Recognition Method and System for Incomplete Squares, Auxiliary Positioning Method and Device

Abstract

The present invention relates to the field of visual measurement technology, and provides a method and system for identifying an incomplete square, an auxiliary positioning method and device. The identification method includes: acquiring an image of an incomplete square in the field of view, and adaptively binarizing the image; extracting all closed contours in the binarized image; retaining the contours that satisfy the first screen size ratio relationship; the first screen size ratio relationship is that the total length of the contour line is between 0.03 times the larger value of the width and height of the image and 4 times the larger value of the width and height of the image; retaining the convex polygon contour of the incomplete square to be identified; finding at least two vertices not on the edge of the screen; supplementing the vertices of the square according to the at least two vertices; and constructing a complete square according to the supplemented vertices of the square. This solution can perform visual positioning, the calculation method is simple, and more position information of the drone can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of visual measurement, and in particular to a method and system for identifying an incomplete square, and an auxiliary positioning method and device. Background Art

[0002] With the gradual diversification of the application fields of unmanned aerial vehicles (UAVs) in the industry, in the face of complex and harsh operating environments, reliability and intelligence have become important factors in measuring the maturity of UAV systems. As a part of the intelligent application of space-based robots, the UAV vision-assisted positioning method can effectively avoid position offsets caused by unreliable satellite navigation signals in some application scenarios.

[0003] The existing CN110989687A proposes to guide the UAV to land with a multi-layer square pattern with a central coincidence; this method has strict requirements for the pattern and can only be used for height calculation, and the obtained position information is too little, which is not conducive to the intelligent application of the aircraft.

[0004] Moreover, when the UAV lands, it will get closer and closer to the square pattern, and it may not be able to capture the entire pattern, resulting in inability to position.

[0005] Therefore, it is necessary to provide a method and system for identifying an incomplete square, and an auxiliary positioning method and device, which can perform visual positioning, the solution method is simple, and more UAV position information can be obtained.

[0006] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present application, and therefore it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] The main purpose of the present invention is to overcome the problem of inability to position or inaccurate positioning when the UAV lands, and to provide a method and system for identifying an incomplete square, and an auxiliary positioning method and device, which can perform visual positioning, the solution method is simple, and more UAV position information can be obtained.

[0008] To achieve the above object, the first aspect of the present invention provides a method for identifying an incomplete square in a field of view, the square having a different color from its background color, comprising the following steps:

[0009] Obtain an image of an incomplete square in the field of view, and adaptively binarize the image;

[0010] Extract all closed contours in the binarized image;

[0011] Retain the contour that satisfies the first screen size ratio relationship, and the contour and vertices cannot form a complete square; the first screen size ratio relationship is that the total length of the contour line is between 0.03 times the larger value of the width and height of the image and 4 times the larger value of the width and height of the image;

[0012] Retain the convex polygon contour of the incomplete square that needs to be recognized;

[0013] Find at least two vertices that are not on the edge of the screen;

[0014] Complete the vertices of the square according to at least two vertices;

[0015] Construct a complete square according to the vertices of the completed square.

[0016] According to an exemplary embodiment of the present invention, the finding at least two vertices that are not on the edge of the screen includes:

[0017] According to the convex polygon contour of the incomplete square that needs to be recognized, obtain a total of 4, 5, or 6 vertices; if the number of vertices is 4 or 6, find two vertices that are not on the edge of the screen; if the number of vertices is 5, find three vertices that are not on the edge of the screen.

[0018] According to an exemplary embodiment of the present invention, the completing the vertices of the square according to at least two vertices includes:

[0019] If the total number of vertices is 4, construct two squares based on the two vertices that are not on the edge of the screen, form two groups of candidate third and fourth vertices, and retain the candidate third and fourth vertices on the same side as the connection line of the two vertices on the edge of the screen as the third and fourth vertices of the incomplete square;

[0020] If the total number of vertices is 5, construct three parallelograms based on the three vertices that are not on the edge of the screen, form three candidate fourth vertices, and retain the candidate fourth vertices whose cross-connection line length between vertices satisfies the first predetermined ratio relationship and the sequential connection line length between vertices satisfies the second predetermined ratio relationship as the fourth vertices of the incomplete square;

[0021] If the total number of vertices is 6, construct a square based on the two vertices that are not on the edge of the screen, form the third and fourth vertices.

