Airport navigation aid lamp inspection method based on unmanned aerial vehicle technology

Abstract

The invention relates to an airport aid-to-navigation light inspection method based on an unmanned aerial vehicle technology, and relates to the technical field of aid-to-navigation light inspection, and the method comprises the following steps: S1, setting standard parameters; s2, the unmanned aerial vehicle executes a flight task, and images are collected from multiple angles; s3, obtaining the mass center of the navigation aid lamp; s4, calculating actual geographic coordinates of the navigation aid lamp; s5, comparing the position coordinate information of the navigation aid lamp; s6, brightness attenuation detection of the navigation aid lamp; and S7, carrying out statistical feedback, and carrying out the step S2. The airport navigational light inspection method has the effects of reducing and improving the airport navigational light inspection cost and improving the airport navigational light inspection efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of inspection of navigation lights, and in particular to an inspection method of airport navigation lights based on unmanned aerial vehicle technology. Background Art

[0002] Aids to navigation lights are an important part of airports that provide visual navigation for aircraft. Currently, there are many types and quantities of aids to navigation lights. The runway lighting series is used to mark the runway range so that the aircraft can accurately align with the runway and land safely. The approach lighting series is used to help the aircraft make precise approaches in severe weather conditions and guide pilots to maintain the correct course and altitude when landing. The taxiway lighting series is used to guide the aircraft to taxi on the ground to ensure that the aircraft can safely leave or enter the runway.

[0003] At present, the inspection of airport navigation lights mainly includes three methods: manual inspection, inspection vehicle inspection and remote monitoring. The manual inspection method mainly relies on ground staff to conduct regular inspections, using special tools to detect the brightness, angle and physical damage of the lights. Although manual inspections are highly reliable, they are less efficient. Inspection vehicles mainly rely on vehicles equipped with high-speed cameras and sensors to conduct inspections, and use image recognition technology to determine whether the lights are working properly. Although the efficiency and coverage of inspections have been improved, there are uncontrollable factors such as the angle of high-speed camera acquisition and vehicle speed. Remote monitoring mainly relies on sensors and monitoring systems installed at the airport to collect real-time data on the working status of navigation lights. The remote monitoring system can automatically detect abnormal situations and issue alarms, reducing the need for manual intervention. Although the inspection method of the remote monitoring system is ideal and safe and reliable, the initial investment of the remote monitoring system is large. The reconstruction and installation of remote monitoring systems in old airports will prolong the reconstruction time and affect normal operations. Therefore, remote monitoring inspections are suitable for newly built airports. Summary of the invention

[0004] In order to reduce the cost of airport navigation light inspection and improve the efficiency of airport navigation light inspection, the present application provides an airport navigation light inspection method based on drone technology.

[0005] The present application provides an airport navigation light inspection method based on drone technology, which adopts the following technical solution: comprising the following steps:

[0006] S1. Set standard parameters;

[0007] S2, the drone performs flight missions and collects images from multiple angles;

[0008] S3, obtaining the centroid of the navigation light;

[0009] S4, calculating the actual geographical coordinates of the navigation lights;

[0010] S5. Compare the position coordinate information of the navigation lights;

[0011] S6, navigation light brightness attenuation detection;

[0012] S7: Statistical feedback, and proceed to step S2.

[0013] By adopting the above technical solution, the drone flies according to a pre-set route, and at the same time, the drone takes photos from multiple angles. The photos are uploaded for processing, and the centroid of the navigation light is extracted. Then, the coordinates of the centroid of the navigation light in the image are solved to obtain the actual geographic coordinates of the navigation light. The actual geographic coordinates of the navigation light are traversed and compared with the standard position coordinates of the navigation light. When the actual geographic coordinates of the navigation light exist in the standard position coordinates of the navigation light, the brightness attenuation detection is performed on the navigation light at that position. When the UAV completes the flight mission, the detected navigation lights are not extinguished, and the undetected navigation lights are extinguished. In the captured image, the navigation lights that are completely extinguished have less distinguishing features from the background. Therefore, by identifying the navigation lights that are not extinguished, the navigation lights that are completely extinguished can be screened out, thereby improving the accuracy of detecting faulty navigation lights. UAVs can easily fly to every corner of the runway and are more flexible than vehicles. At the same time, UAVs can automatically collect images according to pre-set flight missions and routes and upload images for analysis, reducing the investment in human resources. UAVs also have a faster flight speed, which improves the efficiency of navigation light inspections.

