Method for discriminating bumpy areas of low-altitude flight based on machine vision

Abstract

The present invention provides a method for discriminating bumpy areas during low-altitude flight based on machine vision, belonging to the field of area discrimination, including: collecting displacement data based on a displacement sensor of a general aviation aircraft to be discriminated to obtain flight data of the general aviation aircraft to be discriminated and determining the flight state at each flight moment; obtaining model parameters from a general aviation aircraft database to establish an aircraft model and performing low-altitude flight simulation according to the corresponding specified flight mode; obtaining a flight image set of the aircraft model during the complete simulation process and obtaining the flight contour parameters of each flight image; determining the acquisition moment of each flight image and matching the flight state, and combining the flight contour parameters at the corresponding acquisition moment to determine the bumpy area. By collecting displacement data to determine the flight state, establishing a model for simulation, determining the contour parameters according to the simulation results, and determining the bumpy area according to the acquisition moment, it is possible to accurately judge the bumpy area under poor data conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of area discrimination, and particularly to a method for discriminating bumpy areas of low-altitude flight based on machine vision. Background Art

[0002] When general aviation aircraft fly at low altitude in different flight states, different degrees of bumps will occur. Different degrees of bumps during the flight of the aircraft will have different degrees of impact on the flight. Therefore, it is necessary to determine the bumpy areas during the low-altitude flight of the aircraft.

[0003] Most of the existing methods for judging and predicting bumpy areas of flight require relying on multiple continuous and good data, and use the form of nested multiple area weights to discriminate bumpy areas of flight. This discrimination method cannot ensure applicability in general cases and has high requirements for the quality of data.

[0004] Therefore, the present invention provides a method for discriminating bumpy areas of low-altitude flight based on machine vision. Summary of the Invention

[0005] The present invention provides a method for discriminating bumpy areas of low-altitude flight based on machine vision to solve the defects that the existing discrimination methods are not generally applicable and have high requirements for data.

[0006] On the one hand, the present invention provides a method for discriminating bumpy areas of low-altitude flight based on machine vision, including:

[0007] Step 1: Displacement sensors set at preset positions of the general aviation aircraft to be discriminated collect displacement data of the general aviation aircraft to be discriminated during actual low-altitude flight to obtain flight data of the general aviation aircraft to be discriminated, and determine the flight state of the general aviation aircraft to be discriminated at each flight moment;

[0008] Step 2: Obtain the model parameters of the general aviation aircraft to be discriminated from the general aviation aircraft database to establish an aircraft model, and perform low-altitude flight simulation according to the corresponding specified flight mode;

[0009] Step 3: Obtain a flight image set of the aircraft model during the complete simulation process according to the simulation results, and use the contour algorithm of machine vision to obtain the flight contour parameters of each flight image in the flight image set;

[0010] Step 4: Determine the acquisition moment of each flight image in the flight image set to match the flight state at the corresponding flight moment, and combine the flight contour parameters at the corresponding acquisition moment to determine the bumpy areas of the general aviation aircraft to be discriminated during low-altitude flight.

[0011] According to the method for discriminating bumpy areas in low-altitude flight based on machine vision provided by the present invention, determining the flight state of the general aviation aircraft to be discriminated at each flight moment includes:

[0012] Collecting the displacement data of the general aviation aircraft to be discriminated according to the set acquisition time of each displacement sensor;

[0013] Sorting and analyzing all the collected data according to the relative position relationship of all displacement sensors and the relative time relationship of the set acquisition time to determine the flight state of the general aviation aircraft to be discriminated at the corresponding flight moment.

[0014] According to the method for discriminating bumpy areas in low-altitude flight based on machine vision provided by the present invention, sorting and analyzing all the collected data according to the relative position relationship of all displacement sensors and the relative time relationship of the set acquisition time includes:

[0015] Randomly selecting a displacement sensor as the first sensor, determining the relative position relationship between each remaining sensor and the first sensor, and sorting each remaining sensor according to the distance to obtain the number of each displacement sensor;

[0016] Arranging the collected data at the same acquisition moment in the order of the corresponding displacement sensors to obtain the collected data sequence at the same moment, obtaining the sequential relationship of different acquisition moments based on the relative time relationship of the set acquisition time, and sorting the collected data sequences at each same moment from front to back according to the moment sequential relationship to obtain the order of all the collected data.

