Land resource land type pattern spot classification rapid sampling inspection and analysis method

Abstract

The invention provides a land resource land type pattern spot classification rapid sampling inspection and analysis method. The method comprises the following steps: S1, obtaining a land resource land type vertical aerial view image shot by an aerial camera; s2, performing land type contour division on the land type vertical aerial view of the territorial resources in the step S1; s3, performing territorial resource land class labeling on the territorial resource land class vertical aerial view after land class contour division in the step S2; and S4, classifying the vertical aerial view of the land type of the territorial resource marked in the step S3 for rapid sampling inspection and verification. According to the invention, rapid verification and identification can be carried out according to the casual inspection graph, and the efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of land classification sampling inspection, and in particular to a method for rapid sampling inspection, verification and analysis of land resource land classification map classification. Background Art

[0002] As the basic unit for confirming land ownership and dividing land features, map patches are a basic term in cartography. They refer to single-class land parcels with the same or similar landforms and land use types, and are divided by administrative boundaries, land ownership boundaries or linear features. The definition and nature of map patches determine that their boundaries and shapes are often irregular. The conventional verification methods are:

[0003] The first step is to obtain a vertical bird's-eye view of land resources and land types taken by an aerial camera;

[0004] The second step is to divide the land type contours of the vertical bird's-eye view of the land resources in the first step;

[0005] The third step is to label the land resources categories on the vertical bird's-eye view of the land resources categories after the land category outlines are divided in the second step;

[0006] The fourth step is to verify the marked vertical bird's-eye view of land resources types.

[0007] The first three steps are the preliminary sorting steps of land classification, and the fourth step is the verification step. During the verification, it is necessary to judge whether the land classification in the entire bird's-eye view corresponds to the annotation, which may lead to omissions of annotations and is very inefficient. Therefore, this patent solution was created. Summary of the invention

[0008] The present invention aims to at least solve the technical problems existing in the prior art, and in particular innovatively proposes a method for rapid sampling, verification and analysis of land resource land classification map patches.

[0009] In order to achieve the above-mentioned object of the present invention, the present invention provides a method for rapid sampling inspection and analysis of land resources land classification spots, comprising the following steps:

[0010] S1, obtain a vertical bird's-eye view of land resources and land types taken by an aerial camera;

[0011] S2, dividing the land resource classification vertical bird's-eye view map in step S1 into land classification contours;

[0012] S3, marking the land and resources categories on the vertical bird's-eye view of the land and resources categories after the land category outlines are divided in step S2;

[0013] S4, classifying the vertical bird's-eye view of the land resources categories marked in step S3 for rapid sampling and verification.

[0014] In a preferred embodiment of the present invention, step S1 is:

[0015] Obtain M vertical bird's-eye views of land resources taken by an aerial camera, where M is a positive integer greater than or equal to 2;

[0016] Splice M vertical bird's-eye view maps of land and resources classification into one vertical bird's-eye view map of land and resources classification.

[0017] In a preferred embodiment of the present invention, in step S3, the land resource land types include one or any combination of reservoir water surface land types, dry land land types, paddy field land types, tree forest land types, shrub forest land types, hydraulic construction land types, rural residential land types, pond water surface land types, bamboo forest land types, and rural road land types.

[0018] In a preferred embodiment of the present invention, the area photographed by the drone in step S1 is an area based on groups / communities, villages / communities, towns / townships / streets, and districts / counties.

[0019] In a preferred embodiment of the present invention, the land type contour division in step S2 is the land type contour division performed on the vertical bird's-eye view of the land resources land type by drawing software.

[0020] In a preferred embodiment of the present invention, the method for classifying the marked vertical bird's-eye view of land and resources in step S4 comprises the following steps:

[0021] S41, count the number of land resources categories, recorded as N, which are the first land resources category, the second land resources category, the third land resources category, ..., the Nth land resources category;

[0022] The number of contour areas after the land category contour division of the vertical bird's-eye view of the land resources land category is counted, denoted as K, which are the first contour land category area, the second contour land category area, the third contour land category area, ..., the Kth contour land category area, and the area of ​​each contour land category area is S 1 , S 2 , S 3 ,……,S K ; S 1 is the area of ​​the first contour land category, S 2 is the area of ​​the second contour land type area, S 3 is the area of ​​the third contour land category area, S K is the area of ​​the Kth contour land category area;

[0023] Arrange the area of ​​each contour land type in order from small to large as follows:

[0024] S′ min , S′min-1 , S′ min-2 , S′ min-K+1 ,

[0025] Among them, S′ min It means the area of ​​each contour land type area is arranged in order from small to large, and the area at the first place;

[0026] S′ min-1 It means the area of ​​the second place after arranging the areas of each contour land type in order from small to large;

[0027] S′ min-2 It means the area at the third position after arranging the areas of each contour land type in order from small to large;

[0028] S′ min-K+1 It means the area at the Kth position after arranging the areas of each contour land type in order from small to large;

[0029] S42, setting the size of the sampling frame to φ×φ and the number of samplings of each land type contour division area and N sampling maps; N is the number of land resource land types;

[0030] The calculation method for the number of random inspections is:

[0031]