[0014] Preferably, the step S1 includes: setting a route plan and importing standard position coordinates of the navigation lights.

[0015] By adopting the above technical solution, the flight route of the drone is set according to the distribution of the navigation lights inside the airport, and the longitude and latitude of the navigation lights are entered into the system as the coordinates of the navigation light marking position.

[0016] Preferably, step S3 comprises the following steps:

[0017] S31, image preprocessing: grayscale conversion and filtering of the navigation light position image to obtain a preprocessed image retaining the brightness characteristics of the navigation light;

[0018] S32, image threshold segmentation: binarize the preprocessed image and remove small noise points to obtain a threshold image with a highlight area;

[0019] S33, morphological processing: corroding and then dilating the threshold image to obtain a morphological image that retains the shape and continuity of the navigation light;

[0020] S34, connected domain analysis: screening the connected regions in the morphological image to obtain a target image retaining the real navigation lights;

[0021] S35, centroid extraction: extracting the centroid of the navigation light target in the target image.

[0022] By adopting the above technical scheme, the input navigation light position image is first preprocessed, and the color image is converted into a grayscale image through grayscale conversion to reduce the data dimension. At the same time, Gaussian filtering or median filtering is used to remove noise to obtain a preprocessed image that retains the brightness characteristics of the navigation light. Then, the preprocessed image is processed using an adaptive threshold segmentation method, and the image is binarized according to the brightness difference between the navigation light and the background to obtain a threshold image with a highlight area, so as to facilitate the distinction of the position of the navigation light in the image. Then, the threshold image is processed by corrosion and expansion to remove small noise points in the threshold image and connect the broken brightness areas in the threshold image to obtain a morphological image that retains the shape and continuity of the navigation light. Next, the morphological image is subjected to a connected domain analysis, and the connected areas are identified and marked in the morphological image. Then, the features of each connected area are screened, and the parts that do not meet the characteristics of the navigation light are excluded to obtain a target image that retains the real navigation light. Finally, the centroid of the navigation light target in the target image is extracted.

[0023] Preferably, step S4 comprises the following steps:

[0024] S41, determining the image coordinates of the navigation light target by using the centroid of the navigation light target in the target image;

[0025] S42, converting the image coordinates into camera coordinates using camera parameters;

[0026] S43, converting the camera coordinates into Earth-centered Earth-fixed coordinates using the UAV parameters;

[0027] S44. Convert the geocentric coordinates into actual geographic coordinates.

[0028] By adopting the above technical solution, the coordinates of the center of mass of the navigation light in the image, the flight attitude of the UAV, the positioning coordinates of the UAV and the internal parameters of the camera are used to perform coordinate calculation to obtain the actual geographic coordinates of the navigation light.

[0029] Preferably, the step S42 comprises the following steps:

[0030] S421, obtain the camera focal length parameter (f x ,f y ); Get the camera principal point coordinates (u 0 ,v 0 );

[0031] S422, calculate the camera coordinates (x c ,y c ,z c ):

[0032] Preferably, the step S43 comprises the following steps:

[0033] S431. Get the latitude φ of the drone c ; Get the longitude λ of the drone c ; Get the drone height H c ; Get the drone's roll angle φ; Get the drone's pitch angle θ; Get the drone's yaw angle ψ; Get the earth's ellipsoid's major semi-axis earth ellipsoid a; Get the earth's ellipsoid's minor semi-axis earth ellipsoid b.

[0034] S432, calculating the first eccentricity square e2, Calculate the radius of curvature N of the yoke.