[0017] According to the method for discriminating bumpy areas in low-altitude flight based on machine vision provided by the present invention, obtaining the model parameters of the general aviation aircraft to be discriminated from the general aviation aircraft database includes:

[0018] Obtaining all the mechanical information of the general aviation aircraft to be discriminated;

[0019] Matching each mechanical information with the flight database respectively;

[0020] Obtaining the model parameters of the aircraft to be recognized based on all the matching results.

[0021] According to the method for discriminating bumpy areas in low-altitude flight based on machine vision provided by the present invention, establishing an aircraft model and performing low-altitude flight simulation according to the corresponding specified flight mode includes:

[0022] Obtaining a parameter comparison table required for establishing the aircraft model, where the parameter comparison table is a blank comparison table containing several parameter descriptions;

[0023] Fill all model parameters into the blank comparison table respectively to establish an aircraft model;

[0024] Compare the flight state of the general aviation aircraft to be discriminated at each flight moment with the flight state at the previous flight moment to obtain a comparison result;

[0025] Based on the comparison result, determine all flight moments when the flight mode changes, and determine the change situation of the flight mode;

[0026] Based on the change situation, determine the specified flight mode of the general aviation aircraft to be discriminated during the actual flight process, and control the aircraft model to perform flight simulation based on the flight mode.

[0027] According to the low-altitude flight bump area discrimination method based on machine vision provided by the present invention, obtain the flight image set of the aircraft model during the complete simulation process according to the simulation result, and use the contour algorithm of machine vision to obtain the flight contour parameters of each flight image in the flight image set, including:

[0028] Obtain the simulated flight video data of the aircraft model during the complete simulation process, and obtain the acquisition time interval of the displacement data and the initial flight simulation moment corresponding to the initial flight moment;

[0029] Based on the acquisition time interval and the initial flight simulation moment, intercept the images of the simulated flight video data to obtain the flight image set of the aircraft model during the complete simulation process;

[0030] Preprocess each flight image in the flight image set to obtain the grayscale image of each flight image;

[0031] Determine the grayscale value of each pixel point in the single grayscale image, classify all pixel points in the single grayscale image, and determine the contour pixel point set;

[0032] Place the single grayscale image in a preset coordinate system, determine the position coordinates of each pixel point in the contour pixel point set, and combine the principle of similar grayscale values to determine the positional relationship of the position coordinates presented by the preset coordinate system, and then determine multiple contour line segments and the length of each contour line segment;

[0033] According to the principle from left to right of the preset coordinate system, obtain the first distance between the two starting points of every two adjacent contour line segments, the second distance between the two ending points, the third distance between the two closest endpoints, and the included angle between the two adjacent contour line segments, and combine the grayscale values of the two adjacent contour line segments to estimate the clarity of the area where the two adjacent contour line segments are located;

[0034]

[0035]

[0036]

[0037]

[0038] represents the first distance corresponding to two adjacent contour segments; represents the second distance corresponding to two adjacent contour segments; represents the third distance corresponding to two adjacent contour segments; represents the minimum value symbol; represents the minimum distance among all distances corresponding to two adjacent contour segments; represents the included angle formed by two adjacent contour segments; represents the angular relationship function corresponding to two adjacent contour segments; represents the clarity of the estimated area where two adjacent contour segments are located; represents the gray value after pixel normalization for two adjacent contour segments; represents and the corresponding clarity correction function; represents the conversion function between gray value and clarity; represents the initial gray value corresponding to two adjacent contour segments; represents the gray value of the i1-th pixel in the first segment among two adjacent contour segments, and the value range of i1 is [1, n01]; represents the gray value of the i2-th pixel in the second segment among two adjacent contour segments, and the value range of i2 is [1, n02]; represents the minimum gray value in the first segment; represents the maximum gray value in the first segment; represents the minimum gray value in the second segment; represents the maximum gray value in the second segment;

[0039] Perform region annotation on all regions, and determine the final clarity of each individual annotation unit in the grayscale image according to the annotation results;

[0040] Adjust the unit brightness of the grayscale image according to the final clarity of each individual annotation unit, and determine the flight contour parameters according to the adjusted image.