[0022] As the second aspect of the present invention, the present invention provides a recognition system for incomplete squares in the field of view, including a contour extraction module, a vertex completion module, and a square construction module connected in sequence;

[0023] The contour extraction module is used to obtain the image of an incomplete square in the field of view, perform adaptive binarization on the image; extract all closed contours in the binarized image; retain the contours that satisfy the first screen size ratio relationship; retain the convex polygon contours of the incomplete squares to be recognized.

[0024] The vertex completion module is used to find at least two vertices that are not on the edge of the screen; complete the vertices of the square according to the at least two vertices.

[0025] The square construction module is used to construct a complete square according to the completed vertices of the square.

[0026] As the third aspect of the present invention, the present invention provides an auxiliary positioning method for aircraft landing, including the following steps:

[0027] S1: Set nested markers on the ground and obtain the world coordinates of each vertex of each square.

[0028] The nested markers include n squares, the nth square is nested inside the (n - 1)th square, n is a natural number greater than or equal to 2, and the centers of the nth square and the (n - 1)th square are not in the same position.

[0029] S2: Continuously photograph the nested markers from the landing aircraft located above the nested markers, and keep the camera focal length unchanged during the photographing process.

[0030] S3: Identify the squares in sequence starting from the first square. If the largest square in the field of view is incomplete, use the recognition method of the incomplete squares in the field of view to identify the squares.

[0031] S4: Obtain the position of the aircraft according to the image coordinates and world coordinates of the respective vertices corresponding to the (n - 1)th square and the nth square found, where at least 2 vertices of the nth square appear in the photographed image.

[0032] According to an exemplary embodiment of the present invention, in step S1, in the nested markers, two adjacent squares with similar sizes have different colors, or two adjacent squares with similar sizes have the same color and there is an isolation figure with a different color and not a square that is nested with the two adjacent squares with similar sizes respectively.

[0033] According to an exemplary embodiment of the present invention, in step S4, the obtaining the position of the aircraft according to the image coordinates and world coordinates of the respective vertices corresponding to the (n - 1)th square and the nth square found includes:

[0034] Obtain the pointing direction of the camera according to the positional relationship between the (n - 1)th square and the nth square.

[0035] Order the four vertices of the (n - 1)-th square according to the orientation of the camera;

[0036] According to the ordering of the four vertices, solve the PnP to match the pose relationship between the world coordinate system and the image coordinate system of the four vertices, and obtain the relative displacement transformation matrix;

[0037] Obtain the position information of the aircraft according to the relative displacement transformation matrix.

[0038] According to an exemplary embodiment of the present invention, the obtaining the orientation of the camera according to the positional relationship between the (n - 1)-th square and the n-th square includes:

[0039] Perform projective transformation on the respective vertices of the two squares to a standard square;

[0040] Match the relative position of the two squares with the template of the nested flag to obtain the orientation of the camera.

[0041] According to a fourth aspect of the present invention, the present invention provides an auxiliary positioning device for aircraft landing, including: a nested flag, a camera, an identification system for incomplete squares in the field of view, and a position calculation module;

[0042] The nested flag is arranged on the ground; the nested flag includes n squares, the n-th square is nested inside the (n - 1)-th square, n is a natural number greater than or equal to 2, and the centers of the n-th square and the (n - 1)-th square are not in the same position;

[0043] The camera is arranged on the aircraft, and continuously takes pictures of the nested flag from the landing aircraft located above the nested flag, and the camera focal length remains unchanged during the shooting process;

[0044] The position calculation module is used to obtain the world coordinates of the respective vertices of each square; start identifying the squares from the 1st square in sequence. If the largest square in the field of view is incomplete, use the identification system for incomplete squares in the field of view to identify the square; obtain the position of the aircraft according to the image coordinates and world coordinates of the respective corresponding vertices of the (n - 1)-th square and the n-th square found, where at least 2 vertices of the n-th square appear in the shooting picture.

[0045] The advantageous effects of the present invention are:

[0046] The present invention proposes a new eccentric asymmetric square flag, with multiple squares having different positional relationships and proportional relationships, and can identify and locate the position of the aircraft in three cases of combination, nesting, and partial loss, which can cover working conditions such as aircraft cruising, taking off and landing, and abnormal incomplete signs, and can identify incomplete squares, expanding the scope of use. Description of the Drawings

[0047] By describing its exemplary embodiments in detail with reference to the accompanying drawings, the above and other objects, features, and advantages of the present application will become more apparent. The following described drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0048] Figure 1 A schematic diagram schematically shows the first nested mark.

