CN117570881B - Land area measurement system and method for investigation of homeland resources - Google Patents
Land area measurement system and method for investigation of homeland resources Download PDFInfo
- Publication number
- CN117570881B CN117570881B CN202410064807.5A CN202410064807A CN117570881B CN 117570881 B CN117570881 B CN 117570881B CN 202410064807 A CN202410064807 A CN 202410064807A CN 117570881 B CN117570881 B CN 117570881B
- Authority
- CN
- China
- Prior art keywords
- boundary
- image
- land
- points
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005259 measurement Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000011835 investigation Methods 0.000 title claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000004364 calculation method Methods 0.000 claims abstract description 20
- 238000010276 construction Methods 0.000 claims abstract description 20
- 238000000691 measurement method Methods 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 10
- 239000013598 vector Substances 0.000 claims description 50
- 239000003550 marker Substances 0.000 claims description 14
- 229910052739 hydrogen Inorganic materials 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 238000013519 translation Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 3
- 239000002689 soil Substances 0.000 abstract description 5
- 238000010801 machine learning Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 9
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005314 correlation function Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000011524 similarity measure Methods 0.000 description 2
- 238000012271 agricultural production Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/28—Measuring arrangements characterised by the use of optical techniques for measuring areas
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The application belongs to the technical field of land area measurement, and discloses a land area measurement system and a land area measurement method for investigation of homeland resources, wherein the measurement method comprises the following steps: the route planning module plans the route of the unmanned aerial vehicle aerial survey; the data acquisition module acquires a high-definition image through aerial photographing; the image matching module is used for realizing image matching through extraction of image feature points; the pixel points reflecting the same space point are corresponding; the contour construction module constructs a contour image of the measured soil closure; the image processing module is used for carrying out gray level processing on the land contour image obtained by the contour construction module; processing the obtained gray level image by adopting a threshold method, and extracting a boundary of land measurement by extracting a land boundary target from the background; and the intra-boundary area calculation module adopts boundary tracking to extract the outline of the image area to obtain the area of the target area. The application can quickly measure the land area by combining aerial photography with a machine learning algorithm.
Description
Technical Field
The application relates to the technical field of land area measurement, in particular to a land area measurement system and a land area measurement method for investigation of national land resources.
Background
The investigation of land resources is a series of processes for checking the type quantity, quality, space-time distribution and the interrelation and development change rule of the land resources based on the discipline knowledge of land resource science and based on remote sensing, mapping and drawing. Including comprehensive agricultural regions, land resource evaluation, formulation of national economy development planning, scientific management of land resources, and the like.
The land resource investigation aims to provide basic data for land resource management, basic drawing pieces and attribute data for land evaluation and land utilization planning, is an implementation process of dynamic land resource monitoring, and is an important basis for making national economy plans, comprehensive agricultural regions and agricultural production plans.
The prior art publication No. CN219607936U provides a land measuring device for territory resource mapping, including first measuring stick and second measuring stick, first measuring stick lower extreme is equipped with first locating cone, second measuring stick lower extreme is equipped with the second locating cone, second measuring stick upper portion is rotated and is provided with the wire reel, it has the dipperstick to coil on the wire reel, the dipperstick free end is connected with first measuring stick, the wire reel top is equipped with the locking dish, the wire reel up end is equipped with the locking lever, be equipped with first cramp, second cramp and nut on the locking lever, first cramp is fixed to be set up on the locking lever, with locking dish lower terminal surface sliding fit, the second cramp cover is located on the locking lever, be located the top of locking dish, nut threaded connection is on the locking lever for compress tightly second cramp on the locking dish. Through the cooperation of locking dish, locking pole, first clamp piece, second clamp piece and nut, can realize the locking of wire reel to improve measuring precision, read measured data.
The above prior art solution, though realizing the beneficial effects related to the prior art by the structure of the prior art, still has the following drawbacks: when the land is measured, the land is measured manually, so that time and labor are wasted, and the efficiency is low; and is especially not suitable for large land area measurements.
In view of this, we propose a land area measurement system and measurement method for investigation of homeland resources.
Disclosure of Invention
1. The technical problem to be solved.
The application aims to provide a land area measurement system and a land area measurement method for investigation of domestic land resources, which solve the technical problems in the background art and realize the technical effect of quickly measuring the land area.
2. The technical proposal is that.
The technical scheme of the application provides a land area measurement system for investigation of homeland resources, which comprises the following steps of.
The route planning module is used for planning a route of unmanned aerial vehicle aerial survey according to the terrain and the outline of the land to be measured, planning a route along the land boundary, and the planned aerial survey route is a closed route; and placing a calibrated reference object with specific measured size in the land measurement area as a reference for measuring the size of the land. The calibration reference object preferably adopts a square object as a reference object, so that the area of the calibration reference object is convenient to calculate in the later period; according to the requirement, a plurality of boundary points can be arranged on the boundary of the land to be measured, a marker post is arranged on the boundary points, the marker post is used as the boundary point for the land area aerial survey measurement, and the color of the marker post is required to be easily distinguished.
The data acquisition module comprises an unmanned aerial vehicle, a high-definition camera with a laser range finder, a total station and a height measuring instrument; and carrying out aerial photographing on the land boundary according to the planned route by using a high-definition camera with a laser range finder to obtain a high-definition image.
The image matching module is used for matching the images shot by the data acquisition module through the extraction of the image feature points; extracting feature points through Harris operators; the matching of the feature points is to find the pixel points projected by the same three points in the space on different images, and the pixel points reflecting the same space point are corresponding.
The contour construction module adopts a linear trigonometry to realize three-dimensional reconstruction of key points, adopts a rotary workbench method to splice images, and splices three-dimensional space point sets reconstructed by every two images together to splice the images together; and constructing a profile image of the measured soil closure.
The image processing module is used for carrying out gray level processing on the land contour image obtained by the contour construction module; the gray level image obtained is processed by a threshold method, and the boundary of land measurement is extracted by extracting a land boundary target from the background (the process can be assisted by adding a manual selection boundary).
And the area calculation module of the area in the boundary adopts boundary tracking to extract the outline of the image area, and simultaneously records the marks and the searching directions of all boundary points to obtain the area of the target area.
The PLC control module comprises a display screen and is connected with the data acquisition module, the image matching module and the contour construction module in a network mode. The unmanned aerial vehicle aerial image can be checked through the display screen, and the flight route of the unmanned aerial vehicle can be controlled according to actual requirements.