[0041] The method for discriminating bumpy areas during low-altitude flight based on machine vision provided by the present invention determines the bumpy areas of a general aviation aircraft to be discriminated during low-altitude flight in combination with the flight profile parameters at the corresponding acquisition moments, including:

[0042] Based on the model parameters of the general aviation aircraft to be discriminated, determine the total contour line length of the general aviation aircraft to be discriminated from the parameter-length database, and based on the flight profile parameters at each acquisition moment, determine the clear contour line segment length and the clear contour line segment position of the general aviation aircraft to be discriminated;

[0043] Determine the ratio of the clear contour line segment length at each acquisition moment to the total contour line length of the general aviation aircraft to be discriminated. At the same time, determine the position comparison between the corresponding clear contour line segment position and the total contour;

[0044] Based on the ratio result and the position comparison result, determine the bumpy state at each acquisition moment, combine the flight state at each acquisition moment to determine the bumpy level under each different flight state, and calibrate the displacement points during the actual flight process based on the bumpy level to determine the bumpy areas of the general aviation aircraft to be discriminated during low-altitude flight.

[0045] The method for discriminating bumpy areas during low-altitude flight based on machine vision provided by the present invention determines the clear contour line segment length and the clear contour line segment position of the general aviation aircraft to be discriminated based on the flight profile parameters at each acquisition moment, including:

[0046] Determine all the sub-contour line segments of the general aviation aircraft to be discriminated according to the flight profile parameters, and sequentially lock the positions of each point on each sub-contour line segment;

[0047] According to the total number of points with positions locked and in combination with the point length of each locked position point, obtain the corresponding clear contour line segment length.

[0048] The method for discriminating bumpy areas during low-altitude flight based on machine vision provided by the present invention determines the flight state by collecting displacement data, establishes a model for simulation, determines the contour parameters according to the simulation results, and determines the bumpy areas according to the acquisition moments, solving the problem that the current discrimination of flight bumpy areas overly relies on multiple sets of good data. It can effectively utilize the contour algorithm of machine vision to realize the judgment of the aircraft contour, and can accurately judge the bumpy areas through intermittent data, with universal applicability. Brief Description of the Drawings

[0049] To more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0050] Figure 1 It is a schematic flowchart of a method for discriminating bumpy areas during low-altitude flight based on machine vision provided by an embodiment of the present invention;

[0051] Figure 2 It is a structural diagram of a separately marked area provided by an embodiment of the present invention. Detailed implementation manners

[0052] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0053] As Figure 1 shown, the discrimination of bumpy areas during low-altitude flight based on machine vision provided by an embodiment of the present invention mainly includes:

[0054] Step 1: Displacement sensors set based on the preset positions of the general aviation aircraft to be discriminated collect displacement data of the general aviation aircraft to be discriminated during actual low-altitude flight, so as to obtain the flight data of the general aviation aircraft to be discriminated and determine the flight state of the general aviation aircraft to be discriminated at each flight moment;

[0055] Step 2: Obtain the model parameters of the general aviation aircraft to be discriminated from the general aviation aircraft database to establish an aircraft model, and perform low-altitude flight simulation according to the corresponding specified flight mode;

[0056] Step 3: Obtain a set of flight images of the aircraft model during the complete simulation process according to the simulation results, and use the contour algorithm of machine vision to obtain the flight contour parameters of each flight image in the set of flight images;

[0057] Step 4: Determine the acquisition moment of each flight image in the set of flight images to match the flight state at the corresponding flight moment, and combine the flight contour parameters at the corresponding acquisition moment to determine the bumpy area of the general aviation aircraft to be discriminated during low-altitude flight.

[0058] In this embodiment, the general aviation aircraft to be identified is a general aviation aircraft, and the preset position of the general aviation aircraft to be identified is determined by the outer contour of the aircraft and the turbulence characteristics of the aircraft, and is pre-set.

[0059] In this embodiment, the displacement data includes all data recorded by each displacement sensor during the complete actual flight process.

[0060] In this embodiment, the flight data refers to the flight speed, flight acceleration, flight angle, and other data of the aircraft reflected by all data recorded by each displacement sensor.

[0061] In this embodiment, the flight state refers to the operating state of the aircraft, such as accelerating forward, decelerating forward, moving forward at a constant speed, accelerating upward, and the like.

[0062] In this embodiment, the general aviation aircraft database includes all types of aircraft, as well as information such as model parameters corresponding to each aircraft.

[0063] In this embodiment, the aircraft model is established in a virtual environment, and the environmental parameters of the virtual environment are consistent with the actual operating environment of the aircraft.