[0032] Among them, O k Indicates the number of random inspections of the kth contour land type area;

[0033] S k is the area of ​​the kth contour land class area;

[0034] S′ min It means the area of ​​each contour land type area is arranged in order from small to large, and the area at the first place;

[0035] represents the ceiling function;

[0036] <> means taking the decimal part;

[0037] Z represents an integer;

[0038] The size of each sampling image is They represent the number of sizes of sampling frames that can be placed in the sampling image horizontally and vertically respectively; from left to right and from top to bottom, they are the 11th pure white image, the 12th pure white image, the 13th pure white image, ..., the 1φ pure white image, which is the 1st row of pure white images;

[0039] The 21st pure white image, the 22nd pure white image, the 23rd pure white image, ..., the 2φ pure white image, which is the second row of pure white images;

[0040] The 31st pure white image, the 32nd pure white image, the 33rd pure white image, ..., the 3rdφ pure white image, which is the 3rd row of pure white images;

[0041] ……;

[0042] φ1th pure white image, φ2th pure white image, φ3th pure white image, ..., φφth pure white image, which are the φth row of pure white images;

[0043] S43, dividing the vertical bird's-eye view map of land and resources land types after the land type contour division into rows and columns of a sampling frame, wherein the number of pixels between row lines is equal to the number of vertical pixels of the sampling frame, and the number of pixels between column lines is equal to the number of horizontal pixels of the sampling frame. If the number of row pixels or column pixels of the last row or the last column or the first row or the first column is less than the number of horizontal pixels of the sampling frame or the number of vertical pixels of the sampling frame, then the row or the column may be retained or discarded;

[0044] When retaining, corresponding row pixels or column pixels may be added before the corresponding first row or first column or after the last row or last column, so that the number of row pixels or column pixels of the last row or last column or the first row or first column is equal to the number of horizontal pixels of the sampling frame or the number of vertical pixels of the sampling frame; the pixel value of the added row pixels or column pixels is pure white;

[0045] S44, select the contour area of ​​the vertical bird's-eye view of the land resources land category after the land category contour is divided according to the set order, randomly select one or more sampling frames in the selected contour area, if the sampling frames do not completely belong to the selected contour area, then translate the selected sampling frame up and down or left and right or reduce the size of the sampling frame to the size of the selected sampling frame with the center of the selected sampling frame as the center. To reduce the number of pixels, is an even number greater than 1, so that the selected inspection box is completely within the selected contour area;

[0046] S45, overlaying the image in the selected sampling frame onto the sampling map of the corresponding land category, and the center of the image in the selected sampling frame coincides with the center of the overlaid sampling frame, so that the image in the selected sampling frame is completely within the sampling frame in the sampling map;

[0047] S46, executing steps S44 to S45 until all contour areas in the vertical bird's-eye view of the land resources land categories after the land category contours are divided are selected.

[0048] In a preferred embodiment of the present invention, step S5 is also included after step S4. When viewing the sampling map, if a sampling box in the sampling map is selected and viewed, it jumps to the location of the vertical bird's-eye view of the land and resources classification marked corresponding to the image in the sampling box.

[0049] In a preferred embodiment of the present invention, step S5 includes the following steps:

[0050] S51, determining whether a trigger signal of a sampling box in the sampling map is received:

[0051] If a trigger signal of a sampling box in the selected sampling map is received, the next step is executed;

[0052] If no trigger signal for selecting a sampling frame in the sampling map is received, step S51 is executed;

[0053] S52, determining whether a trigger signal for viewing a certain sampling box in the selected sampling map is received:

[0054] If a trigger signal for viewing a certain sampling box in the selected sampling map is received, the next step is executed;

[0055] If no trigger signal for viewing a certain sampling frame in the selected sampling map is received, step S52 is executed;

[0056] S53, obtaining the link pixel coordinates of the image in a certain sampling frame in the selected sampling image, and determining whether the link pixel coordinates of the image in a certain sampling frame in the selected sampling image are obtained:

[0057] If the coordinates of the linked pixel points of the image in a certain sampling frame in the selected sampling image are obtained, step S54 is executed;

[0058] If the link pixel coordinates of the image in a certain sampling frame in the selected sampling image are not obtained, step S53 is executed;

[0059] S54, searching, according to the link pixel point coordinates, for an image position corresponding to the vertical bird's-eye view of the land resources category where the link pixel point coordinates are located.

[0060] The present invention also discloses a computer system, comprising:

[0061] processor;

[0062] a memory for storing processor-executable instructions;

[0063] Wherein, the processor is configured to implement the method for rapid sampling, verification and analysis of land resources land classification map classification when executing the executable instructions.

[0064] The present invention also discloses a computer-readable storage medium, comprising:

[0065] a memory having a computer program stored thereon;

[0066] A processor is used to execute the program in the memory to implement the land resources land classification map rapid sampling verification and analysis method.

[0067] In summary, due to the adoption of the above technical solution, the present invention can quickly verify and identify according to the sampling diagram, thereby improving efficiency and accuracy.

[0068] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0070] Figure 1 It is a schematic block diagram of the process of the present invention.

[0071] Figure 2 It is a bird's-eye view schematic diagram taken by the aerial photography machine of the present invention.