As an alternative of the present invention, the image matching module performs image matching on the image captured by the data acquisition module, including the following steps.
Firstly, corner point extraction of an image is completed through a Harris operator, then initial matching is estimated through a regional gray level correlation algorithm, a cross correlation function is used as similarity measure between two search neighborhoods, and a formula of a characteristic point matching algorithm is as follows.
Aij=（1/N）Σk,l∈Ω[h1(u-k,v-l)-`h1][h2(u'-k,v'-l)-`h2].
Wherein: omega is a neighborhood (which can be 3X3 or 5X5 neighborhood) with u i points as the center, and N is the number of pixel points in the neighborhood; h 1、`h2 is the gray average value of points in the neighborhood of the u i point and the u ' j point; u i=（u,v）T is a point in the first image, its gray value is h 1(u,v),u' j=(U',v')T is any point in the second image, its gray value is h 2(u',v'), u is an abscissa of a point in the first image, and v is an ordinate of a point in the first image; a ij is the gray level similarity of u i and u ' j; t is a translation vector of the high-definition camera; k is the camera internal parameter and l is the gray image template size. The larger the correlation value A ij, the more similar the points u i and u ' j are.
A region correlation coefficient (LACC) is then used to determine whether the two points match.
(Pij)2=(Aij)2/(A1 2A2 2)。
Wherein A i is the standard deviation in the image window, and A1 is the standard deviation in the first image window; a2 Is the standard deviation within the second image window.
(Ai）2=Σk,l∈Ω[hi(u-k,v-l)2]/N-`h1 2。
The LACC transformation ranges from-1 to 1, which indicates that the similarity is from minimum to maximum, and in practical application, the result of the calculation is compared with a certain threshold P ij set in advance.
If the correlation coefficient is larger than the threshold value P ij, the feature point pairs corresponding to the correlation coefficients are reserved to be candidate matching, and if the correlation coefficient is smaller than the threshold value P ij, the feature point pairs are set to 0.
And then matching the image characteristic points by using a robustness algorithm of characteristic point matching, and eliminating mismatching.
As an alternative of the present invention, the processing of the image by the image processing module includes the following steps.
A. Dividing the obtained gray image into L-level gray, counting the gray value of the pixel, and taking the intermediate value as an initial threshold value T 0; and carrying out gray scale treatment on the areas contained in the closed contour image.
B. The image is analyzed into two regions with a selected initial threshold T 0, a pixel component region G1 with all gray values greater than T 0, and a pixel component region G2 with all gray values less than T 0.
C. The gray average values μ 1 and μ 2 in the areas G1 and G2 are calculated.
μ1=(n1+2n2+3n3……+ini)/(n1+n2+n3……+ni);0≤i≤Tj.
μ2=(n1+2n2+3n3……+ini)/(n1+n2+n3……+ni);Tj≤i≤L-1.
Wherein n i is a gray value, i is the number of pixels with the gray value of i, j is the iteration number, tj is a threshold value after the jth iteration, and tj=t 0 is calculated initially; l is the number of gray levels.
D. a new threshold is calculated.
Tj+1=1/2(μ1+μ2)。
E. Repeating steps b-d until the difference between T j+1 and T j is less than the given value. Since the gray values of the image are integers, the difference between the two gray values is also an integer, and we choose the given value to be 1.
The target boundary region and the background may be separated.
F. And judging the areas contained in the target boundary area as the land parts to be measured.
As an alternative of the present invention, the specific area calculation process of the intra-boundary area calculation module is as follows.
(1) Calculating the total number of boundary pixels: and carrying out eight-neighborhood boundary tracking on the target area according to the anticlockwise direction to obtain a group of ordered boundary points, wherein the labels of the boundary points are from small to large in the anticlockwise direction, and the maximum label is the total number of boundary pixels of the image.
(2) The method comprises the steps of recording vector information corresponding to boundary points, namely, according to the sequence number of the boundary point, a path from a previous boundary point (P-1) to a current boundary point (P) is called a forward vector (pv) of the current boundary point, a path from the current boundary point (P) to a next boundary point (P+1) is called a backward vector (nv) of the current boundary point, and vector directions and vector values adopt eight-direction codes to store the vector information and the boundary points correspondingly.
(3) And re-ordering, namely re-ordering the boundary points obtained by boundary tracking in a top-down and left-right order, wherein the relevant vector corresponding to each boundary point is arranged along with the boundary points.
(4) And calculating the total number of pixels in the boundary, namely sequencing from small to large according to boundary points, and sequentially judging the forward vector and the backward vector of each boundary point. If the current boundary point meets the condition that the pixel on the right side is the boundary inner point, the number of pixels between the current boundary point and the next adjacent boundary point is calculated, namely X i+1—Xi -1. Where X i is the column value of the current boundary point and X i+1 is the column value of the next boundary point adjacent to the current boundary point. The pixel to the right of the current boundary point is the in-boundary point.
Nv+.8, with pv=5, or pv <3 and |pv-nv| >4, or pv >5 and |pv-nv| <4.
(5) In-region pixel point calculation, since the boundary pixel point is generally half inside the image and half outside the image, the area measurement is: g Zone(s) = total number of boundary pixels/total number of pixels within 2+ boundary.
Where G Zone(s) is the total number of pixels within the boundary.
(6) Counting the number of pixels occupied by the calibration reference object; and calculating pixel equivalent=DXH/M according to the size of the calibration reference object and the number of pixel points occupied by the calibration reference object.
Wherein D, H is the length and width dimensions of the calibration reference; m is the number of pixels occupied by the reference object in the image.
(7) And calculating the area of the land.
S=g Zone(s) X pixel equivalents, and the area of the land is calculated.
The invention provides a land area measurement method for investigation of homeland resources, which comprises the following steps.
S1, a route planning module plans a route of unmanned aerial vehicle aerial survey according to the terrain and the outline of the land to be measured, and plans the route along the land boundary, wherein the planned aerial survey route is a closed route; placing a calibrated reference object with a specific measured size in a land measurement area as a reference for measuring the size of the land; setting a plurality of boundary points on the boundary of the land to be measured, and setting a marker post on the boundary points to serve as boundary points for land area aerial survey measurement.
S2, the PLC control module controls the data acquisition module to carry out aerial photographing on the land boundary through the high-definition camera with the laser range finder according to the planned route so as to acquire a high-definition image.