[0064] In this embodiment, the prescribed flight mode refers to a flight mode corresponding to the flight state of the aircraft during actual flight, and is a flight mode that can be used for low-altitude flight. Flight simulation refers to controlling the aircraft model to fly according to the prescribed flight mode.

[0065] In this embodiment, the simulation results include simulated flight video data of the aircraft model during the complete simulation process. By capturing images of the video data at corresponding time intervals, a flight image set of the aircraft model during the complete simulation process can be obtained.

[0066] In this embodiment, the flight profile parameter refers to parameter information related to the profile of the aircraft contained in the flight image.

[0067] In this embodiment, the turbulence area of the aircraft refers to an area where the aircraft experiences a certain degree of turbulence during actual flight, and is related to the flight area of the aircraft and the degree of turbulence occurring in the corresponding flight area.

[0068] The beneficial effects of the above technical solution are as follows: By setting displacement sensors at preset positions to collect displacement data and determining the flight state of the general aviation aircraft during low-altitude flight at each flight moment, it is ensured that the displacement data can comprehensively reflect the flight state of the aircraft. By establishing an aircraft model and conducting flight simulations, a flight image set is obtained from the simulation results. Using the contour algorithm of machine vision to judge the contour of the aircraft at different moments, and then matching it with the corresponding flight state to obtain the bumpy area. It can accurately determine the bumpy area of the aircraft during low-altitude flight through discontinuous data, and has general applicability.

[0069] An embodiment of the present invention provides a discrimination method for bumpy areas during low-altitude flight based on machine vision, which determines the flight state of the to-be-discriminated general aviation aircraft at each flight moment, including:

[0070] Collect the displacement data of the to-be-discriminated general aviation aircraft according to the set acquisition time of each displacement sensor;

[0071] According to the relative position relationship of all displacement sensors and the relative time relationship of the set acquisition time, sort and analyze all the collected data to determine the flight state of the to-be-discriminated general aviation aircraft at the corresponding flight moment.

[0072] In this embodiment, the set acquisition time of each displacement sensor is the same, and the setting of the specific acquisition time is related to the type of the aircraft.

[0073] In this embodiment, the sorting and analysis refers to the process of arranging the collected data in the order of the sensor numbers and the sequence of the acquisition moments.

[0074] The beneficial effects of the above technical solution are as follows: Each displacement sensor collects the displacement data of the to-be-discriminated general aviation aircraft according to the set acquisition time, and sorts and analyzes the collected data according to the relative position relationship and the relative time relationship, ensuring the orderliness of the data, further determining the flight state of the to-be-discriminated general aviation aircraft at the corresponding flight moment, and ensuring the accuracy of the determined flight state.

[0075] An embodiment of the present invention provides a discrimination method for bumpy areas during low-altitude flight based on machine vision. According to the relative position relationship of all displacement sensors and the relative time relationship of the set acquisition time, sort and analyze all the collected data, including:

[0076] Randomly select a displacement sensor as the first sensor, determine the relative position relationship between each remaining sensor and the first sensor, and sort each remaining sensor according to the distance from near to far to obtain the numbers of each displacement sensor;

[0077] Arrange the acquisition data at the same acquisition moment in the order of the corresponding displacement sensors to obtain the acquisition data sequence at the same moment. Based on the time relative relationship of the set acquisition time, obtain the order relationship of different acquisition moments, and sort all the acquisition data sequences at the same moment from front to back according to the moment order relationship to obtain the order of all the acquisition data.

[0078] In this embodiment, the two adjacent displacement sensors closest to each displacement sensor can be determined from the relative position relationship.

[0079] The beneficial effects of the above technical solution are as follows: Randomly selecting a displacement sensor as the first displacement sensor ensures the randomness of data sorting, and sorting the acquisition data at all the same moments according to the time sequence ensures the orderliness of the acquisition data, further ensuring the accuracy of the flight state.

[0080] The embodiment of the present invention provides a method for discriminating bumpy areas in low-altitude flight based on machine vision. Obtain the model parameters of the general aviation aircraft to be discriminated from the general aviation aircraft database, including:

[0081] Obtain all the mechanical information of the general aviation aircraft to be discriminated;

[0082] Match each mechanical information with the flight database respectively;

[0083] Based on all the matching results, obtain the model parameters of the aircraft to be identified.

[0084] In this embodiment, the mechanical information includes the specific information of all components of the general aviation aircraft to be discriminated, such as the airfoil and wingspan of the wing and the information of the materials used, etc.

