outline; The blur threshold is obtained based on the second number of database image maps; The evaluation mean difference and fuzzy threshold are used to determine whether the real-time mapping grayscale image is qualified. If it is not qualified, new real-time mapping images are repeatedly acquired until it is qualified. The qualified real-time mapping images are marked as qualified mapping images. Obtain qualified survey images for each survey point, and stitch together the qualified survey images for each survey point to obtain the overall survey report map; Obtaining the first subject threshold and the second subject threshold based on the first number of surveyed subject images includes the following sub-steps: Obtain the first number of images containing only the main body of the survey, and mark them as the main body of the survey; The main survey map is converted to grayscale to obtain a grayscale image of the main survey map. Obtain the grayscale value of each pixel in the grayscale image of the main subject and mark it as the grayscale value of the main subject. The grayscale values ​​of the surveyed subjects are sorted from smallest to largest and labeled as Lym1 to Lym. i ; The number of grayscale values ​​of the main subject obtained is marked as the grayscale number of the main subject. The first position ratio is calculated as: Wzb1 = m1 * Hzt, where Wzb1 is the first position ratio, m1 is the first position coefficient, the range of m1 is (0, 0.5), and Hzt is the gray level of the subject being surveyed. If the ratio of the first position is an integer, then Lym （Wzb1） Mark it as the first position value; if the first position ratio is not an integer, obtain the integers on the left and right sides of Wzb1, and mark them as Wzb1n and Wzb1x respectively, and calculate Lym. （Wzb1n） With Lym （Wzb1x） The mean value is marked as the first position value; The second position ratio is obtained as: Wzb2 = m2 * Hzt, where Wzb2 is the second position ratio, m2 is the second position coefficient, and the range of m2 is [0.5, 1). If the ratio of the second position is an integer, then Lym （Wzb2） Mark the value as the second position value; if the second position ratio is not an integer, obtain the integers on the left and right sides of Wzb2, and mark them as Wzb2n and Wzb2x respectively, and calculate Lym. （Wzb2n） With Lym （Wzb2x） The mean value is marked as the second position value; Create a u*u pixel grid at one corner of the grayscale image of the main subject and mark it as the average pixel grid. Move the average pixel grid with a step size of N. The initial movement direction is along one side of the grayscale image of the main subject. When it moves to the edge of the grayscale image of the main subject, it moves one step along this edge. Then it moves in the opposite direction of the initial movement direction. When it moves to the edge of the grayscale image of the main subject again, it moves one step along this edge. Repeat the above operation until it can no longer move in the grayscale image of the main subject. The average grayscale value of the main subject within the average pixel square is obtained for each movement and marked as the average grayscale value of the square. Obtain the number of average gray values ​​of the squares between the first position value and the second position value, and mark them as the number of intermediate distributions; Get the number of squares with average gray values ​​less than the first position value, and mark them as the smaller distribution number; Get the number of squares with an average gray value greater than the first position value, and mark them as the number of larger distributions; The proportion of the first distribution is calculated as: B1 = Fn / Fz; where B1 is the proportion of the first distribution, Fn is the number of the smaller distribution, and Fz is the number of the middle distribution. The proportion of the second distribution is calculated as: B2 = Fx / Fz; where B2 is the proportion of the second distribution and Fx is the number of the larger distribution. The threshold for the first subject is calculated as: Lz1 = Wzz1 - B1 * (Wzz2 - Wzz1); The second subject threshold is calculated as: Lz2 = Wzz2 + B2 * (Wzz2 - Wzz1); where Lz1 is the first subject threshold, Lz2 is the second subject threshold, Wzz1 is the first position value, and Wzz2 is the second position value; Binarizing the real-time mapping grayscale image based on the first subject threshold and the second subject threshold to obtain the real-time mapping binarized image includes the following sub-steps: Set the grayscale values ​​of the surveying subjects in the real-time surveying grayscale image that are greater than or equal to the first subject threshold and less than or equal to the second subject threshold to 0, and set the grayscale values ​​of the surveying subjects in the real-time surveying grayscale image that are less than the first subject threshold or greater than the second subject threshold to 255 to obtain a real-time surveying binarized image.