S3, the image matching module is used for matching the images shot by the data acquisition module through extraction of the image feature points; extracting feature points through Harris operators; the matching of the feature points is to find the pixel points projected by the same three points in the space on different images, and the pixel points reflecting the same space point are corresponding; the method specifically comprises the following steps.
S31, extracting corner points of the image through a Harris operator, and estimating initial matching by using a region gray correlation algorithm, wherein a formula of a characteristic point matching algorithm is as follows.
Aij=（1/N）Σk,l∈Ω[h1(u-k,v-l)-`h1][h2(u'-k,v'-l)-`h2].
Wherein: omega is a neighborhood (which can be 3X3 or 5X5 neighborhood) with u i points as the center, and N is the number of pixel points in the neighborhood; h 1、`h2 is the gray average value of points in the neighborhood of the u i point and the u ' j point; u i=（u,v）T is a point in the first image, its gray value is h 1 (u, v), u is an abscissa of a point in the first image, and v is an ordinate of a point in the first image; u ' j=(u',v')T is any point in the second image, the gray value of which is h 2(u',v'),Aij is the gray similarity between u i and u ' j points; t is a translation vector of the high-definition camera; k is the camera internal parameter and l is the gray image template size. The larger the correlation value A ij, the more similar the points u i and u ' j are.
S32, then use the region correlation coefficient (LACC) to determine if the two points match.
(Pij)2=(Aij)2/(A1 2A2 2)。
Wherein A i is the standard deviation within the image window: a1 Is the standard deviation within the first image window; a2 Is the standard deviation within the second image window.
(Ai）2=Σk,l∈Ω[hi(u-k,v-l)2]/N-`h1 2。
The LACC transformation ranges from-1 to 1, which indicates that the similarity is from minimum to maximum, and in practical application, the result of the calculation is compared with a certain threshold P ij set in advance.
If the correlation coefficient is larger than the threshold value P ij, the feature point pairs corresponding to the correlation coefficients are reserved to be candidate matching, and if the correlation coefficient is smaller than the threshold value P ij, the feature point pairs are set to 0.
And S33, matching the image characteristic points by using a robustness algorithm of characteristic point matching, and eliminating mismatching.
S4, the contour construction module adopts a linear trigonometry to realize three-dimensional reconstruction of key points, adopts a rotary workbench method to splice images, and splices three-dimensional space point sets reconstructed by every two images together to splice the images together; and constructing a profile image of the measured soil closure.
S5, the image processing module carries out gray scale processing on the land contour image obtained by the contour construction module; the gray level image obtained is processed by a threshold method, and the boundary of land measurement is extracted by extracting a land boundary target from the background (the process can be assisted by adding a manual selection boundary).
S6, an area calculation module in the boundary adopts boundary tracking to extract the outline of the image area, and simultaneously records the marks and the searching directions of all boundary points to obtain the area of the target area; the method specifically comprises the following steps.
S61, calculating the total number of boundary pixels: and carrying out eight-neighborhood boundary tracking on the target area according to the anticlockwise direction to obtain a group of ordered boundary points, wherein the labels of the boundary points are from small to large in the anticlockwise direction, and the maximum label is the total number of boundary pixels of the image.
And S62, recording vector information corresponding to the boundary point, namely, according to the sequence number of the boundary point, a path from the previous boundary point (P-1) to the current boundary point (P) is called a forward vector (pv) of the current boundary point, a path from the current boundary point (P) to the next boundary point (P+1) is called a backward vector (nv) of the current boundary point, and the vector direction and the vector value adopt eight-direction codes to store the vector information corresponding to the boundary point.
S63, reordering, namely reordering the boundary points obtained by boundary tracking in a sequence from top to bottom and from left to right, wherein the relevant vector corresponding to each boundary point is arranged along with the boundary point.
S64, calculating the total number of pixels in the boundary, namely sequencing from small to large according to boundary points, and judging the forward vector and the backward vector of each boundary point in sequence. If the current boundary point meets the condition that the pixel on the right side is the boundary inner point, the number of pixels between the current boundary point and the next adjacent boundary point is calculated, namely X i+1—Xi -1. Where X i is the column value of the current boundary point and X i+1 is the column value of the next boundary point adjacent to the current boundary point. The pixel to the right of the current boundary point is the in-boundary point.
Nv+.8, with pv=5, or pv <3 and |pv-nv| >4, or pv >5 and |pv-nv| <4.
S65, calculating pixel points in the area, wherein the boundary pixel points are half of the inside of the image and half of the outside of the image, so that the area is measured as G Zone(s) =total number of boundary pixels/total number of 2+ pixels in the boundary.
Where G Zone(s) is the total number of pixels within the boundary.
S66, counting the number of pixels occupied by the calibration reference object; and calculating pixel equivalent=DXH/M according to the size of the calibration reference object and the number of pixel points occupied by the calibration reference object.
Wherein D, H is the length and width dimensions of the calibration reference; m is the number of pixels occupied by the reference object in the image.
S67, calculating the area of the land.
S=g Zone(s) ×pixel equivalent, and the area of land is calculated.
3. Has the beneficial effects of.
One or more of the technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages.
1. The invention can rapidly take photo of large land and measure land area, thereby saving manpower and material resources and improving working efficiency.
2. Through boundary route planning and unmanned aerial vehicle aerial photography, a boundary image of the land to be measured can be quickly constructed.
3. The boundary profile of the land to be measured can be obtained rapidly through a machine learning algorithm, so that the land area can be measured conveniently.
Drawings
Fig. 1 is a schematic diagram of a land area measurement system for investigation of homeland resources according to a preferred embodiment of the present application.
FIG. 2 is a flow chart of area calculation performed by the intra-boundary area calculation module of the land area measurement system for investigation of land resources according to a preferred embodiment of the present application.
Detailed Description
The application is described in further detail below with reference to the drawings.
Referring to fig. 1 to 2, the present invention provides a land area measurement system for investigation of homeland resources, comprising.
The route planning module is used for planning a route of unmanned aerial vehicle aerial survey according to the terrain and the outline of the land to be measured, planning a route along the land boundary, and the planned aerial survey route is a closed route; and placing a calibrated reference object with specific measured size in the land measurement area as a reference for measuring the size of the land. The calibration reference object preferably adopts a square object as the reference object, so that the area of the calibration reference object can be conveniently calculated in the later period.