[0085] In this embodiment, the flight database contains all the mechanical information (i.e., structural components) of each part of the general aviation aircraft. By matching the mechanical information of each part, the specific model parameters of the aircraft can be obtained, and the model parameters are used to establish an aircraft model.

[0086] The beneficial effects of the above technical solution are as follows: By obtaining all the mechanical information of the general aviation aircraft to be discriminated and matching each mechanical information with the flight database to obtain the matching results, and then determining the model parameters of the aircraft to be identified through the matching results, the matching of the aircraft model parameters with the general aviation aircraft to be discriminated is ensured, and the consistency between the aircraft model and the general aviation aircraft to be discriminated is further ensured.

[0087] The embodiment of the present invention provides a method for discriminating bumpy areas in low-altitude flight based on machine vision. Establish an aircraft model and conduct low-altitude flight simulation according to the corresponding specified flight mode, including:

[0088] Obtain a parameter comparison table required for establishing an aircraft model, where the parameter comparison table is a blank comparison table containing several parameter descriptions;

[0089] Fill all model parameters into the blank comparison table respectively to establish an aircraft model;

[0090] Compare the flight state of the general aviation aircraft to be discriminated at each flight moment with the flight state at the previous flight moment to obtain a comparison result;

[0091] Based on the comparison result, determine all flight moments when the flight mode changes, and determine the change situation of the flight mode;

[0092] Based on the change situation, determine the specified flight mode of the general aviation aircraft to be discriminated during actual flight, and based on the flight mode, control the aircraft model to perform low-altitude flight simulation.

[0093] In this embodiment, the parameter comparison table includes all parameter information required for establishing an aircraft model.

[0094] In this embodiment, the comparison result includes the specific change situation of all flight states and all time nodes when the flight state changes.

[0095] In this embodiment, the specified flight mode refers to the flight mode adopted by the general aviation aircraft to be discriminated during actual low-altitude flight, including the flight state at each time stage, the change nodes of each flight state, and the flight environment information of the general aviation aircraft to be discriminated, etc.

[0096] The beneficial effects of the above technical solution are: By establishing a parameter comparison table and filling the model parameters into the comparison table to establish an aircraft model, the consistency between the aircraft model and the general aviation aircraft to be discriminated is ensured. By comparing the flight state of the general aviation aircraft to be discriminated at each flight moment with the flight state at the previous moment, all change moments and specific change situations of all flight states can be obtained, and the flight simulation of the aircraft model can be accurately controlled.

[0097] The embodiment of the present invention provides a method for discriminating low-altitude flight bumpy areas based on machine vision. According to the simulation result, obtain the flight image set of the aircraft model during the complete simulation process, and use the contour algorithm of machine vision to obtain the flight contour parameters of each flight image in the flight image set, including:

[0098] Obtain the simulated flight video data of the aircraft model during the complete simulation process, and obtain the acquisition time interval of the displacement data and the initial flight simulation moment corresponding to the initial flight moment;

[0099] Intercept images from the simulated flight video data based on the acquisition time interval and the initial flight simulation time to obtain a set of flight images of the aircraft model during the complete simulation process;

[0100] Preprocess each flight image in the set of flight images to obtain a grayscale image of each flight image;

[0101] Determine the grayscale value of each pixel point in the individual grayscale image, classify all pixel points in the individual grayscale image, and determine a set of contour pixel points;

[0102] Place the individual grayscale image in a preset coordinate system, determine the position coordinates of each pixel point in the set of contour pixel points, and in combination with the principle of similar grayscale values, determine the positional relationship of the position coordinates presented by the preset coordinate system, and further determine multiple contour line segments and the length of each contour line segment;

[0103] According to the principle from left to right of the preset coordinate system, obtain the first distance between the two starting endpoints of every two adjacent contour line segments, the second distance between the two ending endpoints, the third distance between the two closest endpoints, and the included angle between the two adjacent contour line segments, and in combination with the grayscale values of the two adjacent contour line segments, estimate the clarity of the area where the two adjacent contour line segments are located;

[0104]

[0105]

[0106]

[0107]