2. The mapping method based on remote sensing big data analysis according to claim 1, characterized in that, Obtaining a real-time grayscale image by converting a real-time surveyed image to grayscale includes the following sub-steps: Obtain the RGB value of each pixel in the real-time mapping image and mark it as the real-time mapping RGB value; The weighted average grayscale conversion formula is used to convert all real-time RGB values ​​into grayscale values ​​to obtain a real-time grayscale image.

3. The mapping method based on remote sensing big data analysis according to claim 2, characterized in that, Obtaining the real-time subject outline based on the real-time mapping binarized image includes the following sub-steps: In the real-time mapping binarized image, pixels with a grayscale value of 0 are marked as main pixels. The largest continuous region composed of main pixels is marked as the main region. The part adjacent to the region with a grayscale value of 255 pixels in the real-time mapping binarized image is marked as the real-time main outline.

4. The mapping method based on remote sensing big data analysis according to claim 3, characterized in that, The evaluation mean difference based on the real-time subject contour includes the following sub-steps: Establish a Cartesian coordinate system, labeled as the first analysis coordinate system. Place the real-time mapping binarized image in the first quadrant of the first analysis coordinate system, while ensuring that the two sides of the real-time mapping binarized image coincide with the horizontal and vertical axes of the first analysis coordinate system, respectively, to obtain the coordinates of the real-time main outline. Establish a Cartesian coordinate system, labeled as the second analysis coordinate system, and place the real-time grayscale image in the first quadrant, while ensuring that the two sides of the real-time grayscale image coincide with the horizontal and vertical axes of the second analysis coordinate system, respectively. The real-time subject outline is drawn on the real-time grayscale map based on the coordinates of the real-time subject outline. Obtain the pixels on the real-time subject outline in the real-time grayscale image and mark them as subject outline pixels; In the real-time grayscale image, the main outline pixels and the pixels contained in the main outline pixels are marked as real-time main pixels. A 3x3 pixel matrix is ​​established centered on each main outline pixel, labeled as the dilation matrix. Pixels in the dilation matrix that are not real-time main pixels are identified and labeled as first-dilated pixels. The average grayscale value of the first-dilated pixels is obtained and labeled as Pz1. A dilation matrix is ​​then established centered on each first-dilated pixel. Pixels in the dilation matrix that are not real-time main pixels and the first-dilated pixels are identified and labeled as second-dilated pixels. The average grayscale value of the second-dilated pixels is obtained and labeled as Pz2, and so on, until Pz1 is obtained. t ; Transfer Pz1 to Pz t Marked as Pz j , where j is an integer from 1 to t; The mean difference of the evaluation is calculated as follows: C1 represents the average difference in evaluation.

5. The mapping method based on remote sensing big data analysis according to claim 4, characterized in that, Obtaining the blur threshold based on the second number of database mapping images includes the following sub-steps: Acquire a second number of images of the same type as the main body of the survey and clearly marked survey points, and mark them as database survey image maps; The step of obtaining the evaluation mean difference based on real-time surveying and mapping images is to obtain the evaluation mean difference of the database surveying and mapping images and mark it as the database evaluation mean difference; Find the minimum average difference among all database evaluations and mark it as the fuzzy threshold.

6. The mapping method based on remote sensing big data analysis according to claim 5, characterized in that, The process of determining whether a real-time grayscale image is qualified based on the evaluation mean difference and fuzzy threshold is as follows: If it is not qualified, new real-time grayscale images are repeatedly acquired until they are qualified. Marking qualified real-time grayscale images as qualified images includes the following sub-steps: Determine if the average difference of the evaluation is less than the fuzzy threshold. If it is less, mark the real-time grayscale image as unqualified. Repeat the acquisition of new real-time image images until the average difference of the evaluation of the new real-time image images is greater than or equal to the fuzzy threshold. Mark the real-time image images with an average difference of evaluation greater than or equal to the fuzzy threshold as qualified image images.

Images