According to the requirement, a plurality of boundary points can be arranged on the boundary of the land to be measured, a marker post is arranged on the boundary points, the marker post is used as the boundary point for the land area aerial survey measurement, and the color of the marker post is required to be easily distinguished.
The data acquisition module comprises an unmanned aerial vehicle, a high-definition camera with a laser range finder, a total station and a height measuring instrument; and carrying out aerial photographing on the land boundary according to the planned route by using a high-definition camera with a laser range finder to obtain a high-definition image.
The image matching module is used for matching the images shot by the data acquisition module through the extraction of the image feature points; extracting feature points through Harris operators; the matching of the feature points is to find the pixel points projected by the same three points in the space on different images, and the pixel points reflecting the same space point are corresponding.
The contour construction module adopts a linear trigonometry to realize three-dimensional reconstruction of key points, adopts a rotary workbench method to splice images, and splices three-dimensional space point sets reconstructed by every two images together to splice the images together; and constructing a profile image of the measured soil closure.
The image processing module is used for carrying out gray level processing on the land contour image obtained by the contour construction module; the gray level image obtained is processed by a threshold method, and the boundary of land measurement is extracted by extracting a land boundary target from the background (the process can be assisted by adding a manual selection boundary).
And the area calculation module of the area in the boundary adopts boundary tracking to extract the outline of the image area, and simultaneously records the marks and the searching directions of all boundary points to obtain the area of the target area.
The PLC control module comprises a display screen and is connected with the data acquisition module, the image matching module and the contour construction module in a network mode. The unmanned aerial vehicle aerial image can be checked through the display screen, and the flight route of the unmanned aerial vehicle can be controlled according to actual requirements.
Further, the image matching module performs image matching on the image captured by the data acquisition module, and the image matching module comprises the following steps.
Firstly, corner point extraction of an image is completed through a Harris operator, then initial matching is estimated through a regional gray level correlation algorithm, a cross correlation function is used as similarity measure between two search neighborhoods, and a formula of a characteristic point matching algorithm is as follows.
Aij=（1/N）Σk,l∈Ω[h1(u-k,v-l)-`h1][h2(u'-k,v'-l)-`h2].
Wherein: omega is a neighborhood (which can be 3X3 or 5X5 neighborhood) with u i points as the center, and N is the number of pixel points in the neighborhood; h 1、`h2 is the gray average value of points in the neighborhood of the u i point and the u ' j point; u i=（u,v）T is a point in the first image, its gray value is h 1 (u, v), u is an abscissa of a point in the first image, and v is an ordinate of a point in the first image; u ' j=(u',v')T is any point in the second image, the gray value of which is h 2(u',v'),Aij is the gray similarity between u i and u ' j points; t is a translation vector of the high-definition camera; k is the camera internal parameter and l is the gray image template size. The larger the correlation value A ij, the more similar the points u i and u ' j are.
A region correlation coefficient (LACC) is then used to determine whether the two points match.
(Pij)2=(Aij)2/(A1 2A2 2)。
Wherein A i is the standard deviation in the image window; a1 Is the standard deviation within the first image window; a2 Is the standard deviation within the second image window.
(Ai）2=Σk,l∈Ω[hi(u-k,v-l)2]/N-`h1 2。
The LACC transformation ranges from-1 to 1, which indicates that the similarity is from minimum to maximum, and in practical application, the result of the calculation is compared with a certain threshold P ij set in advance.
If the correlation coefficient is larger than the threshold value P ij, the feature point pairs corresponding to the correlation coefficients are reserved to be candidate matching, and if the correlation coefficient is smaller than the threshold value P ij, the feature point pairs are set to 0.
And then matching the image characteristic points by using a robustness algorithm of characteristic point matching, and eliminating mismatching.
The processing of the image by the image processing module comprises the following steps.
A. Dividing the obtained gray image into L-level gray, counting the gray value of the pixel, and taking the intermediate value as an initial threshold value T 0; and carrying out gray scale treatment on the areas contained in the closed contour image.
B. The image is analyzed into two regions with a selected initial threshold T 0, a pixel component region G1 with all gray values greater than T 0, and a pixel component region G2 with all gray values less than T 0.
C. The gray average values μ 1 and μ 2 in the areas G1 and G2 are calculated.
μ1=(n1+2n2+3n3……+ini)/(n1+n2+n3……+ni);0≤i≤Tj.
μ2=(n1+2n2+3n3……+ini)/(n1+n2+n3……+ni);Tj≤i≤L-1.
Wherein n i is a gray value, i is the number of pixels with the gray value of i, j is the iteration number, tj is a threshold value after the jth iteration, and tj=t 0 is calculated initially; l is the number of gray levels.
D. a new threshold is calculated.
Tj+1=1/2(μ1+μ2)。
E. Repeating steps b-d until the difference between T j+1 and T j is less than the given value. Since the gray values of the image are integers, the difference between the two gray values is also an integer, and we choose the given value to be 1.
The target boundary region and the background may be separated.
F. And judging the areas contained in the target boundary area as the land parts to be measured.
The specific area calculation process of the area calculation module in the boundary is as follows.
(1) Calculating the total number of boundary pixels: and carrying out eight-neighborhood boundary tracking on the target area according to the anticlockwise direction to obtain a group of ordered boundary points, wherein the labels of the boundary points are from small to large in the anticlockwise direction, and the maximum label is the total number of boundary pixels of the image.
(2) The method comprises the steps of recording vector information corresponding to boundary points, namely, according to the sequence number of the boundary point, a path from a previous boundary point (P-1) to a current boundary point (P) is called a forward vector (pv) of the current boundary point, a path from the current boundary point (P) to a next boundary point (P+1) is called a backward vector (nv) of the current boundary point, and vector directions and vector values adopt eight-direction codes to store the vector information and the boundary points correspondingly.
(3) And re-ordering, namely re-ordering the boundary points obtained by boundary tracking in a top-down and left-right order, wherein the relevant vector corresponding to each boundary point is arranged along with the boundary points.
(4) And calculating the total number of pixels in the boundary, namely sequencing from small to large according to boundary points, and sequentially judging the forward vector and the backward vector of each boundary point. If the current boundary point meets the condition that the pixel on the right side is the boundary inner point, the number of pixels between the current boundary point and the next adjacent boundary point is calculated, namely X i+1—Xi -1. Where X i is the column value of the current boundary point and X i+1 is the column value of the next boundary point adjacent to the current boundary point. The pixel to the right of the current boundary point is the in-boundary point.