[0049] Figure 2 A schematic diagram schematically shows the second nested mark.

[0050] Figure 3 A schematic diagram schematically shows the respective vertices of the square of the positioning mark for identifying the first aircraft.

[0051] Figure 4 A schematic diagram schematically shows the respective vertices of the square of the positioning mark for identifying the second aircraft.

[0052] Figure 5 A schematic diagram schematically shows the step diagram of an auxiliary positioning method for an aircraft to land.

[0053] Figure 6 A schematic diagram schematically shows the vertices of the first incomplete square.

[0054] Figure 7 A schematic diagram schematically shows the vertices of the second incomplete square.

[0055] Figure 8 A schematic diagram schematically shows the vertices of the third incomplete square.

[0056] Figure 9 A schematic diagram schematically shows the four - direction diagram of the nested mark.

[0057] Figure 10 A schematic diagram schematically shows the actual picture of the aircraft shooting the nested mark.

[0058] Figure 11 A schematic diagram schematically shows the actual picture of the aircraft shooting the nested mark (another case). Detailed implementation manners

[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. Like reference numerals in the figures denote like or similar parts, and thus their repetitive description will be omitted.

[0060] In addition, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of this application. However, those skilled in the art will realize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be used. In other cases, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of this application.

[0061] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0062] The flowcharts shown in the drawings are merely illustrative and do not necessarily include all the content and operations / steps, nor are they necessarily executed in the order described. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may change according to the actual situation.

[0063] It should be understood that although terms such as first, second, and third may be used herein to describe various components, these components should not be limited by these terms. These terms are used to distinguish one component from another. Thus, the first component discussed below can be referred to as the second component without departing from the teachings of the concept of this application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0064] Those skilled in the art can understand that the drawings are only schematic diagrams of the example embodiments, and the modules or processes in the drawings are not necessarily essential for implementing this application, so they cannot be used to limit the protection scope of this application.

[0065] According to the first specific embodiment of the present invention, the present invention provides an auxiliary positioning device for aircraft landing, including: a nested marker, a camera, an identification system for an incomplete square within the field of view, and a position calculation module.

[0066] The nested marker is set on the ground; the nested marker includes n squares, the nth square is nested within the (n - 1)th square, where n is a natural number greater than or equal to 2, and the centers of the nth square and the (n - 1)th square are not in the same position. The recognized figure is a square, which is convenient for recognition because a square has the characteristics of equal side lengths and equal diagonal lengths, while the characteristics of a rectangle are relatively less obvious and are prone to false detection, especially when observed at an inclined angle. Similarly, for example, marker points are mostly regular circles instead of ellipses.

[0067] In the nested marker, two adjacent nested squares with similar sizes have different colors, or two adjacent nested squares with similar sizes have the same color and there is an isolating figure with a different color and not in the shape of a square that is nested with each of the two adjacent nested squares with similar sizes respectively.

[0068] The ratio range of the side length of the (n - 1)th square to the nth square is (3 - 6):1. As Figure 1 shown, Figure 1 For the positioning marker in the first case, the first square is a black square, the second square is a white square, the second square is nested within the first square and is not in the same position as the center of the first square, and the two together form an eccentric and asymmetric square, Figure 3 and the vertices of the marked square can be clearly seen. As Figure 2 shown, Figure 2 For the positioning marker in the second case, the first square is a large black square, the isolating figure is a white rectangle, and the second square is a small black square, Figure 4 and the vertices of the marked square can be clearly seen. The isolating figure is nested within the first square, and the second square is nested within the isolating figure. The reason for making the colors of the two squares different or setting an isolating figure in the middle is to be able to clearly identify the sides and vertices of the square.

[0069] The camera is set on the aircraft, and the nested marker is continuously photographed from the landing aircraft located above the nested marker, with the camera focal length remaining unchanged during the photographing process. Real-time lens focal length information is required for the calculation of the auxiliary positioning position, and using a fixed focal length is for the convenience of position calculation during the flight process.