[0072] Figure 3 It is a schematic diagram of the land type contour division of the present invention.

[0073] Figure 4 This invention is Figure 3 Time marking diagram.

[0074] Figure 5 It is a schematic diagram of land classification marking according to the present invention.

[0075] Figure 6 It is a schematic diagram of a sampling inspection frame formed by dividing rows and columns in the present invention.

[0076] Figure 7 The present invention Figure 6 Remove the land class labeling diagram.

[0077] Figure 8 It is a schematic display diagram of the dry land sampling map after the sampling inspection of the present invention.

[0078] Fig. 9 The present invention randomly selects a rural homestead sampling map to show a schematic display diagram.

[0079] Fig.10 The present invention provides two schematic diagrams of rural homestead sampling maps.

[0080] Fig.11The present invention provides a schematic display of three rural homestead sampling maps.

[0081] Fig.12 It is a schematic display diagram of the sampling map of rural homesteads after the sampling inspection of the present invention.

[0082] Fig.13 This invention is Fig.12 Time marking diagram.

[0083] Fig.14 It is a schematic display diagram of the paddy field sampling diagram after the sampling inspection of the present invention. DETAILED DESCRIPTION

[0084] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0085] The present invention discloses a method for rapid sampling, checking and analyzing land resource land classification spots. Figure 1 As shown, the following steps are included:

[0086] S1, obtain a vertical bird's-eye view of land resources and land types taken by an aerial camera;

[0087] S2, dividing the land resource classification vertical bird's-eye view map in step S1 into land classification contours;

[0088] S3, marking the land and resources categories on the vertical bird's-eye view of the land and resources categories after the land category outlines are divided in step S2;

[0089] S4, classifying the vertical bird's-eye view of the land resources categories marked in step S3 for rapid sampling and verification.

[0090] In a preferred embodiment of the present invention, step S1 is:

[0091] Obtain M vertical bird's-eye views of land resources taken by an aerial camera, where M is a positive integer greater than or equal to 2;

[0092] Splice M vertical bird's-eye view maps of land and resources classification into one vertical bird's-eye view map of land and resources classification.

[0093] In a preferred embodiment of the present invention, in step S3, the land resource land types include one or any combination of reservoir water surface land types, dry land land types, paddy field land types, tree forest land types, shrub forest land types, hydraulic construction land types, rural residential land types, pond water surface land types, bamboo forest land types, and rural road land types.

[0094] In a preferred embodiment of the present invention, the area photographed by the drone in step S1 is an area based on groups / communities, villages / communities, towns / townships / streets, and districts / counties.

[0095] In a preferred embodiment of the present invention, in step S2, the land class contour division is performed on the vertical bird's-eye view of the land resources land class by using a drawing software (such as Photoshop, Adobe Illustrator, CorelDRAW, etc., to outline the land class contour of the corresponding land class using a brush tool).

[0096] In a preferred embodiment of the present invention, the method for classifying the marked vertical bird's-eye view of land and resources in step S4 comprises the following steps:

[0097] S41, count the number of land resource categories, denoted as N, which are the first land resource category, the second land resource category, the third land resource category, ..., the Nth land resource category; in this embodiment, N=10, and the corresponding first land resource category is dry land, the second land resource category is rural residential land, the third land resource category is paddy field, the fourth land resource category is reservoir water surface, the fifth land resource category is tree forest, the sixth land resource category is shrub forest, the seventh land resource category is hydraulic construction land, the eighth land resource category is pond water surface, the ninth land resource category is bamboo forest, and the tenth land resource category is rural road.

[0098] The number of contour areas after the land category contour division of the vertical bird's-eye view of the land resources land category is counted, denoted as K, which are the first contour land category area, the second contour land category area, the third contour land category area, ..., the Kth contour land category area, and the area of ​​each contour land category area is S 1 , S 2 , S 3 ,……,S K ; S 1 is the area of ​​the first contour land category, S 2 is the area of ​​the second contour land type area, S 3 is the area of ​​the third contour land category area, S K is the area of ​​the Kth contour land category area;

[0099] Arrange the area of ​​each contour land type in order from small to large as follows:

[0100] S′ min , S′ min-1 , S′ min-2 , S′ min-K+1 ,

[0101] Among them, S′min It means the area of ​​each contour land type area is arranged in order from small to large, and the area at the first place;

[0102] S′ min-1 It means the area of ​​the second place after arranging the areas of each contour land type in order from small to large;

[0103] S′ min-2 It means the area at the third position after arranging the areas of each contour land type in order from small to large;

[0104] S′ min-K+1 It means the area at the Kth position after arranging the areas of each contour land type in order from small to large;

[0105] S42, set the size of the sampling frame (φ×φ, preferably 31×31, that is, 31 pixels horizontally and 31 pixels vertically, a square sampling frame of 31 pixels) and the number of samplings of each land type contour division area and N sampling maps; N is the number of land resources land types; the calculation method of the number of samplings is:

[0106]

[0107] Among them, O k Indicates the number of random inspections of the kth contour land type area;

[0108] S k is the area of ​​the kth contour land class area;