[0035] S433, obtain the UAV's Earth-centered Earth-fixed coordinates E(X e ,Y e ,Z e ),

[0036] Preferably, the step S44 comprises the following steps:

[0037] S441, calculate the roll angle matrix R roll ,

[0038] Calculate the pitch angle matrix R pitch ,

[0039] Calculate the yaw angle matrix R yaw ,

[0040] Calculate the first parameter A,

[0041] Calculate the second parameter B,

[0042] Calculate the third parameter C,

[0043] S442, calculate the first rotation matrix R b2n , R b2n =R yaw ·R pitch ·R roll ;

[0044] Calculate the second rotation matrix R n2e ,

[0045] S443, calculate the total rotation matrix R, R = R n2e ·Rb2n ;

[0046] Calculate the camera direction vector d c ,

[0047] S444. Calculate the geocentric direction vector d ecef , d ecef =R·d c ;

[0048] S445, calculating the first distance parameter t1,

[0049] Calculate the second distance parameter t2,

[0050] S446, if t2<0, the fourth parameter t is the first distance parameter t1, and step S448 is performed; if t2≥0, step S447 is performed;

[0051] S447, if t2≤t1, the fourth parameter t is the first distance parameter t2; if t2>t1, the fourth parameter t is the first distance parameter t1;

[0052] S448, calculate the Earth-centered Earth-fixed geographic coordinates P(X p , Y p , Z p )=E+t·d ecef ;

[0053] Calculate the actual geographic coordinate angle β of the navigation light;

[0054] S449, calculate the actual geographic coordinates (λ p ,φ p ), λ p =arctan2(Y p ,X p ),

[0055] Preferably, step S5 comprises the following steps:

[0056] S51, comparing the actual geographic coordinates with the standard position coordinates of the navigation light;

[0057] S52. If the actual geographic coordinates exist in the standard position coordinates of the navigation light, proceed to step S6; if the actual geographic coordinates do not exist in the standard position coordinates of the navigation light, feedback is given that the navigation light does not exist, and proceed to step S7.

[0058] By adopting the above technical solution, when it is detected that the navigation light at the standard position is not extinguished, the brightness attenuation detection of the navigation light is performed, so as to timely discover the aging problem of the navigation light and shorten the detection time of the faulty navigation light.

[0059] Preferably, step S6 comprises the following steps:

[0060] S61, visible light camera collects visible light images of navigation lights, and thermal infrared camera collects thermal images of navigation lights;

[0061] S62, obtaining the brightness and attenuation rate of the navigation lights through the visible light image and thermal image of the navigation lights;

[0062] S63, when the brightness of the navigation light is lower than the brightness threshold, feedback is given that the navigation light is too dim, and step S7 is performed; when the brightness of the navigation light is not lower than the brightness threshold, step S64 is performed;

[0063] S64. When the attenuation rate of the navigation light is lower than the brightness attenuation threshold, feedback is given that the navigation light is rapidly aging, and step S7 is performed; when the brightness attenuation rate of the navigation light is not lower than the brightness attenuation threshold, feedback is given that the navigation light is normal, and step S7 is performed.

[0064] By adopting the above technical solution, visible light cameras are used to collect visible light images of the navigation lights during the day and at night, and thermal infrared cameras are used to obtain thermal images of the navigation lights, so as to monitor the thermal radiation intensity of the navigation lights. By combining visible light and infrared spectra, multiple dimensions such as the brightness and temperature of the navigation lights are monitored, thereby improving the accuracy of judging the brightness attenuation of the navigation lights.

[0065] In summary, the present application includes at least one of the following beneficial technical effects:

[0066] 1. In the captured images, the completely extinguished navigation lights have little distinguishing features from the background. Therefore, by identifying the navigation lights that are not extinguished, the completely extinguished navigation lights can be screened out, thereby improving the accuracy of detecting faulty navigation lights. UAVs can easily fly to every corner of the runway, which is more flexible than vehicles. At the same time, UAVs can automatically collect images according to pre-set flight missions and routes and upload images for analysis, reducing the investment in human resources. In addition, UAVs have a faster flight speed, which improves the efficiency of navigation light inspections.

[0067] 2. Reduce the data dimension in the image through grayscale conversion and filtering, use the adaptive threshold segmentation method to distinguish the highlight area in the image, and then make the highlight area more obvious through corrosion and expansion. Next, obtain the target image that retains the real navigation light through connected domain analysis, so as to facilitate the extraction of the center of mass of the navigation light target in the image.