[0108] Represents the first distance corresponding to two adjacent contour line segments; Represents the second distance corresponding to two adjacent contour line segments; Represents the third distance corresponding to two adjacent contour line segments; Represents the minimum value symbol; Represents the minimum distance among all distances corresponding to two adjacent contour line segments; Represents the included angle formed by two adjacent contour line segments; Represents the angle relationship function corresponding to two adjacent contour line segments; Represents the estimated clarity of the area where two adjacent contour line segments are located; Represents the grayscale value after pixel normalization for two adjacent contour line segments; Represents corresponding to two adjacent contour line segments And The corresponding clarity correction function; A conversion function representing the conversion between grayscale values and clarity; Represents the initial grayscale value corresponding to two adjacent contour line segments; Represents the grayscale value of the i1-th pixel point in the first line segment among two adjacent contour line segments, and the value range of i1 is [1, n01]; Represents the grayscale value of the i2-th pixel point in the second line segment among two adjacent contour line segments, and the value range of i2 is [1, n02]; Represents the minimum grayscale value in the first line segment; Represents the maximum grayscale value in the first line segment; Represents the minimum grayscale value in the second line segment; Represents the maximum grayscale value in the second line segment;

[0109] Perform region annotation on all regions, and determine the final clarity of each individual annotation unit in the grayscale image according to the annotation results;

[0110] Adjust the unit brightness of the grayscale image according to the final clarity of each individual annotation unit, and determine the flight contour parameters according to the adjusted image.

[0111] In this embodiment, image capture is a process of determining the flight image of the aircraft model corresponding to the displacement sensor information acquisition time on the general aviation aircraft to be discriminated according to the acquisition time interval and the initial flight simulation time. Each image in the flight image set corresponds to the displacement data collected by the displacement sensor of the general aviation aircraft to be discriminated.

[0112] In this embodiment, the contour pixel point set contains all pixel points whose grayscale values meet the requirements of becoming a contour.

[0113] In this embodiment, the region where it is located generally refers to the region outlined by a rectangular box corresponding to two adjacent contour lines. Region annotation refers to the process of annotating all regions according to clarity.

[0114] In this embodiment, when two adjacent contour line segments have an intersection point, the angle corresponding to the intersection point is the included angle. When there is no intersection point, the line segments are extended until there is an intersection point. When they are parallel lines, the included angle is regarded as 0°.

[0115] In this embodiment, there are line segment 01 and line segment 02. The left endpoint of line segment 01 is a1, the right endpoint of line segment 01 is a2, the left endpoint of line segment 02 is b1, and the right endpoint of line segment 02 is b2. At this time, the distance between a1 and b1 is the first distance, and the distance between a2 and b2 is the second distance.

[0116] In this embodiment, there are area u1 and area u2 where the object is located. There is an overlapping area u3 between area 01 and area 02 where the object is located. At this time, 001, 002, and 003 are separate marked areas.

[0117] The beneficial effects of the above technical solution are as follows: By obtaining the simulated flight video data during the complete simulation process and intercepting images from the simulated flight video data according to the acquisition time to obtain the image set during the complete simulation process, it is ensured that each image in the image set is consistent with the acquisition data of the displacement sensor during the actual flight process. By determining the gray value of each pixel point in the gray image to determine the final clarity of each separate marked unit in the gray image, and further determining the flight profile parameters, it is ensured that each profile parameter corresponds one-to-one with the actual acquisition data.

[0118] The embodiment of the present invention provides a method for discriminating bumpy areas during low-altitude flight based on machine vision. By combining the flight profile parameters at the corresponding acquisition time, it determines the bumpy areas of the general aviation aircraft to be discriminated during low-altitude flight, including:

[0119] Based on the model parameters of the general aviation aircraft to be discriminated, determine the total contour line length of the general aviation aircraft to be discriminated from the parameter-length database, and based on the flight profile parameters at each acquisition time, determine the clear contour line segment length and the position of the clear contour line segment of the general aviation aircraft to be discriminated;

[0120] Determine the ratio of the clear contour line segment length at each acquisition time to the total contour line length of the general aviation aircraft to be discriminated. At the same time, determine the position comparison between the corresponding clear contour line segment position and the total contour;

[0121] Based on the ratio result and the position comparison result, determine the bumpy state at each acquisition time, combine the flight state at each acquisition time to determine the bumpy level under each different flight state, and calibrate the displacement points during the actual flight process based on the bumpy level to determine the bumpy areas of the general aviation aircraft to be discriminated during low-altitude flight.

[0122] In this embodiment, the parameter-length database contains the total contour line lengths of the aircraft corresponding to the model parameters of all aircraft.

[0123] In this embodiment, the bumpy state refers to the bumpy situation and degree of the general aviation aircraft to be discriminated during low-altitude flight at the corresponding acquisition time.