[0070] An identification system for incomplete squares within the field of view, comprising a contour extraction module, a vertex completion module, and a square construction module connected in sequence. The contour extraction module is used to obtain an image of an incomplete square within the field of view, perform image adaptive binarization; extract all closed contours in the binarized image; retain contours that satisfy the first screen size ratio relationship; retain the convex polygon contours of the incomplete squares to be identified. The vertex completion module is used to find at least two vertices not on the edge of the screen; complete the vertices of the square based on the at least two vertices. The square construction module is used to construct a complete square based on the completed vertices of the square.

[0071] The position calculation module is used to obtain the world coordinates of each vertex of each square; start identifying squares sequentially from the first square. If the largest square within the field of view is incomplete, use the identification system for incomplete squares within the field of view to identify the square; obtain the position of the aircraft based on the image coordinates and world coordinates of the corresponding vertices of the (n - 1)th square and the nth square found, where at least two vertices of the nth square appear in the captured image.

[0072] According to the second specific embodiment of the present invention, the present invention provides an auxiliary positioning method for aircraft landing, using the auxiliary positioning device for aircraft landing in the first specific embodiment, as Figure 5 shown, including the following steps:

[0073] S1: Set nested markers on the ground and obtain the world coordinates of each vertex of each square.

[0074] The nested markers include n squares, the nth square is nested within the (n - 1)th square, n is a natural number greater than or equal to 2, and the centers of the nth square and the (n - 1)th square are not in the same position.

[0075] Among the nested markers, two adjacent squares with similar sizes have different colors, or two adjacent squares with similar sizes have the same color and there is an isolation figure with a different color and not a square that is nested with each of the two adjacent squares with similar sizes respectively.

[0076] S2: Continuously capture the nested markers from an aircraft landing above the nested markers, with the camera focal length remaining unchanged during the capture process.

[0077] S3: Start identifying squares sequentially from the first square. If the largest square within the field of view is incomplete, use the identification method for incomplete squares within the field of view to identify the square.

[0078] An identification method for incomplete squares within the field of view, where the square has a different color from its background color, including the following steps:

[0079] S31: Obtain the image of an incomplete square within the field of view, and perform adaptive binarization on the image.

[0080] S32: Extract all closed contours in the binarized image.

[0081] S33: Retain the contours that satisfy the first screen size ratio relationship and whose contours and vertices cannot form a complete square.

[0082] The first screen size ratio relationship is that the total length of the contour line is between 0.03 times the larger value of the width and height of the image and 4 times the larger value of the width and height of the image; 0.03×max(w, h) ≤ total contour line length ≤ 4×max(w, h), where w represents the image width and h represents the image height; squares that are too large or too small within the screen will not be recognized to avoid false detection.

[0083] The incomplete square appears due to the field of view problem, that is, it appears because the square mark falls on the edge of the screen. By filtering, it is possible to prevent the detection of irrelevant squares within the screen because too many irrelevant squares need to be excluded, and they will affect the calculation efficiency.

[0084] The combination of vertex and contour features can be considered an incomplete square.

[0085] S34: Retain the convex polygon contour of the incomplete square to be recognized.

[0086] S35: Find at least two vertices that are not on the edge of the screen.

[0087] Finding at least two vertices that are not on the edge of the screen includes:

[0088] Based on the convex polygon contour of the incomplete square to be recognized, obtain a total of 4, 5, or 6 vertices; if the number of vertices is 4 or 6, then find two vertices that are not on the edge of the screen; if the number of vertices is 5, then find three vertices that are not on the edge of the screen.

[0089] S36: Complete the vertices of the square according to at least two vertices.

[0090] Completing the vertices of the square according to at least two vertices includes:

[0091] As Figure 6 shown, if the total number of vertices is 4, then construct two squares based on the two vertices that are not on the edge of the screen, forming two groups of candidate third and fourth vertices, and retain the candidate third and fourth vertices on the same side as the connection line of the two vertices on the edge of the screen as the third and fourth vertices of this incomplete square. Figure 6 It is a 2V1E structure, that is, there are only 2 vertices and 1 complete side. Figure 6The four points among them are two groups of candidate third vertices and fourth vertices. The solid dot located above is used as the third vertex and the fourth vertex, and the hollow dots located below are excluded.