[0068] 3. By combining visible light and infrared spectra, the brightness and temperature of navigation lights can be monitored in multiple dimensions to improve the accuracy of the judgment of the brightness attenuation of navigation lights. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a flow chart of an airport navigation light inspection method based on drone technology in an embodiment of the present application.

[0070] Figure 2 It is a flow chart of the brightness attenuation detection process of the navigation light in the embodiment of the present application. DETAILED DESCRIPTION

[0071] The following is combined with Figure 1-2 This application is described in further detail.

[0072] The present application embodiment discloses a method for inspecting airport navigation lights based on drone technology. Figure 1 and Figure 2 , including the following steps:

[0073] S1. Set standard parameters: Set the flight route of the drone according to the distribution of the navigation lights inside the airport, and enter the longitude and latitude of the navigation lights as the coordinates of the navigation light marking position into the system.

[0074] S2. The drone performs flight missions and collects images from multiple angles.

[0075] S3, obtaining the image coordinates of the navigation light. Step S3 includes the following steps.

[0076] S31. Image preprocessing: grayscale conversion is performed on the navigation light position image. The color image is converted into a grayscale image through grayscale conversion to reduce the data dimension. At the same time, a high-pass filter is used or the image is filtered to obtain a preprocessed image that retains the brightness characteristics of the navigation light.

[0077] S32, image threshold segmentation: using an adaptive threshold segmentation method to process the preprocessed image, binarizing the image according to the brightness difference between the navigation light and the background, and obtaining a threshold image with a highlight area;

[0078] S33, morphological processing: corroding and then dilating the threshold image, removing small noise points in the threshold image and connecting the broken brightness areas in the threshold image, so as to obtain a morphological image that retains the shape and continuity of the navigation light;

[0079] S34, connected domain analysis: screening the connected regions in the morphological image, identifying and marking the connected regions in the morphological image, and then screening the features of each connected region, excluding the parts that do not meet the features of the navigation lights, and obtaining the target image that retains the real navigation lights;

[0080] S35, centroid extraction: extracting the centroid of the navigation light target in the target image.

[0081] S4, calculating the actual geographic coordinates of the navigation lights, S4 includes the following steps.

[0082] S41, determining the image coordinates (u, v) of the navigation light target by using the centroid of the navigation light target in the target image;

[0083] S42. Convert the image coordinates into camera coordinates using camera parameters.

[0084] S421, obtain the camera focal length parameter (f x ,f y ); Get the camera principal point coordinates (u 0 ,v 0 );

[0085] S422, calculate the camera coordinates (x c ,y c ,z c ):

[0086] S43. Obtain the Earth-centered Earth-fixed coordinates of the UAV using the UAV parameters.

[0087] S431. Get the latitude φ of the drone c ; Get the longitude λ of the drone c ; Get the drone height H c ; Get the drone's roll angle φ; Get the drone's pitch angle θ; Get the drone's yaw angle ψ; Get the earth's ellipsoid's major semi-axis earth ellipsoid a; Get the earth's ellipsoid's minor semi-axis earth ellipsoid b.

[0088] S432, calculate the first eccentricity square e 2 , Calculate the radius of curvature N of the yoke.

[0089] S433, obtain the UAV's Earth-centered Earth-fixed coordinates E(X e , Y e , Z e ),

[0090] S44. Calculate the actual geographic coordinates of the navigation lights.

[0091] S441, calculate the roll angle matrix R roll ,

[0092] Calculate the pitch angle matrix R pitch ,

[0093] Calculate the yaw angle matrix R yaw ,

[0094] Calculate the first parameter A,

[0095] Calculate the second parameter B,

[0096] Calculate the third parameter C,

[0097] S442, calculate the first rotation matrix R b2n , R b2n =R yaw ·R pitch ·R roll ;

[0098] Calculate the second rotation matrix R n2e ,

[0099] S443, calculate the total rotation matrix R, R = R n2e ·R b2n ;

[0100] Calculate the camera direction vector d c ,

[0101] S444. Calculate the geocentric direction vector d ecef , d ecef =R·d c ;

[0102] S445, calculating the first distance parameter t1,

[0103] Calculate the second distance parameter t2,

[0104] S446, if t2<0, the fourth parameter t is the first distance parameter t1, and step S448 is performed; if t2≥0, step S447 is performed;

[0105] S447. If t2≤t1, the fourth parameter t is the first distance parameter t2; if t2>t1, the fourth parameter t is the first distance parameter t1;

[0106] S448, calculate the Earth-centered Earth-fixed geographic coordinates P(X p , Y p , Z p )=E+t·d ecef ;

[0107] Calculate the actual geographic coordinate angle β of the navigation light;

[0108] S449, calculate the actual geographic coordinates (λ p ,φ p ), λ p =arctan2(Y p ,X p ),

[0109] S5. Compare the navigation light position coordinate information. Step S5 includes the following steps.