[0109] S′ min It means the area of ​​each contour land type area is arranged in order from small to large, and the area at the first place;

[0110] represents the ceiling function;

[0111] <> means taking the decimal part;

[0112] Z represents an integer;

[0113] The size of each sampling image is The preferred resolution is 310×310. Respectively represent the number of sampling frames that can be placed horizontally and vertically in the sampling image; the sampling image is composed of 10×10 pure white images of the size of the sampling frame, and the size of each pure white image is consistent with the size of the sampling frame. From left to right, from top to bottom (set order, the setting order can also be from top to bottom, from left to right, or other orders), they are the 11th pure white image, the 12th pure white image, the 13th pure white image, ..., the 1φ pure white image, which is the 1st row of pure white images;

[0114] The 21st pure white image, the 22nd pure white image, the 23rd pure white image, ..., the 2φ pure white image, which is the second row of pure white images;

[0115] The 31st pure white image, the 32nd pure white image, the 33rd pure white image, ..., the 3rdφ pure white image, which is the 3rd row of pure white images;

[0116] ……;

[0117] φ1th pure white image, φ2th pure white image, φ3th pure white image, ..., φφth pure white image, which are the φth row of pure white images;

[0118] When φ is 10, from left to right and from top to bottom, they are the 11th pure white image, the 12th pure white image, the 13th pure white image, ..., the 110th pure white image, which is the first row of pure white images;

[0119] The 21st pure white image, the 22nd pure white image, the 23rd pure white image, ..., the 210th pure white image, which are the second row of pure white images;

[0120] The 31st pure white image, the 32nd pure white image, the 33rd pure white image, ..., the 310th pure white image, which are the 3rd row of pure white images;

[0121] ……;

[0122] The 101st pure white image, the 102nd pure white image, the 103rd pure white image, ..., the 1010th pure white image, which is the 10th row of pure white images;

[0123] S43, divide the vertical bird's-eye view of the land resources classification after the land classification contour division in step S2 into rows and columns for sampling frames, wherein the number of pixels between row lines is equal to the number of vertical pixels in the sampling frame, and the number of pixels between column lines is equal to the number of horizontal pixels in the sampling frame. If the number of row pixels or column pixels of the last row or the last column (the first row or the first column) is less than the number of horizontal pixels or the number of vertical pixels in the sampling frame, then the row or the column can be retained or discarded.

[0124] When retaining, corresponding row pixels or column pixels can be added before the corresponding first row or first column or after the last row or last column, so that the number of row pixels or column pixels of the last row or last column (first row or first column) is equal to the number of horizontal pixels of the sampling frame or the number of vertical pixels of the sampling frame; the pixel value of the added row pixel or column pixel is pure white.

[0125] S44, select the contour area of ​​the vertical bird's-eye view of the land resources land category after the land category contour division in step S2 according to the set order (the set order can be from left to right, from top to bottom, from top to bottom, from left to right, or randomly), and randomly select one or more sampling frames in the selected contour area (the number of sampling frames here is determined by the number of samplings of each land category contour division area). If the sampling frames do not completely belong to the selected contour area, the selected sampling frame is translated up, down, left, and right (one pixel distance each time) or the size of the sampling frame is reduced to the center of the selected sampling frame. To reduce the number of pixels, is an even number greater than 1, so that the selected inspection box is completely within the selected contour area;

[0126] S45, copy (overlay) the image in the selected sampling frame to the sampling map of the corresponding land class category, and the center of the image in the selected sampling frame coincides with the center of the overlaid sampling frame, so that the image in the selected sampling frame is completely in the sampling frame in the sampling map; this step also includes obtaining the link pixel point coordinates of the image in the selected sampling frame, and storing the obtained link pixel point coordinates in the image in the selected sampling frame, so as to facilitate the subsequent query based on the pixel point coordinates to obtain the position of the image in the vertical bird's-eye view of the land and resources category, so as to view the land class labeling.

[0127] S46, executing steps S44 to S45 until all contour areas in the vertical bird's-eye view of the land resources land category after the land category contour division in step S2 are selected.

[0128] In a preferred embodiment of the present invention, step S5 is also included after step S4. When viewing the sampling map, if a sampling box in the sampling map is selected and viewed, the program jumps to the location of the vertical bird's-eye view of the land resources classification marked in step S3 corresponding to the image in the sampling box to determine whether the classification is correct.

[0129] In a preferred embodiment of the present invention, step S5 includes the following steps:

[0130] S51, determining whether a trigger signal of a sampling box in the sampling map is received:

[0131] If a trigger signal of a sampling box in the selected sampling map is received, the next step is executed;

[0132] If no trigger signal for selecting a sampling frame in the sampling map is received, step S51 is executed;

[0133] S52, determining whether a trigger signal for viewing a certain sampling box in the selected sampling map is received:

[0134] If a trigger signal for viewing a certain sampling box in the selected sampling map is received, the next step is executed;

[0135] If no trigger signal for viewing a certain sampling frame in the selected sampling map is received, step S52 is executed;

[0136] S53, obtaining the link pixel coordinates of the image in a certain sampling frame in the selected sampling image, and determining whether the link pixel coordinates of the image in a certain sampling frame in the selected sampling image are obtained:

[0137] If the coordinates of the linked pixel points of the image in a certain sampling frame in the selected sampling image are obtained, step S54 is executed;

[0138] If the link pixel coordinates of the image in a certain sampling frame in the selected sampling image are not obtained, step S53 is executed;

[0139] S54, searching, according to the link pixel point coordinates, for an image position corresponding to the vertical bird's-eye view of the land resources category where the link pixel point coordinates are located;

[0140] The step S55 is also included. After searching the image position corresponding to the vertical bird's-eye view of the land resources land category where the link pixel point coordinates are located according to the link pixel point coordinates, the image contained in the contour area where the image position is located can be judged by the inspector whether the land category corresponding to the image contained in the contour area is the land category marked in the contour area:

[0141] If the inspector determines that the land category corresponding to the image contained in the contour area is not the land category marked in the contour area, the contour area will be marked (the mark may be marking the contour area with the text "to be confirmed"), or the image in a certain inspection box in the inspection map selected for viewing will be marked (the mark may be marking the contour area with the text "to be confirmed"), so as to facilitate subsequent searching.

[0142] The present invention discloses a method for rapid sampling, checking and analyzing land resource land classification spots, which specifically comprises the following steps:

[0143] The first step is to obtain a vertical bird's-eye view of land resources taken by an aerial camera; Figure 2 shown.

[0144] The second step is to use Photoshop's brush tool to outline the land class contours of the vertical bird's-eye view of the land resources land class in the first step and divide the land class contours; Figure 3 As shown in the figure, there are 88 land class contour areas.

[0145] The third step is to label the land resources categories on the vertical bird's-eye view of the land resources categories after the land category outlines are divided in the second step; Figure 5 As shown in the figure, there is 1 land type outline area for the reservoir water surface land type, 20 land type outline areas for the dry land land type, 15 land type outline areas for the paddy field land type, 14 land type outline areas for the arbor forest land type, 6 land type outline areas for the shrub forest land type, 1 land type outline area for the hydraulic construction land type, 20 land type outline areas for the rural residential land type, 2 land type outline areas for the pond water surface land type, 7 land type outline areas for the bamboo forest land type, and 2 land type outline areas for the rural road land type.

[0146] The fourth step is to count the number of land resource categories, denoted as N=10, namely, the first land resource category is dry land, the second land resource category is rural residential land, the third land resource category is paddy field, the fourth land resource category is reservoir water surface, the fifth land resource category is tree forest, the sixth land resource category is shrub forest, the seventh land resource category is hydraulic construction land, the eighth land resource category is pond water surface, the ninth land resource category is bamboo forest, and the tenth land resource category is rural road.

[0147] The fifth step is to count the number of contour areas after the land category contour division of the vertical bird's-eye view of the land resources land category, denoted as K = 88, which are the first contour land category area, the second contour land category area, the third contour land category area, ..., the 88th contour land category area, and the area of ​​each contour land category area is S 1 , S 2 , S 3 ,……,S 88 ;

[0148] The 88 contour land types are arranged in order from small to large:

[0149] S′ min , S′ min-1 , S′ min-2 , S′ min-K+1 ,

[0150] The sixth step is to set the size of the sampling frame to 31×31 and 10 sampling maps; the 10 sampling maps are: the first sampling map is the dry land classification sampling map, the second sampling map is the rural homestead land classification sampling map, the third sampling map is the paddy field land classification sampling map, the fourth sampling map is the reservoir water surface land classification sampling map, the fifth sampling map is the tree forest land classification sampling map, the sixth sampling map is the shrub forest land classification sampling map, the seventh sampling map is the hydraulic construction land classification sampling map, the eighth sampling map is the pond water surface land classification sampling map, the ninth sampling map is the bamboo forest land classification sampling map, and the tenth sampling map is the rural road land classification sampling map.

[0151] The size of each sampling image is 310×310;

[0152] The seventh step is to divide the vertical bird's-eye view of the land resources and resources marked in the third step into rows and columns for random inspection, such as Figure 7 As shown, the number of pixels between rows is equal to the number of vertical pixels in the sampling frame, and the number of pixels between columns is equal to the number of horizontal pixels in the sampling frame. Alternatively, the vertical bird's-eye view of the land resources land category after the land category contour division in the second step is divided into rows and columns for the sampling frame, as shown in Figure 6 shown.

[0153] In the eighth step, the vertical bird's-eye view map of the land resources land category after the land category contour division in the seventh step is randomly selected, and one or more random inspection frames are randomly selected in the selected contour area. If the random inspection frames do not completely belong to the selected contour area, the selected random inspection frames are moved up, down, left, and right by one pixel each time, so that the selected random inspection frames are completely within the selected contour area;

[0154] In the ninth step, if the land category marked in the image in the selected sampling frame is the homestead land category, the link pixel coordinates of the image in the selected sampling frame are obtained and the link pixel coordinates are stored in the image in the selected sampling frame, and the image is overlaid on the 11th pure white image in the second sampling map (rural homestead land category sampling map), such as Fig. 9 shown.