Nv+.8, with pv=5, or pv <3 and |pv-nv| >4, or pv >5 and |pv-nv| <4.
(5) Since the boundary pixels are typically half inside the image and half outside the image, the area is measured as G Zone(s) =total boundary pixels/total 2+ total pixels within the boundary.
Where G Zone(s) is the total number of pixels within the boundary.
(6) Counting the number of pixels occupied by the calibration reference object; and calculating pixel equivalent=DXH/M according to the size of the calibration reference object and the number of pixel points occupied by the calibration reference object.
Wherein D, H is the length and width dimensions of the calibration reference; m is the number of pixels occupied by the reference object in the image.
(7) And calculating the area of the land.
S=g Zone(s) ×pixel equivalent, and the area of land is calculated.
The invention provides a land area measurement method for investigation of homeland resources, which comprises the following steps.
S1, a route planning module plans a route of unmanned aerial vehicle aerial survey according to the terrain and the outline of the land to be measured, and plans the route along the land boundary, wherein the planned aerial survey route is a closed route; placing a calibrated reference object with a specific measured size in a land measurement area as a reference for measuring the size of the land; a plurality of boundary points are arranged on the boundary of the land to be measured, a marker post is arranged on the boundary points, the marker post is used as the boundary point of the land aerial survey measurement, and the color of the marker post is required to be easily distinguished.
S2, the PLC control module controls the data acquisition module to carry out aerial photographing on the land boundary through the high-definition camera with the laser range finder according to the planned route so as to acquire a high-definition image.
S3, the image matching module is used for matching the images shot by the data acquisition module through extraction of the image feature points; extracting feature points through Harris operators; the matching of the feature points is to find the pixel points projected by the same three points in the space on different images, and the pixel points reflecting the same space point are corresponding; the method specifically comprises the following steps.
S31, extracting corner points of the image through a Harris operator, and estimating initial matching by using a region gray correlation algorithm, wherein a formula of a characteristic point matching algorithm is as follows.
Aij=（1/N）Σk,l∈Ω[h1(u-k,v-l)-`h1][h2(u'-k,v'-l)-`h2].
Wherein: omega is a neighborhood (which can be 3X3 or 5X5 neighborhood) with u i points as the center, and N is the number of pixel points in the neighborhood; h 1、`h2 is the gray average value of points in the neighborhood of the u i point and the u ' j point; u i=（u,v）T is a point in the first image, its gray value is h 1 (u, v), u is an abscissa of a point in the first image, and v is an ordinate of a point in the first image; u ' j=(u',v')T is any point in the second image, the gray value of which is h 2(u',v'),Aij is the gray similarity between u i and u ' j points; t is a translation vector of the high-definition camera; k is a camera internal parameter and l is a gray image template size, e.g., 5X5 or 7X7. The larger the correlation value A ij, the more similar the points u i and u ' j are.
S32, then use the region correlation coefficient (LACC) to determine if the two points match.
(Pij)2=(Aij)2/(A1 2A2 2)。
Wherein A i is the standard deviation in the image window, and A1 is the standard deviation in the first image window; a2 Is the standard deviation within the second image window.
(Ai）2=Σk,l∈Ω[hi(u-k,v-l)2]/N-`h1 2。
The LACC transformation ranges from-1 to 1, which indicates that the similarity is from minimum to maximum, and in practical application, the result of the calculation is compared with a certain threshold P ij set in advance.
If the correlation coefficient is larger than the threshold value P ij, the feature point pairs corresponding to the correlation coefficients are reserved to be candidate matching, and if the correlation coefficient is smaller than the threshold value P ij, the feature point pairs are set to 0.
And S33, matching the image characteristic points by using a robustness algorithm of characteristic point matching, and eliminating mismatching.
S4, the contour construction module adopts a linear trigonometry to realize three-dimensional reconstruction of key points, adopts a rotary workbench method to splice images, and splices three-dimensional space point sets reconstructed by every two images together to splice the images together; and constructing a profile image of the measured soil closure.
S5, the image processing module carries out gray scale processing on the land contour image obtained by the contour construction module; the gray level image obtained is processed by a threshold method, and the boundary of land measurement is extracted by extracting a land boundary target from the background (the process can be assisted by adding a manual selection boundary).
S6, an area calculation module in the boundary adopts boundary tracking to extract the outline of the image area, and simultaneously records the marks and the searching directions of all boundary points to obtain the area of the target area; the method specifically comprises the following steps.
S61, calculating the total number of boundary pixels: and carrying out eight-neighborhood boundary tracking on the target area according to the anticlockwise direction to obtain a group of ordered boundary points, wherein the labels of the boundary points are from small to large in the anticlockwise direction, and the maximum label is the total number of boundary pixels of the image.
And S62, recording vector information corresponding to the boundary point, namely, according to the sequence number of the boundary point, a path from the previous boundary point (P-1) to the current boundary point (P) is called a forward vector (pv) of the current boundary point, a path from the current boundary point (P) to the next boundary point (P+1) is called a backward vector (nv) of the current boundary point, and the vector direction and the vector value adopt eight-direction codes to store the vector information corresponding to the boundary point.
S63, reordering, namely reordering the boundary points obtained by boundary tracking in a sequence from top to bottom and from left to right, wherein the relevant vector corresponding to each boundary point is arranged along with the boundary point.
S64, calculating the total number of pixels in the boundary, namely sequencing from small to large according to boundary points, and judging the forward vector and the backward vector of each boundary point in sequence. If the current boundary point meets the condition that the pixel on the right side is the boundary inner point, the number of pixels between the current boundary point and the next adjacent boundary point is calculated, namely X i+1—Xi -1. Where X i is the column value of the current boundary point and X i+1 is the column value of the next boundary point adjacent to the current boundary point. The pixel to the right of the current boundary point is the in-boundary point.
Nv+.8, with pv=5, or pv <3 and |pv-nv| >4, or pv >5 and |pv-nv| <4.
S65, calculating pixel points in the area, wherein the boundary pixel points are half of the inside of the image and half of the outside of the image, so that the area is measured as G Zone(s) =total number of boundary pixels/total number of 2+ pixels in the boundary.
Where G Zone(s) is the total number of pixels within the boundary.