[0110] S51, the actual geographic coordinates (λ p ,φ p ) is compared with the standard position coordinates of the navigation lights;

[0111] S52, if the actual geographic coordinates (λ p ,φ p ), then proceed to step S6; if the actual geographic coordinates (λ) do not exist in the standard position coordinates of the navigation light p ,φ p ), then proceed to step S7.

[0112] S6, navigation light brightness attenuation detection, step S6 follows the following steps.

[0113] S61, visible light camera collects visible light images of navigation lights, and thermal infrared camera collects thermal images of navigation lights;

[0114] S62, obtaining the brightness and attenuation rate of the navigation lights through the visible light image and thermal imaging of the navigation lights;

[0115] S63, when the brightness of the navigation light is lower than the brightness threshold, feedback is given that the navigation light is too dim, and step S7 is performed; when the brightness of the navigation light is not lower than the brightness threshold, step S64 is performed;

[0116] S64. When the attenuation rate of the navigation light is lower than the brightness attenuation threshold, feedback is given that the navigation light is rapidly aging, and step S7 is performed; when the brightness attenuation rate of the navigation light is not lower than the brightness attenuation threshold, feedback is given that the navigation light is normal, and step S7 is performed.

[0117] The implementation principle of an airport navigation light inspection method based on drone technology in an embodiment of the present application is as follows: the drone flies according to a pre-set route, and at the same time the drone takes multi-angle photos. The photos taken are uploaded for processing, and the centroid of the navigation light is extracted. After that, the coordinates of the centroid of the navigation light in the image are solved to obtain the actual geographic coordinates of the navigation light. The actual geographic coordinates of the navigation light are traversed and compared with the standard position coordinates of the navigation light. When the actual geographic coordinates of the navigation light exist in the standard position coordinates of the navigation light, the brightness attenuation detection is performed on the navigation light at that position. When the UAV completes the flight mission, the detected navigation lights are not extinguished, and the undetected navigation lights are extinguished. In the captured image, the navigation lights that are completely extinguished have less distinguishing features from the background. Therefore, by identifying the navigation lights that are not extinguished, the navigation lights that are completely extinguished can be screened out, thereby improving the accuracy of detecting faulty navigation lights. UAVs can easily fly to every corner of the runway and are more flexible than vehicles. At the same time, UAVs can automatically collect images according to pre-set flight missions and routes and upload images for analysis, reducing the investment in human resources. UAVs also have a faster flight speed, which improves the efficiency of navigation light inspections.

[0118] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for inspecting airport navigation lights based on drone technology, characterized in that: The following steps are involved: S1. Set standard parameters; S2, the drone performs flight missions and collects images from multiple angles; S3, obtaining the centroid of the navigation light; S4, calculating the actual geographical coordinates of the navigation lights; S5. Compare the position coordinate information of the navigation lights; S6, navigation light brightness attenuation detection; S7: Statistical feedback, and proceed to step S2.

2. The method for inspecting airport navigation lights based on drone technology according to claim 1 is characterized by: The step S1 includes: setting a route plan and importing the standard position coordinates of the navigation lights.

3. The method for inspecting airport navigation lights based on drone technology according to claim 1 is characterized by: The step S3 comprises the following steps: S31, image preprocessing: grayscale conversion and filtering of the navigation light position image to obtain a preprocessed image retaining the brightness characteristics of the navigation light; S32, image threshold segmentation: binarize the preprocessed image and remove small noise points to obtain a threshold image with a highlight area; S33, morphological processing: corroding and then dilating the threshold image to obtain a morphological image that retains the shape and continuity of the navigation light; S34, connected domain analysis: screening the connected regions in the morphological image to obtain a target image retaining the real navigation lights; S35, centroid extraction: extracting the centroid of the navigation light target in the target image.