[0155] Step 10. If the land category of the image in the sampling frame selected again is the homestead land category, then obtain the link pixel coordinates of the image in the selected sampling frame and store the link pixel coordinates in the image in the selected sampling frame, and overlay the image on the 12th pure white image in the second sampling map (rural homestead land category sampling map), as shown in Fig.10 shown.

[0156] In the eleventh step, if the land category marked in the image in the sampling frame selected again is also the homestead land category, then obtain the link pixel coordinates of the image in the selected sampling frame and store the link pixel coordinates in the image in the selected sampling frame, and overlay the image on the 13th pure white image in the second sampling map (rural homestead land category sampling map), as shown in Fig.11 shown.

[0157] ……;

[0158] In the twelfth step, after all the contour areas in the vertical bird's-eye view of the land resources land category after the land category contour division are selected, Fig.12 Shown is a schematic diagram of the random inspection of rural homesteads after the inspection.

[0159] like Figure 8 Shown is a schematic display of the dry land sampling map after the sampling inspection.

[0160] like Fig.14 Shown is a schematic diagram of the paddy field inspection after random inspection.

[0161] Other sampling maps are not displayed, such as reservoir water surface land classification sampling maps, tree forest land classification sampling maps, shrub forest land classification sampling maps, hydraulic construction land classification sampling maps, pond water surface land classification sampling maps, bamboo forest land classification sampling maps, rural road land classification sampling maps, etc.

[0162] The following steps are verification steps (based on the rural homestead sampling map):

[0163] The first step is to open Fig.12 The rural housing land sampling map shown;

[0164] In the second step, if you select the sampling box in the first row and fifth column of the sampling map; Fig.13 shown.

[0165] In the third step, if you choose to view the sampling frame in the first row and fifth column of the selected sampling image (at this time, the sampling inspector may think that the image does not belong to the rural homestead land category image); then obtain the link pixel point coordinates of the image in the sampling frame in the first row and fifth column of the selected sampling image;

[0166] The fourth step is to obtain the link pixel coordinates, and then search for the image position corresponding to the vertical bird's-eye view of the land resources category where the link pixel coordinates are located according to the link pixel coordinates; Figure 4 shown.

[0167] In the fifth step, the inspector can determine whether the land category corresponding to the image contained in the contour area is the land category marked in the contour area:

[0168] If the land category corresponding to the image contained in the contour area is not the land category marked in the contour area, the contour area can be marked, or the image in the sampling box in the 1st row and 5th column of the selected sampling map can be marked.

[0169] The present invention also discloses a computer system, comprising:

[0170] processor;

[0171] a memory for storing processor-executable instructions;

[0172] Wherein, the processor is configured to implement the method for rapid sampling, verification and analysis of land resources land classification map classification when executing the executable instructions.

[0173] The present invention also discloses a computer-readable storage medium, comprising:

[0174] a memory having a computer program stored thereon;

[0175] A processor is used to execute the program in the memory to implement the land resources land classification map rapid sampling verification and analysis method.

[0176] In a preferred embodiment of the present invention, step S1 also includes the following steps:

[0177] S11, convert the vertical bird's-eye view of land resources classification into a land survey slope classification map;

[0178] S12, smoothing the slope classification map of the land survey;

[0179] S13, displays the processed land survey slope classification map.

[0180] In a preferred embodiment of the present invention, the method for performing smoothness processing on the land survey slope classification map in step S12 comprises the following steps:

[0181] S121, setting the pixel search box size to P×P;

[0182] S122, determining the position of the land survey slope classification map where the current pixel search frame is located; the pixel coordinates of the center point of the land survey slope classification map where the current pixel search frame is located are recorded as (i, j), and the pixel coordinates of the land survey slope classification map where the current pixel search frame is located are determined as:

[0183]

[0184] It can also be written as:

[0185]

[0186] The pixel coordinates of the left side of the land survey slope classification map where the current pixel search box is located are determined as follows:

[0187] It can also be written as:

[0188]

[0189] The pixel coordinates of the upper side of the land survey slope classification map where the current pixel search box is located are determined as follows:

[0190] It can also be written as:

[0191]

[0192] The pixel coordinates of the right side of the land survey slope classification map where the current pixel search box is located are determined as follows:

[0193] It can also be written as:

[0194]

[0195] The pixel coordinates of the lower side of the land survey slope classification map where the current pixel search box is located are determined as follows:

[0196]

[0197] It can also be written as:

[0198]

[0199] S123, obtaining the slope classification value of each pixel coordinate under the current pixel search box; according to all the obtained slope classification values, the number of slope classification values ​​at the left position is counted, and the maximum number is recorded as A Left , A Left It indicates the maximum value after counting the number of values ​​of the slope classification values ​​at the left position;

[0200] The number of values ​​of the upper slope classification value is counted, and the maximum number is recorded as A Up , A Up It indicates the maximum value after counting the number of values ​​of the slope classification values ​​at the upper side;

[0201] Count the number of slope classification values ​​at the right position, and record the maximum number as A Right , A RightIt indicates the maximum value after counting the number of values ​​of the slope classification values ​​at the right position;

[0202] Count the number of slope classification values ​​at the left position, and record the maximum number as A. Down , A Down It indicates the maximum value after counting the number of values ​​of the slope classification value at the lower side;