[0092] As Figure 7 shown, if the total number of vertices is 5, three parallelograms are constructed based on the three vertices not on the edge of the screen, forming three candidate fourth vertices. The candidate fourth vertices whose cross-connection line lengths satisfy the first predetermined proportional relationship and whose sequential connection line lengths between vertices satisfy the second predetermined proportional relationship are retained as the fourth vertex of the incomplete square. The parallelogram is used because it is more convenient in engineering to establish a parallelogram with parallel lines instead of points and angles. As a preferred implementation, a square or a rectangle is constructed based on the three vertices not on the edge of the screen. Figure 7 It is a 3V2E structure, that is, there are only 3 vertices and 2 complete edges. Figure 7 Among them, the three points are candidate fourth vertices. The solid dot located above is used as the fourth vertex, and the remaining hollow dots are excluded. The first predetermined proportional relationship: the length ratio of the cross-line X / Y is between 0.5 and 2, so that a rhombus will not be misdetected. X and Y respectively represent the two cross-lines. The second predetermined proportional relationship: the lengths of the sequential connection lines A / B / C / D of the vertices each account for more than 20% of the total length of A + B + C + D, so that an irregular quadrilateral will not be misdetected. A, B, C, and D respectively represent the four vertices.

[0093] As Figure 8 shown, if the total number of vertices is 6, a square is constructed based on the two vertices not on the edge of the screen, forming the third vertex and the fourth vertex. Figure 8 It is a 2V0E structure, that is, there are only 2 vertices and 0 complete edges. Figure 8 Among them, the two solid dots are respectively the third vertex and the fourth vertex.

[0094] S37: Construct a complete square according to the vertices of the completed square.

[0095] S4: Obtain the position of the aircraft according to the image coordinates and world coordinates of the respective vertices corresponding to the (n - 1)-th square and the n-th square, where at least 2 vertices of the n-th square appear in the captured image.

[0096] Obtaining the position of the aircraft according to the image coordinates and world coordinates of the respective vertices corresponding to the (n - 1)-th square and the n-th square includes:

[0097] S41: Obtain the pointing direction of the camera according to the positional relationship between the (n - 1)-th square and the n-th square.

[0098] Obtaining the pointing direction of the camera according to the positional relationship between the (n - 1)-th square and the n-th square includes:

[0099] S411: Project the vertices of the two squares through projective transformation to the standard square;

[0100] S412: Match the relative positions of the two squares with the template of the nesting flag to obtain the camera pointing direction.

[0101] What needs to be confirmed for the positional relationship is the positional relationship between the (n - 1)-th square and the n-th square.

[0102] The connection lines of the vertices are a processing step in the subsequent recognition process. The recognition has always been implemented based on the four vertices. To determine whether the four vertices are the four vertices of a square, the length information of the total four connection lines of two adjacent vertices is required for judgment. Define the template of the nesting flag as the standard pointing diagram, and match the template of the positioning flag of the aircraft according to the relationship between the side length and the diagonal. After the picture rotates, it is easy to recognize the change in the pointing direction according to the relative relationship between the inner and outer squares.

[0103] As Figure 9 shown, Figure 9 is Figure 1 a schematic diagram of the positioning flags in four directions. Figure 1 In, the second square is close to the upper edge of the first square. Define the camera direction at this time as pointing upward. Figure 9 (a) shows the camera pointing upward. Figure 9 (b) shows the camera pointing to the right. Figure 9 (c) shows the camera pointing downward. Figure 9 (d) shows the camera pointing to the left. The advantage of the eccentric square is that the camera direction in the picture can be known.

[0104] S42: Sequentially number the four vertices of the (n - 1)-th square according to the camera pointing direction.

[0105] The four sequentially numbered points are #0, #1, #2, and #3 respectively.

[0106] The sequential numbering is based on the template of the nesting flag. As Figure 3 shown, the vertex at the upper left corner of the first square is #0, the vertex at the upper right corner is #1, the vertex at the lower right corner is #2, and the vertex at the lower left corner is #3; the vertex at the upper left corner of the second square is #0, the vertex at the upper right corner is #1, the vertex at the lower right corner is #2, and the vertex at the lower left corner is #3. Similarly, Figure 4Among them, for the first square, the vertex at the upper left corner is #0, the vertex at the upper right corner is #1, the vertex at the lower right corner is #2, and the vertex at the lower left corner is #3; for the second square, the vertex at the upper left corner is #0, the vertex at the upper right corner is #1, the vertex at the lower right corner is #2, and the vertex at the lower left corner is #3; for the third square, the vertex at the upper left corner is #0, the vertex at the upper right corner is #1, the vertex at the lower right corner is #2, and the vertex at the lower left corner is #3.