S66, counting the number of pixels occupied by the calibration reference object; and calculating pixel equivalent=DXH/M according to the size of the calibration reference object and the number of pixel points occupied by the calibration reference object.
Wherein D, H is the length and width dimensions of the calibration reference; m is the number of pixels occupied by the reference object in the image.
S67, calculating the area of the land.
S=g Zone(s) X pixel equivalents, and the area of the land is calculated.
The invention relates to a land area measurement system for investigation of homeland resources, which has the working principle that.
Firstly, a route planning module plans a route of unmanned aerial vehicle aerial survey according to the terrain and the outline of the land to be measured, plans the route along the land boundary, and the planned aerial survey route is a closed route; placing a calibrated reference object with a specific measured size in a land measurement area as a reference for measuring the size of the land; a plurality of boundary points are arranged on the boundary of the land to be measured, a marker post is arranged on the boundary points, the marker post is used as the boundary point of the land aerial survey measurement, and the color of the marker post is required to be easily distinguished.
And the PLC control module is used for controlling the data acquisition module to aerial photograph the land boundary through the high-definition camera with the laser range finder according to the planned route so as to acquire a high-definition image.
Then the image matching module is used for matching the images shot by the data acquisition module through the extraction of the image characteristic points; extracting feature points through Harris operators; the matching of the feature points is to find the pixel points projected by the same three points in the space on different images, and the pixel points reflecting the same space point are corresponding; the contour construction module adopts a linear trigonometry to realize three-dimensional reconstruction of key points, adopts a rotary workbench method to splice images, and splices three-dimensional space point sets reconstructed by every two images together so as to splice the images together; constructing a profile image of the measured land closure; then the image processing module carries out gray scale processing on the land contour image obtained by the contour construction module; processing the obtained gray level image by a threshold method, and extracting a boundary of land measurement by extracting a land boundary target from the background (the process can be assisted by adding a manual selection boundary); and extracting the outline of the image region by adopting boundary tracking through a region area calculation module in the boundary, and simultaneously recording the marks and the searching directions of all boundary points to obtain the area of the target region.
The invention can rapidly take photo of large land and measure land area, thereby improving working efficiency; through boundary route planning and unmanned aerial vehicle aerial photography, a boundary image of the land to be measured can be quickly constructed; the boundary profile of the land to be measured can be obtained rapidly through a machine learning algorithm, so that the land area can be measured conveniently.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (9)
1. The land area measurement method for investigation of the homeland resources is characterized by comprising the following steps of:
s1, a route planning module plans a route of unmanned aerial vehicle aerial survey according to the terrain and the outline of the land to be measured;
S2, the PLC control module controls the data acquisition module to carry out aerial photographing on the land boundary according to the planned route to acquire a high-definition image;
S3, the image matching module is used for matching the images shot by the data acquisition module through extraction of the image feature points; extracting feature points through Harris operators; the pixel points reflecting the same space point are corresponding;
S4, the contour construction module adopts a linear trigonometry to realize three-dimensional reconstruction of key points, and adopts a rotary workbench method to splice images so as to splice the images; constructing a profile image of the measured land closure;
S5, the image processing module carries out gray scale processing on the land contour image obtained by the contour construction module; processing the obtained gray level image by adopting a threshold method, and extracting a boundary of land measurement by extracting a land boundary target from the background;
s6, an area calculation module in the boundary adopts boundary tracking to extract the outline of the image area, and simultaneously records the labels and the searching directions of all boundary points to obtain the area of the target area, and the method comprises the following steps:
s61, calculating the total number of boundary pixels: performing eight-neighborhood boundary tracking on the target area according to the anticlockwise direction to obtain a group of ordered boundary points, wherein the labels of the boundary points are from small to large in the anticlockwise direction, and the maximum label is the total number of boundary pixels of the image;
S62, recording vector information corresponding to boundary points, namely, according to the sequence number of the boundary points, a path from a previous boundary point (P-1) to a current boundary point (P) is called as a forward vector (pv) of the current boundary point, a path from the current boundary point (P) to a next boundary point (P+1) is called as a backward vector (nv) of the current boundary point, and vector directions and vector values adopt eight-direction codes, so that the vector information and the boundary points are correspondingly stored;
S63, reordering the boundary points obtained by boundary tracking according to the sequence from top to bottom and from left to right, wherein the relevant vector corresponding to each boundary point is arranged along with the boundary point;
S64, calculating the total number of pixels in the boundary, namely sequencing from small to large according to boundary points, and judging forward vectors and backward vectors of each boundary point in sequence; if the current boundary point meets the condition that the pixel on the right side is the boundary inner point, calculating the number of pixels between the current boundary point and the adjacent next boundary point, namely X i+1—Xi -1; wherein, X i is the column value of the current boundary point, and X i+1 is the column value of the next boundary point adjacent to the current boundary point; the condition that the pixel on the right side of the current boundary point is the in-boundary point is:
nv+.8, with pv=5, or pv <3 and |pv-nv| >4, or pv >5 and |pv-nv| <4;
S65, calculating pixel points in the area, wherein the boundary pixel points are half of the boundary pixel points in the image and half of the boundary pixel points are outside the image, so that the measurement value of the area is G Zone(s) = total boundary pixels/total 2+ total pixels in the boundary;
Wherein G Zone(s) is the total number of pixel points in the boundary;
s66, counting the number of pixels occupied by the calibration reference object; calculating pixel equivalent = D x H/M according to the size of the calibration reference and the number of pixel points occupied by the calibration reference;
Wherein D, H is the length and width dimensions of the calibration reference; m is the number of pixels occupied by the marked reference object in the image;
S67, calculating the area of the land:
s=g Zone(s) ×pixel equivalent, and the area of land is calculated.
2. The land area measurement method for investigation of homeland resources according to claim 1, wherein: in step S1, a calibration reference object with specific measured size is placed in a land measurement area; a plurality of boundary points are arranged on the boundary of the land, and a marker post is arranged on the boundary points and used as the boundary points for the land area aerial survey.