[0203] S124, A Left , A Up , A Right , A Down The four values ​​are arranged in order from largest to smallest: A max , A max-1 , A max-2 , A max-3 ;

[0204] Among them, A max Indicates that A Left , A Up , A Right , A Down The first value after these four values ​​are arranged in order from largest to smallest;

[0205] A max-1 Indicates that A Left , A Up , A Right , A Down The second value after these four values ​​are arranged in order from largest to smallest;

[0206] A max-2 Indicates that A Left , A Up , A Right , A Down The third value after these four values ​​are arranged in order from largest to smallest;

[0207] A max-3 Indicates that A Left , A Up , A Right , A Down The fourth value after these four values ​​are arranged in order from largest to smallest;

[0208] If A max >A max-1 , then the slope classification value of the pixel coordinates of the center point of the location of the land survey slope classification map where the current pixel search box is located is replaced with the slope classification value corresponding to the value in the first place in the statistics;

[0209] If A max =Amax-1 , then proceed to the next step;

[0210] S125, Get A max The corresponding position of the slope classification value and A max-1 The position corresponding to the slope classification value is A max The corresponding position of the slope classification value and A max-1 The positions corresponding to the corresponding slope classification values ​​constitute a position set;

[0211] If there is a left position in the position set, the slope classification value of the center pixel coordinates of the position of the land survey slope classification map where the current pixel search box is located is replaced with the slope classification value with the largest number of values ​​of the slope classification values ​​of the left position;

[0212] If there is no left position in the position set, but there is an upper position, then the slope classification value of the pixel point coordinates of the center point of the position of the land survey slope classification map where the current pixel search box is located is replaced with the slope classification value with the largest number of values ​​of the slope classification value of the upper position;

[0213] If there are right and lower positions in the position set, the slope classification value of the pixel point coordinates of the center point of the position of the land survey slope classification map where the current pixel search box is located is replaced with the slope classification value with the largest number of slope classification values ​​of the right position;

[0214] S126, the pixel search box executes the next position until the entire land survey slope classification map is traversed.

[0215] In a preferred embodiment of the present invention, in step S121, P is an odd number greater than or equal to 3.

[0216] In a preferred implementation of the present invention, in step S121, the pixel search box size is 5×5.

[0217] In a preferred embodiment of the present invention, the relationship between P and p in step S122 is expressed as:

[0218]

[0219] Wherein, P represents the number of horizontal / vertical pixels in the pixel search box;

[0220] Indicates the floor symbol;

[0221] p represents the number of pixels from the center of the pixel search box to the left bounding box / upper bounding box / right bounding box / lower bounding box.

[0222] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for rapid sampling, verification and analysis of land resource land classification, characterized in that: The following steps are involved: S1, obtain a vertical bird's-eye view of land resources and land types taken by an aerial camera; S2, dividing the land resource classification vertical bird's-eye view map in step S1 into land classification contours; S3, marking the land and resources categories on the vertical bird's-eye view of the land and resources categories after the land category outlines are divided in step S2; S4, classifying the vertical bird's-eye view of the land resources categories marked in step S3 for rapid sampling and verification.

2. The method for rapid sampling, verification and analysis of land resource land classification according to claim 1 is characterized in that: Step S1 is: Obtain M vertical bird's-eye views of land resources taken by an aerial camera, where M is a positive integer greater than or equal to 2; Splice M vertical bird's-eye view maps of land and resources classification into one vertical bird's-eye view map of land and resources classification.

3. The method for rapid sampling, verification and analysis of land resource land classification patterns according to claim 1 is characterized in that: In step S3, the land resource land categories include one or any combination of reservoir water surface land categories, dry land land categories, paddy field land categories, tree forest land categories, shrub forest land categories, hydraulic construction land categories, rural residential land categories, pond water surface land categories, bamboo forest land categories, and rural road land categories.

4. The method for rapid sampling, verification and analysis of land resource land classification patterns according to claim 1 is characterized in that: In step S1, the area photographed by the drone is an area based on groups / communities, villages / communities, towns / townships / streets, and districts / counties.

5. The method for rapid sampling, verification and analysis of land resource land classification according to claim 1 is characterized in that: In step S2, the land type contour division is performed on the vertical bird's-eye view of the land resources land type by using drawing software.