[0107] Figure 10 is Figure 1 a real - shot image of the positioning mark. Among them, for the second square (the white square) on the upper side, it can be known that the camera points upward at this time. The point where the first square (the black square) is recognized to be on the left side of the white square is 17#0. According to the clockwise numbering, the remaining corners are 17#1, 17#2, and 17#3 respectively. Figure 11 is Figure 1 another real - shot image of the positioning mark. Figures 10 to 11 , the positioning mark is from far to near and rotated. For the second square (the white square) on the upper side, it can be known that the camera points upward at this time. The point where the first square (the black square) is recognized to be on the upper side of the white square is 17#0. According to the clockwise numbering, the remaining corners are 17#1, 17#2, and 17#3 respectively.

[0108] S43: According to the fixed order of the four vertices, solve the pose relationship between the world coordinate system and the image coordinate system of the four vertices through PnP algorithm to obtain the relative displacement transformation matrix.

[0109] In the camera view, directly or indirectly extract the 2D coordinates of the square vertices on the image. Through PnP algorithm, match the pose relationship between the 3D world coordinates and the 2D image coordinates to obtain the relative displacement transformation matrix R / T, where R is the rotation matrix and T is the translation vector.

[0110] S44: Obtain the position information of the aircraft according to the relative displacement transformation matrix.

[0111] This solution proposes a new eccentric asymmetric square mark with multiple squares having different position relationships and proportional relationships. It can identify and locate the position of the aircraft in three cases: combination, nesting, and partial loss, which can cover working conditions such as aircraft cruising, take - off and landing, and abnormal situations with incomplete marks, and can identify incomplete squares, expanding the scope of use.

[0112] The above has specifically shown and described the exemplary embodiments of the present invention. It should be understood that the present invention is not limited to the detailed structures, setting methods, or implementation methods described here; on the contrary, the present invention is intended to cover various modifications and equivalent settings included within the spirit and scope of the appended claims.

Claims

1. A method for identifying an incomplete square in a field of view, wherein the square is different in color from its background, characterized in that: The following steps are involved: Acquire an incomplete square image within the field of view and perform adaptive binarization of the image; Extract all closed contours in the binary image; Retaining contours that satisfy a first screen size ratio relationship and whose contours and vertices cannot form a complete square; the first screen size ratio relationship is that the total length of the contour line is between 0.03 times the larger value of the width and height of the image and 4 times the larger value of the width and height of the image; Keep the convex polygon outline of the incomplete square that needs to be recognized; Find at least two vertices that are not on the edge of the picture; Fill in the vertices of the square based on at least two vertices; Construct a complete square from the vertices of the completed square; The step of finding at least two vertices that are not on the edge of the picture comprises: According to the convex polygonal outline of the incomplete square to be identified, the number of vertices is obtained to be 4, 5 or 6 in total; if the number of vertices is 4 or 6, two vertices that are not on the edge of the picture are found; if the number of vertices is 5, three vertices that are not on the edge of the picture are found; The step of filling in the vertices of a square according to at least two vertices comprises: If the total number of vertices is 4, construct two squares based on the two vertices that are not at the edge of the picture to form two groups of candidate third vertices and fourth vertices, and retain the candidate third vertices and fourth vertices on the same side of the line connecting the two vertices at the edge of the picture as the third vertex and fourth vertex of the incomplete square; If the total number of vertices is 5, three parallelograms are constructed based on the three vertices that are not on the edge of the screen to form three candidate fourth vertices, and the candidate fourth vertices whose intersecting line lengths satisfy the first predetermined ratio and whose sequential line lengths satisfy the second predetermined ratio are retained as the fourth vertices of the incomplete square; If the number of vertices is 6 in total, a square is constructed based on two vertices that are not on the edge of the picture, forming a third vertex and a fourth vertex.