[0124] In this embodiment, by combining the bumpy state with the flight state, the total bumpy degree under each flight state can be determined, and the bumpy level can be determined according to different total bumpy degrees.

[0125] In this embodiment, the bumpy area is an area where the aircraft experiences bumps related to the flight state, which is determined by calibrating the bump level with the displacement points during the actual flight and through comparison.

[0126] The beneficial effects of the above technical solution are as follows: The total contour line length is determined through model parameters, and the proportion of the total contour line length of the general aviation aircraft to be discriminated is further determined by determining the flight contour parameters at each acquisition moment. The bump state at each acquisition moment is determined and combined with the flight state and the positional relationship between the clear contour line and the total contour line to determine the bump level, ensuring the accuracy of the bumpy area of the aircraft.

[0127] The embodiment of the present invention provides a method for discriminating the bumpy area of low-altitude flight based on machine vision. Based on the flight contour parameters at each acquisition moment, the length and position of the clear contour line segment of the general aviation aircraft to be discriminated are determined, including:

[0128] All sub-contour line segments of the general aviation aircraft to be discriminated are determined according to the flight contour parameters, and the position of each point on each sub-contour line segment is locked in sequence;

[0129] Based on the total number of points with locked positions and combined with the point length of each locked position point, the corresponding clear contour line segment length is obtained.