3. The land area measurement method for investigation of homeland resources according to claim 1, wherein: step S3 comprises the steps of:
s31, extracting corner points of an image through a Harris operator, and estimating initial matching by using a region gray correlation algorithm, wherein the formula of a characteristic point matching algorithm is as follows:
Aij=（1/N）Σk,l∈Ω[h1(u-k,v-l)-`h1][h2(u'-k,v'-l)-`h2]
Wherein: omega is a neighborhood with u i points as the center, and N is the number of pixel points in the neighborhood; h 1、`h2 is the gray average value of points in the neighborhood of the u i point and the u ' j point; u i=（u,v）T is a point in the first image, its gray value is h 1 (u, v), u is an abscissa of a point in the first image, and v is an ordinate of a point in the first image; u ' j=(u',v')T is any point in the second image, the gray value of which is h 2(u',v'),Aij is the gray similarity between u i and u ' j points; t is a translation vector of the high-definition camera; k is the internal parameters of the camera, and l is the size of the gray image template;
s32, judging whether two points match or not by using the region correlation coefficient:
(Pij)2=(Aij)2/(A1 2A2 2)；
Wherein A 1 is the standard deviation in the first image window; a 2 is the standard deviation in the second image window;
the transformation range of the area correlation coefficient is from-1 to 1, the similarity is from the minimum to the maximum, and in practical application, the calculated result is compared with a certain preset threshold value P ij;
If the correlation coefficient is larger than the threshold value P ij, reserving the feature point pairs corresponding to the correlation coefficients as candidate matching, and if the correlation coefficient is smaller than the threshold value P ij, setting the feature point pairs to be 0;
and S33, matching the image characteristic points by using a robustness algorithm of characteristic point matching, and eliminating mismatching.
4. The land area measurement method for investigation of homeland resources according to claim 1, wherein: step S5 comprises the steps of:
S51, dividing the obtained gray image into L-level gray, counting the gray value of the pixel, and taking the intermediate value as an initial threshold value T 0; carrying out gray scale treatment on the areas contained in the closed contour image;
S52, analyzing the image into two areas by using the selected initial threshold T 0, wherein all gray values are larger than a pixel composition area G1 of T 0, and all gray values are smaller than a pixel composition area G2 of T 0;
S53, calculating gray average values mu 1 and mu 2 in the areas G1 and G2;
μ1=(n1+2n2+3n3……+ini)/(n1+n2+n3……+ni);0≤i≤Tj;
μ2=(n1+2n2+3n3……+ini)/(n1+n2+n3……+ni);Tj≤i≤L-1;
wherein n i is a gray value, i is the number of pixels with the gray value of i, j is the iteration number, tj is a threshold value after the jth iteration, and tj=t 0 is calculated initially; l is the number of gray levels;
s54, calculating a new threshold value:
Tj+1=1/2(μ1+μ2)
S55, repeating the steps S52-S54 until the difference between T j+1 and T j is smaller than a given value; since the gray values of the image are integers, the difference between the two gray values is also an integer, and we select the given value to be 1; separating the target boundary region from the background;
S56, judging all the areas contained in the target boundary area as land parts to be measured.
5. The land area measurement method for investigation of homeland resources according to claim 1, wherein: and the data acquisition module comprises an unmanned aerial vehicle, a high-definition camera with a laser range finder, a total station and a height measuring instrument.
6. The land area measurement method for investigation of homeland resources according to claim 1, wherein: the PLC control module comprises a display screen, the aerial image of the unmanned aerial vehicle is checked through the display screen, and the flight route of the unmanned aerial vehicle is controlled according to actual needs.
7. The land area measurement method for investigation of homeland resources according to claim 1, wherein: in step S5, the boundary of the measurement land is determined in an assisted manner by manually selecting the boundary.
8. The land area measurement method for investigation of homeland resources according to claim 2, wherein: the calibration reference object is square, so that the area of the calibration reference object is conveniently calculated.
9. A land area measurement system for territory resource investigation using the land area measurement method for territory resource investigation of claim 1, comprising: the system comprises a route planning module, a data acquisition module, an image matching module, a contour construction module, an image processing module, an intra-boundary area calculation module and a PLC control module; the method is characterized in that:
The route planning module is used for planning a route of unmanned aerial vehicle aerial survey according to the terrain and the outline of the land to be measured, planning a route along the land boundary, and the planned aerial survey route is a closed route;
the data acquisition module is used for performing aerial photographing on the land boundary according to the planned route to acquire a high-definition image;
the image matching module is used for matching the images shot by the data acquisition module through the extraction of the image feature points; extracting feature points through Harris operators; the pixel points reflecting the same space point are corresponding;
The contour construction module adopts a linear trigonometry to realize three-dimensional reconstruction of key points, adopts a rotary workbench method to splice images, and splices three-dimensional space point sets reconstructed by every two images together to splice the images together; constructing a profile image of the measured land closure;
the image processing module is used for carrying out gray level processing on the land contour image obtained by the contour construction module; processing the obtained gray level image by adopting a threshold method, and extracting a boundary of land measurement by extracting a land boundary target from the background;
the area calculation module of the area in the boundary adopts boundary tracking to extract the outline of the image area, and simultaneously records the label and the searching direction of each boundary point to obtain the area of the target area:
the PLC control module is connected with the data acquisition module, the image matching module and the outline construction module in a network mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410064807.5A CN117570881B (en) | 2024-01-17 | 2024-01-17 | Land area measurement system and method for investigation of homeland resources |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410064807.5A CN117570881B (en) | 2024-01-17 | 2024-01-17 | Land area measurement system and method for investigation of homeland resources |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117570881A CN117570881A (en) | 2024-02-20 |
| CN117570881B true CN117570881B (en) | 2024-05-17 |