6. The method for rapid sampling, verification and analysis of land resource land classification according to claim 1 is characterized in that: The method for classifying the marked vertical bird's-eye view of land and resources in step S4 includes the following steps: S41, count the number of land resources categories, recorded as N, which are the first land resources category, the second land resources category, the third land resources category, ..., the Nth land resources category; The number of contour areas after the land category contour division of the vertical bird's-eye view of the land resources land category is counted, denoted as K, which are the first contour land category area, the second contour land category area, the third contour land category area, ..., the Kth contour land category area, and the area of ​​each contour land category area is S1, S2, S3, ..., S K ; S1 is the area of ​​the first contour land class area, S2 is the area of ​​the second contour land class area, S3 is the area of ​​the third contour land class area, S K is the area of ​​the Kth contour land category area; Arrange the area of ​​each contour land type in order from small to large as follows: S′ min 、S′ min-1 、S′ min-2 、S′ min-K+1 , Among them, S′ min It means the area of ​​each contour land type area is arranged in order from small to large, and the area at the first place; S′ min-1 It means the area of ​​the second place after arranging the areas of each contour land type in order from small to large; S′ min-2 It means the area at the third position after arranging the areas of each contour land type in order from small to large; S′ min-K+1 It means the area at the Kth position after arranging the areas of each contour land type in order from small to large; S42, setting the size of the sampling frame to φ×φ and the number of samplings of each land type contour division area and N sampling maps; N is the number of land resource land types; The calculation method for the number of random inspections is: Among them, O k Indicates the number of random inspections of the kth contour land type area; S k is the area of ​​the kth contour land class area; S′ min It means the area of ​​each contour land type area is arranged in order from small to large, and the area at the first place; represents the ceiling function; <> means taking the decimal part; Z represents an integer; The size of each sampling image is They represent the number of sizes of sampling frames that can be placed in the sampling image horizontally and vertically respectively; from left to right and from top to bottom, they are the 11th pure white image, the 12th pure white image, the 13th pure white image, ..., the 1φ pure white image, which is the 1st row of pure white images; The 21st pure white image, the 22nd pure white image, the 23rd pure white image, ..., the 2φ pure white image, which is the second row of pure white images; The 31st pure white image, the 32nd pure white image, the 33rd pure white image, ..., the 3rdφ pure white image, which is the 3rd row of pure white images; ……； φ1th pure white image, φ2th pure white image, φ3th pure white image, ..., φφth pure white image, which are the φth row of pure white images; S43, dividing the vertical bird's-eye view map of land and resources land types after the land type contour division into rows and columns of a sampling frame, wherein the number of pixels between row lines is equal to the number of vertical pixels of the sampling frame, and the number of pixels between column lines is equal to the number of horizontal pixels of the sampling frame. If the number of row pixels or column pixels of the last row or the last column or the first row or the first column is less than the number of horizontal pixels of the sampling frame or the number of vertical pixels of the sampling frame, then the row or the column may be retained or discarded; When retaining, corresponding row pixels or column pixels may be added before the corresponding first row or first column or after the last row or last column, so that the number of row pixels or column pixels of the last row or last column or the first row or first column is equal to the number of horizontal pixels of the sampling frame or the number of vertical pixels of the sampling frame; the pixel value of the added row pixels or column pixels is pure white; S44, select the contour area of ​​the vertical bird's-eye view of the land resources land category after the land category contour is divided according to the set order, randomly select one or more sampling frames in the selected contour area, if the sampling frames do not completely belong to the selected contour area, then translate the selected sampling frame up and down or left and right or reduce the size of the sampling frame to the size of the selected sampling frame with the center of the selected sampling frame as the center. To reduce the number of pixels, is an even number greater than 1, so that the selected inspection box is completely within the selected contour area; S45, overlaying the image in the selected sampling frame onto the sampling map of the corresponding land category, and the center of the image in the selected sampling frame coincides with the center of the overlaid sampling frame, so that the image in the selected sampling frame is completely within the sampling frame in the sampling map; S46, executing steps S44 to S45 until all contour areas in the vertical bird's-eye view of the land resources land categories after the land category contours are divided are selected.

7. The method for rapid sampling, verification and analysis of land resource land classification according to claim 1 is characterized in that: After step S4, step S5 is also included. When checking the sampling map, if a sampling box in the sampling map is selected and checked, it jumps to the location of the vertical bird's-eye view of the land and resources category marked corresponding to the image in the sampling box.

8. The method for rapid sampling, verification and analysis of land resource land classification according to claim 1 is characterized in that: Step S5 includes the following steps: S51, determining whether a trigger signal of a sampling box in the sampling map is received: If a trigger signal of a sampling box in the selected sampling map is received, the next step is executed; If no trigger signal for selecting a sampling frame in the sampling map is received, step S51 is executed; S52, determining whether a trigger signal for viewing a certain sampling box in the selected sampling map is received: If a trigger signal for viewing a certain sampling box in the selected sampling map is received, the next step is executed; If no trigger signal for viewing a certain sampling frame in the selected sampling map is received, step S52 is executed; S53, obtaining the link pixel coordinates of the image in a certain sampling frame in the selected sampling image, and determining whether the link pixel coordinates of the image in a certain sampling frame in the selected sampling image are obtained: If the coordinates of the linked pixel points of the image in a certain sampling frame in the selected sampling image are obtained, step S54 is executed; If the link pixel coordinates of the image in a certain sampling frame in the selected sampling image are not obtained, step S53 is executed; S54, searching, according to the link pixel point coordinates, for an image position corresponding to the vertical bird's-eye view of the land resources category where the link pixel point coordinates are located.

9. A computer system, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to implement the method for rapid sampling, verification and analysis of land resources land classification map classification as described in one of claims 1 to 8 when executing the executable instructions.

10. A computer-readable storage medium, characterized in that: include: a memory having a computer program stored thereon; A processor is used to execute the program in the memory to implement the method for rapid sampling, verification and analysis of land resources land classification map classification as described in any one of claims 1 to 8.

Images