[0130] The beneficial effects of the above technical solution are as follows: All sub-contour line segments of the general aviation aircraft to be discriminated are determined through the flight contour parameters, and the position of each point on each sub-contour line segment is locked, which provides convenience for subsequent position comparison. Then, based on the total number of points with locked positions and the point length of each locked position point, the length of the clear contour line segment is obtained, ensuring the accuracy of the total length of the clear contour line segment.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for discriminating bumpy areas in low-altitude flight based on machine vision, characterized in that, Including: Step 1: The displacement sensors set based on the preset position of the general aviation aircraft to be discriminated collect the displacement data of the general aviation aircraft to be discriminated during the actual low-altitude flight process, so as to obtain the flight data of the general aviation aircraft to be discriminated, and determine the flight state of the general aviation aircraft to be discriminated at each flight moment; Step 2: Obtain the model parameters of the general aviation aircraft to be discriminated from the general aviation aircraft database to establish an aircraft model, and perform low-altitude flight simulation according to the corresponding specified flight mode; Step 3: Obtain the flight image set of the aircraft model during the complete simulation process according to the simulation results, and use the contour algorithm of machine vision to obtain the flight contour parameters of each flight image in the flight image set; Step 4: Determine the acquisition moment of each flight image in the flight image set to match the flight state at the corresponding flight moment, and combine the flight contour parameters at the corresponding acquisition moment to determine the bumpy area of the general aviation aircraft to be discriminated during low-altitude flight; Among them, determining the flight state of the general aviation aircraft to be discriminated at each flight moment includes: Collect the displacement data of the general aviation aircraft to be discriminated according to the set acquisition time of each displacement sensor; Sort and analyze all the collected data according to the relative position relationship of all displacement sensors and the time relative relationship of the set acquisition time, and determine the flight state of the general aviation aircraft to be discriminated at the corresponding flight moment; Among them, sorting and analyzing all the collected data according to the relative position relationship of all displacement sensors and the time relative relationship of the set acquisition time includes: Randomly select a displacement sensor as the first sensor, determine the relative position relationship between each remaining sensor and the first sensor, and sort each remaining sensor according to the distance to obtain the number of each displacement sensor; Arrange the collected data at the same acquisition moment in the order of the corresponding displacement sensors to obtain the collected data sequence at the same moment, obtain the order relationship of different acquisition moments based on the time relative relationship of the set acquisition time, and sort the collected data sequences at each same moment from front to back according to the moment order relationship to obtain the order of all the collected data; Among them, obtaining the model parameters of the general aviation aircraft to be discriminated from the general aviation aircraft database includes: Obtain all the mechanical information of the general aviation aircraft to be discriminated; Match each mechanical information with the flight database respectively; Obtain the model parameters of the aircraft to be recognized based on all the matching results; Among them, establishing an aircraft model and performing low-altitude flight simulation according to the corresponding specified flight mode includes: Obtain the parameter comparison table required for establishing the aircraft model, where the parameter comparison table is a blank comparison table containing several parameter descriptions; Fill all the model parameters into the blank comparison table to establish an aircraft model; Compare the flight state of the general aviation aircraft to be discriminated at each flight moment with the flight state at the previous flight moment to obtain a comparison result; Determine all flight times when the flight mode changes based on the comparison result, and determine the change situation of the flight mode; Determine the specified flight mode of the general aviation aircraft to be discriminated during actual flight based on the change situation, and control the aircraft model to perform low-altitude flight simulation based on the flight mode; Among them, obtain the flight image set of the aircraft model during the complete simulation process according to the simulation result, and use the contour algorithm of machine vision to obtain the flight contour parameters of each flight image in the flight image set, including: Obtain the simulated flight video data of the aircraft model during the complete simulation process, and obtain the acquisition time interval of the displacement data and the initial flight simulation time corresponding to the initial flight time; Based on the acquisition time interval and the initial flight simulation time, perform image interception on the simulated flight video data to obtain the flight image set of the aircraft model during the complete simulation process; Preprocess each flight image in the flight image set to obtain the grayscale image of each flight image; Determine the grayscale value of each pixel point in the individual grayscale image, and classify all pixel points in the individual grayscale image to determine the contour pixel point set; Place the individual grayscale image in a preset coordinate system, determine the position coordinates of each pixel point in the contour pixel point set, and combine the principle of similar grayscale values to determine the positional relationship of the position coordinates presented by the preset coordinate system, and then determine multiple contour line segments and the length of each contour line segment; According to the principle from left to right of the preset coordinate system, obtain the first distance between the two start points of every two adjacent contour line segments, the second distance between the two end points, the third distance between the two closest end points, and the included angle between the two adjacent contour line segments, and combine the grayscale values of the two adjacent contour line segments to estimate the clarity of the area where the two adjacent contour line segments are located; Among them, represents the first distance under two adjacent contour line segments; represents the second distance under two adjacent contour line segments; represents the third distance under two adjacent contour line segments; represents the minimum value symbol; represents the minimum distance among all distances under two adjacent contour line segments; represents the included angle formed by two adjacent contour line segments; represents the angular relationship function under two adjacent contour line segments; represents the clarity of the area where two adjacent contour line segments are located as estimated; represents the gray value after pixel normalization for two adjacent contour line segments; represents and the corresponding clarity correction function; represents the conversion function between gray value and clarity; represents the initial gray value under two adjacent contour line segments; represents the gray value of the i1-th pixel point in the first line segment under two adjacent contour line segments, and the value range of i1 is [1, n01]; represents the gray value of the i2-th pixel point in the second line segment under two adjacent contour line segments, and the value range of i2 is [1, n02]; represents the minimum gray value in the first line segment; represents the maximum gray value in the first line segment; represents the minimum gray value in the second line segment; represents the maximum gray value in the second line segment; Perform area annotation on all areas where they are located, and determine the final clarity of each individual annotation unit in the grayscale image according to the annotation result; Adjust the unit clarity of the grayscale image according to the final clarity of each individual annotation unit, and determine the flight contour parameters according to the adjusted image; Among them, in combination with the flight contour parameters at the corresponding acquisition time, determine the bumpy area of the general aviation aircraft to be discriminated during low-altitude flight, including: Based on the model parameters of the general aviation aircraft to be discriminated, determine the total contour line length of the general aviation aircraft to be discriminated from the parameter-length database, and based on the flight contour parameters at each acquisition time, determine the clear contour line segment length and the position of the clear contour line segment of the general aviation aircraft to be discriminated; Determine the ratio of the clear contour line segment length at each acquisition time to the total contour line length of the general aviation aircraft to be discriminated. At the same time, determine the position comparison between the corresponding clear contour line segment position and the total contour; Determine the bump state at each acquisition moment based on the proportion result and the position comparison result, determine the bump level under each different flight state in combination with the flight state at each acquisition moment, and calibrate the displacement points during the actual flight process based on the bump level to determine the bump area of the general aviation aircraft to be discriminated during low-altitude flight; Among them, determining the clear contour line segment length and the clear contour line segment position of the general aviation aircraft to be discriminated includes: Determine all sub-contour line segments of the general aviation aircraft to be discriminated according to the flight contour parameters, and lock the positions of each point on each sub-contour line segment in turn; Obtain the corresponding clear contour line segment length according to the total number of points with locked positions and in combination with the point length of each locked position point.

Images