4. The method for inspecting airport navigation lights based on drone technology according to claim 3 is characterized by: The step S4 comprises the following steps: S41, determining the image coordinates of the navigation light target by using the centroid of the navigation light target in the target image; S42, converting the image coordinates into camera coordinates using camera parameters; S43, converting the camera coordinates into Earth-centered Earth-fixed coordinates using the UAV parameters; S44. Convert the geocentric coordinates into actual geographic coordinates.

5. The method for inspecting airport navigation lights based on drone technology according to claim 4 is characterized by: The step S42 comprises the following steps: S421, obtain the camera focal length parameter (f x ,f y ); Get the camera principal point coordinates (u0, v0); S422, calculate the camera coordinates (x c ,y c ,z c ):

6. The method for inspecting airport navigation lights based on drone technology according to claim 5 is characterized by: The step S43 comprises the following steps: S431. Get the latitude φ of the drone c ; Get the longitude λ of the drone c ; Get the drone height H c ; Get the UAV roll angle φ; Get the UAV pitch angle θ; Get the UAV yaw angle segment; Get the earth ellipsoid major semi-axis earth ellipsoid a; Get the earth ellipsoid minor semi-axis earth ellipsoid b. S432, calculate the first eccentricity square e 2 , Calculate the radius of curvature N of the yoke. S433, obtain the UAV's Earth-centered Earth-fixed coordinates E(X e ,Y e ,Z e ), 7. The method for inspecting airport navigation lights based on drone technology according to claim 6 is characterized by: The step S44 comprises the following steps: S441, calculate the roll angle matrix R roll , Calculate the pitch angle matrix R pitch , Calculate the yaw angle matrix R yaw , Calculate the first parameter A, Calculate the second parameter B, Calculate the third parameter C, S442, calculate the first rotation matrix R b2n , R b2n =R yaw ·R pitch ·R roll ; Calculate the second rotation matrix R n2e , S443, calculate the total rotation matrix R, R = R n2e ·R b2n ; Calculate the camera direction vector d c , S444. Calculate the geocentric direction vector d ecef , d ecef =R·d c ; S445, calculating the first distance parameter t1, Calculate the second distance parameter t2, S446, if t2<0, the fourth parameter t is the first distance parameter t1, and step S448 is performed; if t2≥0, step S447 is performed; S447, if t2≤t1, the fourth parameter t is the first distance parameter t2; if t2>t1, the fourth parameter t is the first distance parameter t1; S448, calculate the Earth-centered Earth-fixed geographic coordinates P(X p ,Y p ,Z p )=E+t·d ecef ; Calculate the actual geographic coordinate angle β of the navigation light; S449, calculate the actual geographic coordinates (λ p ,φ p ), λ p =arctan2(Y p ,X p ), 8. The method for inspecting airport navigation lights based on drone technology according to claim 1, characterized in that: The step S5 comprises the following steps: S51, comparing the actual geographic coordinates with the standard position coordinates of the navigation light; S52. If the actual geographic coordinates exist in the standard position coordinates of the navigation light, proceed to step S6; if the actual geographic coordinates do not exist in the standard position coordinates of the navigation light, feedback is given that the navigation light does not exist, and proceed to step S7.

9. The method for inspecting airport navigation lights based on drone technology according to claim 1, characterized in that: The step S6 comprises the following steps: S61, visible light camera collects visible light images of navigation lights, and thermal infrared camera collects thermal images of navigation lights; S62, obtaining the brightness and attenuation rate of the navigation lights through the visible light image and thermal image of the navigation lights; S63, when the brightness of the navigation light is lower than the brightness threshold, feedback is given that the navigation light is too dim, and step S7 is performed; when the brightness of the navigation light is not lower than the brightness threshold, step S64 is performed; S64. When the attenuation rate of the navigation light is lower than the brightness attenuation threshold, feedback is given that the navigation light is rapidly aging, and step S7 is performed; when the brightness attenuation rate of the navigation light is not lower than the brightness attenuation threshold, feedback is given that the navigation light is normal, and step S7 is performed.

Images