2. A system for identifying an incomplete square in a field of view, characterized in that: It includes a contour extraction module, a vertex completion module and a square construction module which are connected in sequence; The contour extraction module is used to obtain an image of an incomplete square in the field of view, and the image is adaptively binarized; all closed contours are extracted from the binarized image; contours that meet a first picture size ratio relationship are retained; and convex polygonal contours of the incomplete square to be identified are retained; the first picture size ratio relationship is that the total length of the contour line is between 0.03 times the larger value of the width and height of the image and 4 times the larger value of the width and height of the image; The vertex completion module is used to find at least two vertices that are not on the edge of the picture; and complete the vertices of the square according to the at least two vertices; The square construction module is used to construct a complete square according to the vertices of the padded square; The step of finding at least two vertices that are not on the edge of the picture comprises: According to the convex polygonal outline of the incomplete square to be identified, the number of vertices is obtained to be 4, 5 or 6 in total; if the number of vertices is 4 or 6, two vertices that are not on the edge of the picture are found; if the number of vertices is 5, three vertices that are not on the edge of the picture are found; The step of filling in the vertices of a square according to at least two vertices comprises: If the total number of vertices is 4, construct two squares based on the two vertices that are not at the edge of the picture to form two groups of candidate third vertices and fourth vertices, and retain the candidate third vertices and fourth vertices on the same side of the line connecting the two vertices at the edge of the picture as the third vertex and fourth vertex of the incomplete square; If the total number of vertices is 5, three parallelograms are constructed based on the three vertices that are not on the edge of the screen to form three candidate fourth vertices, and the candidate fourth vertices whose intersecting line lengths satisfy the first predetermined ratio and whose sequential line lengths satisfy the second predetermined ratio are retained as the fourth vertices of the incomplete square; If the number of vertices is 6 in total, a square is constructed based on two vertices that are not on the edge of the picture, forming a third vertex and a fourth vertex.

3. An auxiliary positioning method for aircraft landing, characterized in that: The following steps are involved: S1: Set the nested mark on the ground and obtain the world coordinates of each vertex of each square; The nested mark includes n squares, the nth square is nested in the n-1th square, n is a natural number greater than or equal to 2, and the centers of the nth square and the n-1th square are not in the same position; S2: Continuously photograph the nesting mark from a landing aircraft located above the nesting mark, and the camera focal length remains unchanged during the photographing process; S3: identifying squares in sequence starting from the first square, and if the largest square in the field of view is incomplete, using the method for identifying an incomplete square in the field of view of claim 1 to identify the squares; S4: Obtain the position of the aircraft according to the image coordinates and world coordinates of each vertex corresponding to the found n-1th square and the nth square, wherein the nth square has at least two vertices in the shooting picture.

4. The auxiliary positioning method for aircraft landing according to claim 3, characterized in that: In step S1, in the nested mark, two nested squares of similar size have different colors, or two nested squares of similar size have the same color and there is an isolation figure of different color and not a square nested with the two squares of similar size.

5. The auxiliary positioning method for aircraft landing according to claim 3, characterized in that: In step S4, obtaining the position of the aircraft according to the image coordinates and world coordinates of each vertex corresponding to the found n-1th square and the nth square includes: Get the direction of the camera based on the positional relationship between the n-1th square and the nth square; According to the direction of the camera, order the four vertices of the n-1th square; According to the order of the four vertices, the pose relationship between the world coordinate system and the image coordinate system of the four vertices is matched through PnP solution to obtain the relative displacement transformation matrix; The position information of the aircraft is obtained according to the relative displacement transformation matrix.

6. The auxiliary positioning method for aircraft landing according to claim 5, characterized in that: The step of obtaining the direction of the camera according to the positional relationship between the n-1th square and the nth square includes: Project the vertices of the two squares to the standard square; The relative positions of the two squares are matched with the template of the nested logo to get the camera's pointing direction.

7. An auxiliary positioning device for aircraft landing, characterized in that: include: Nested markers, a camera, a recognition system for an incomplete square in the field of view as claimed in claim 2, and a position solving module; The nested mark is set on the ground; the nested mark includes n squares, the nth square is nested in the n-1th square, n is a natural number greater than or equal to 2, and the centers of the nth square and the n-1th square are not in the same position; The camera is set on the aircraft, and the nesting mark is continuously photographed from the aircraft that is located above the nesting mark and is landing. The focal length of the camera does not change during the photographing process; The position solution module is used to obtain the world coordinates of each vertex of each square; starting from the first square, the squares are identified in sequence. If the largest square in the field of view is incomplete, the incomplete square recognition system in the field of view is used to identify the square; the position of the aircraft is obtained according to the image coordinates and world coordinates of each vertex corresponding to the found n-1th square and the nth square, wherein the nth square has at least 2 vertices in the shooting picture.

Images