Family
ID=89888552
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410064807.5A Active CN117570881B (en) | 2024-01-17 | 2024-01-17 | Land area measurement system and method for investigation of homeland resources |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117570881B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117876621A (en) * | 2024-03-07 | 2024-04-12 | 贵州省第一测绘院(贵州省北斗导航位置服务中心) | Homeland mapping method based on high-resolution remote sensing image and topographic data |
| CN117907242A (en) * | 2024-03-15 | 2024-04-19 | 贵州省第一测绘院(贵州省北斗导航位置服务中心) | Homeland mapping method, system and storage medium based on dynamic remote sensing technology |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015079412A (en) * | 2013-10-18 | 2015-04-23 | 三井住友建設株式会社 | Method for calculating land area by use and device for calculating land area by use |
| CN105277144A (en) * | 2015-10-16 | 2016-01-27 | 浙江工业大学 | Land area rapid detection method based on binocular vision and detection device thereof |
| CN105698715A (en) * | 2016-02-29 | 2016-06-22 | 北方民族大学 | Single-light-beam land area measurement system based on unmanned aerial vehicle and measurement method thereof |
| CN205808364U (en) * | 2016-02-29 | 2016-12-14 | 北方民族大学 | A kind of single beam Land area measure system based on unmanned plane |
| CN108007395A (en) * | 2017-10-17 | 2018-05-08 | 天津飞眼无人机科技有限公司 | A kind of Land area measure system based on unmanned plane |
| CN108805915A (en) * | 2018-04-19 | 2018-11-13 | 南京市测绘勘察研究院股份有限公司 | A kind of close-range image provincial characteristics matching process of anti-visual angle change |
| CN212007102U (en) * | 2020-04-28 | 2020-11-24 | 南京信息工程大学 | Gas leakage fire catching unmanned aerial vehicle intelligent monitoring and processing system based on FPGA and ARM |
| CN112179294A (en) * | 2020-09-21 | 2021-01-05 | 杭州星网测绘技术有限公司 | Land proofreading method, device and system |
| CN113449594A (en) * | 2021-05-25 | 2021-09-28 | 湖南省国土资源规划院 | Multilayer network combined remote sensing image ground semantic segmentation and area calculation method |
| KR20220064073A (en) * | 2020-11-11 | 2022-05-18 | (주)한국공간정보통신 | Methods for automatic calculation of ecological area ratio using GIS and land cover road |
| CN115235386A (en) * | 2022-07-30 | 2022-10-25 | 中国科学院西北生态环境资源研究院 | Glacier area change detection method based on aerial photography |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105466364A (en) * | 2014-09-11 | 2016-04-06 | 富泰华工业（深圳）有限公司 | Area measuring system and method for irregular figure |
-
2024
- 2024-01-17 CN CN202410064807.5A patent/CN117570881B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015079412A (en) * | 2013-10-18 | 2015-04-23 | 三井住友建設株式会社 | Method for calculating land area by use and device for calculating land area by use |
| CN105277144A (en) * | 2015-10-16 | 2016-01-27 | 浙江工业大学 | Land area rapid detection method based on binocular vision and detection device thereof |
| CN105698715A (en) * | 2016-02-29 | 2016-06-22 | 北方民族大学 | Single-light-beam land area measurement system based on unmanned aerial vehicle and measurement method thereof |
| CN205808364U (en) * | 2016-02-29 | 2016-12-14 | 北方民族大学 | A kind of single beam Land area measure system based on unmanned plane |
| CN108007395A (en) * | 2017-10-17 | 2018-05-08 | 天津飞眼无人机科技有限公司 | A kind of Land area measure system based on unmanned plane |
| CN108805915A (en) * | 2018-04-19 | 2018-11-13 | 南京市测绘勘察研究院股份有限公司 | A kind of close-range image provincial characteristics matching process of anti-visual angle change |
| CN212007102U (en) * | 2020-04-28 | 2020-11-24 | 南京信息工程大学 | Gas leakage fire catching unmanned aerial vehicle intelligent monitoring and processing system based on FPGA and ARM |
| CN112179294A (en) * | 2020-09-21 | 2021-01-05 | 杭州星网测绘技术有限公司 | Land proofreading method, device and system |
| KR20220064073A (en) * | 2020-11-11 | 2022-05-18 | (주)한국공간정보통신 | Methods for automatic calculation of ecological area ratio using GIS and land cover road |
| CN113449594A (en) * | 2021-05-25 | 2021-09-28 | 湖南省国土资源规划院 | Multilayer network combined remote sensing image ground semantic segmentation and area calculation method |
| CN115235386A (en) * | 2022-07-30 | 2022-10-25 | 中国科学院西北生态环境资源研究院 | Glacier area change detection method based on aerial photography |
Non-Patent Citations (2)
| Title |
|---|
| 地籍空间数据库建设与质量控制;朱光兴;李龙强;湛国毅;;科技创新导报;20100801(第22期);全文 * |
| 采煤塌陷积水区面积无人机采集方法;迟臣鑫;陈伟强;朱鹏程;查剑锋;殷和健;马晨阳;;金属矿山;20200815(第08期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117570881A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117570881B (en) | Land area measurement system and method for investigation of homeland resources | |
| CN104484648B (en) | Robot variable visual angle obstacle detection method based on outline identification | |
| CN110263717B (en) | Method for determining land utilization category of street view image | |
| CN109886939B (en) | Bridge crack detection method based on tensor voting | |
| CN109840553B (en) | Extraction method and system of cultivated land crop type, storage medium and electronic equipment | |
| CN110223389B (en) | Scene modeling method, system and device fusing image and laser data | |
| CN111915746B (en) | Weak-labeling-based three-dimensional point cloud target detection method and labeling tool | |
| CN107424142A (en) | A kind of weld joint recognition method based on saliency detection | |
| CN108648274A (en) | A kind of cognition point cloud map creation system of vision SLAM | |
| CN109544501A (en) | A kind of transmission facility defect inspection method based on unmanned plane multi-source image characteristic matching | |
| CN109900274B (en) | Image matching method and system | |
| CN111104850B (en) | Remote sensing image building automatic extraction method and system based on residual error network | |
| CN109829476A (en) | End-to-end three-dimension object detection method based on YOLO | |
| CN113313047A (en) | Lane line detection method and system based on lane structure prior | |
| CN110689512A (en) | Method for quickly splicing and fusing annular images of panoramic video in hole into image | |
| CN116758234A (en) | Mountain terrain modeling method based on multipoint cloud data fusion | |
| CN113658147B (en) | Workpiece size measuring device and method based on deep learning | |
| CN110349209A (en) | Vibrating spear localization method based on binocular vision | |
| CN112215893B (en) | Method, device and equipment for determining target two-dimensional center coordinate point and ranging system | |
| CN104700359A (en) | Super-resolution reconstruction method of image sequence in different polar axis directions of image plane | |
| CN116188943A (en) | Solar radio spectrum burst information detection method and device | |
| CN112819900B (en) | Method for calibrating internal azimuth, relative orientation and distortion coefficient of intelligent stereography | |
| CN109883400A (en) | Fixed station Automatic Targets and space-location method based on YOLO-SITCOL | |
| CN112164118B (en) | Geographic image processing system and method | |
| CN115661694A (en) | Intelligent detection method, system, storage medium and electronic equipment for light-weight main transformer focusing on key characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |