Airport navigation aid light detection method and system

Abstract

The invention discloses an airport navigation-aid light detection method and system, and the method comprises the following steps: inputting the position coordinates of a navigation-aid lamp into a data processing module, and generating the twinborn information of the lamp in the data processing module; a high-speed camera loaded with a positioning module is installed on a carrier, and in the process that the carrier moves along the center line of a runway, the high-speed camera performs continuous imaging, obtains continuous video images of light spots and sends the video images to a data processing module; the data processing module acquires illuminance values of light spots of a plurality of columns of lamps which sequentially pass through the same channel at different positions by utilizing image recognition and position comparison; and generating a luminous intensity detection report of the lamp. According to the method, the visible light high-speed camera moves according to the preset line, all the aid-to-navigation lamps in the same lane where the high-speed camera passes are continuously imaged, the light intensity and the light distribution of multiple columns of aid-to-navigation lamps in the same lane can be rapidly detected in real time, and the detection efficiency is high.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photoelectric detection, and particularly relates to an airport navigation light detection method and system. BACKGROUND

[0002] The airport navigation light system, as the core facility of the visual navigation project of the airport, plays a crucial role in air transportation. A perfect navigation light system can not only provide a clear outline of the runway, taxiway and approach path for the aircraft, helping the pilot to accurately determine the position and direction of the aircraft under complex weather conditions, but also can reduce the risk of accidents, improve the efficiency of airport operation and reduce the waiting time of the aircraft on the ground. However, with the continuous expansion of the airport scale and the increase of the flight density, the normal operation and maintenance of the airport navigation light system also face many challenges, especially the monitoring of the light-emitting state of the navigation light.

[0003] At present, the monitoring technology of the airport navigation light still has many deficiencies. The traditional monitoring method mainly relies on manual inspection, which is not only low in efficiency, but also limited by time and environment. Manual inspection can usually be carried out only after night flight, and it is difficult to monitor the state of the navigation light in real time. In addition, manual inspection also has safety risks, especially during the operation of the airport, which is easy to conflict with the operation of the aircraft. With the development of sensor technology, some airports have begun to use automatic monitoring technology, such as sensor installation, power carrier communication and wireless sensor network. However, these technologies also have obvious defects. For example, the sensor installation cost is high, and it needs to be physically modified to the navigation light, which is complex to construct. At the same time, the sensor is easy to be affected by bad weather and complex environment in outdoor environment, resulting in monitoring circuit failure or misjudgment. In recent years, using fixed installation of high-definition camera combined with image processing and analysis technology can also automatically identify the light-emitting state of the navigation light. Although this method does not need to physically modify the navigation light, the actual use effect is general.

[0004] In order to overcome the deficiencies of the above monitoring methods, the special photometric vehicle for the airport navigation light is a more reasonable solution. The special photometric vehicle is a mobile device integrating multiple detection technologies, which can detect the light-emitting state of the navigation light on the runway and taxiway of the airport. Its main advantage is that it can realize rapid deployment and flexible operation, and at the same time avoid many limitations of the traditional detection methods.

[0005] However, the general photometric vehicle adopts an illuminance sensor to detect the light-emitting state of the light, although its detection accuracy is high, but in each detection process, the photometric vehicle can only detect a column of longitudinally arranged lamps, and cannot simultaneously detect the comprehensive detection of the multiple columns of lamps such as centerline lamps and side lamps contained in the same lane, thereby resulting in poor effect of improving the detection efficiency. The lamps in the area where the vehicle cannot pass are helpless, and it is difficult to realize comprehensive coverage detection.

[0006] To overcome the limitations of the above monitoring methods, this invention proposes a navigation light monitoring system based on a high-speed camera and image recognition algorithm. Summary of the Invention

[0007] The purpose of this invention is to propose a method for real-time on-site detection of the light intensity and light distribution of airport navigation lights. Under appropriate conditions, it can quickly and continuously detect all navigation lights in the entire airport to ensure the safe operation of the airport.

[0008] To achieve the above objectives, the present invention is specifically implemented through the following technical solutions:

[0009] On one hand, the present invention provides a method for detecting airport navigation lights, comprising the following steps:

[0010] S1. Input the location coordinates of the navigation lights into the data processing module, and generate twin information of the lights in the data processing module, including number, type and location;

[0011] S2. Install a high-speed camera with a positioning module on the vehicle. As the vehicle moves along the center line of the runway, the high-speed camera continuously images and acquires continuous video images of each light point, and sends the video images to the data processing module.

[0012] S3. The data processing module uses image recognition and position comparison to obtain the illuminance value E of the light points of multiple rows of lights passing through the same lane at different positions.

[0013] S4. Based on the illuminance value, position, and light coefficient F of each light point, the system automatically generates a luminous intensity detection report for the light fixture, including the minimum luminous intensity at a specific point of each beam and the average luminous intensity value of the main beam.

[0014] Furthermore, the vehicle is a drone or a ground vehicle; the high-speed camera is a single-lens camera, an array camera, a time-of-flight camera, or a binocular camera.

[0015] Furthermore, in step S3, the data processing module uses image recognition and position comparison methods as follows: based on the position map, the real-time position of the vehicle and the video signal, it uses AI algorithms to perform multimodal matching of light spots and lamps, automatically calculates the lamp corresponding to a single light spot, and obtains the real-time three-dimensional relative position of the high-speed camera and the lamp.

[0016] Further, in step S3, the data processing module utilizes image recognition and position comparison methods, including matching pixels in the video image with the coordinates of the lamp positions within the selected range; the specific process of coordinate matching is: converting the world coordinates of the lamps... pass Transformed into camera coordinates, then processed through a projection matrix. Convert coordinates to image coordinates Then, the image coordinates are transformed using a quadratic transformation matrix S. The coordinates of the pixel Using image drawing functions, the coordinate information of the lamps can be added as text labels to images or videos.

[0017] Furthermore, before performing the coordinate matching, the image is denoised and filtered, the image is segmented using an edge detection algorithm, the maximum contour of the light source image is extracted, and then the center pixel is calculated according to the moment of the contour, thereby obtaining the pixel position coordinates corresponding to the light spot.

[0018] Furthermore, the method for selecting the locations of the lighting fixtures involved in the matching process:

[0019] With the camera's numerical aperture as 2α, the camera's mounting height as H, the camera's position denoted by C, and the distance L between the camera and each light fixture:

[0020]

[0021] in, This refers to the sequential numbering of the lights along the direction the camera is traveling. This refers to the number of the lamps in the direction perpendicular to the direction of travel;

[0022] By continuously acquiring the camera's position coordinates and The motion direction vector of the camera can be obtained as follows:

[0023] ;

[0024] For any given light fixture, its direction vector relative to the camera is:

[0025] ;

[0026] Therefore, the angle β between the direction of motion of the lamp and the camera can be calculated as follows:

[0027] ;

[0028] in, ,

[0029]

[0030] By comparison, if It is greater than The luminaire can be imaged by the camera, and the position coordinates of the luminaire are selected to match the light spot; otherwise, it cannot be imaged and will not participate in the matching.

[0031] Further, in step S4, the calculation method of the light coefficient F is as follows: select a certain number of lamps of a certain type, measure their actual average light intensity value of the main beam, use them as positive samples to train the calculation model as necessary, and compare the obtained measurement values ​​to calculate the coefficient F.

[0032] Furthermore, the high-speed camera is deviated from the centerline by a certain distance to obtain the light coefficient F of different rows of lights; in actual measurements, the high-speed camera travels along different routes to obtain complete light intensity and light distribution data.

[0033] Further, in step S4, the calculation method for the luminous intensity is as follows: a high-speed camera moves along the runway centerline from far to near, capturing video images from above a certain light fixture's light point. The brightness and positional changes of the light fixture's light point in the video images are continuously collected to obtain a set of luminous intensity and positional distribution data for the light fixture. Given the relative positions of the high-speed camera and the light fixture in the horizontal (x, y) and vertical (z) directions, R is calculated using simple trigonometric functions based on the relative position data. The luminous intensity I of the point light source is then calculated using the formula I=E*R. 2 The luminous intensity I of the light spot of the lamp can be obtained by .

[0034] On the other hand, the present invention provides an airport navigation lighting detection system, comprising:

[0035] A high-speed camera is mounted on the top front center of the vehicle to capture the light emission signals of the airport navigation lights. By capturing the light emission signals of the navigation lights in real time, it generates continuous video images for subsequent light intensity distribution and colorimetric monitoring.

[0036] The positioning module is used to acquire the position, altitude, and orientation information of the monitoring device on the airport flight area pavement in real time, providing spatial reference for matching the light spot with the lamp;

[0037] The data processing module is used to receive, process, and analyze data from the high-speed camera and the positioning module; it includes a lamp position coordinate input module, a light spot and lamp matching module, a lamp light intensity and light distribution monitoring module, and a data fusion and result output module.

[0038] This invention utilizes a high-speed visible light camera that moves along a preset route to continuously image all navigation lights within the same lane. Using a built-in AI algorithm, it simulates the perspective of an aircraft pilot to continuously detect the light intensity and distribution of the navigation lights. The device, suspended from a drone, can inspect all lights on the entire field, including approach lights installed in areas inaccessible by vehicles. It enables rapid, real-time detection of the light intensity and distribution of multiple rows of navigation lights within the same lane, with a detection efficiency far exceeding that of current mainstream navigation light inspection vehicles that can only test one row of lights at a time. The detection results closely reflect the actual experience of an aircraft pilot. Mounted on a drone, it can also detect approach lights that are inaccessible to other conventional inspection methods and have a greater impact on safety during landing.

[0039] This invention continuously monitors the light intensity distribution and chromaticity of all navigation lights within the area traversed by the system from a simulated driver's perspective, enabling real-time, rapid, and efficient assessment of the operational status of the navigation lights. The advantages of this invention are also reflected in the following aspects:

[0040] (1) The core of this invention is to use a high-speed camera combined with image recognition, positioning and matching algorithms to achieve accurate monitoring of navigation lights. The high-speed camera can capture the illumination state of the navigation lights at a high frame rate, ensuring the continuity and real-time nature of the monitoring process. It can obtain key parameters such as the light intensity distribution and chromaticity of the navigation lights and match the corresponding lights according to the positioning module.

[0041] (2) To meet the needs of different application scenarios, the monitoring equipment of the present invention adopts a modular design, which has high flexibility and scalability. The modular design allows the equipment to be quickly assembled and disassembled according to specific needs. It can be installed on vehicles for routine monitoring of airport runways and taxiways, or it can be used as a pod for UAVs to achieve rapid inspection of approach lights and the surrounding area of ​​the airport. This multi-purpose design not only improves the utilization rate of the equipment, but also significantly reduces airport operating costs.

[0042] (3) It has the synergistic advantages of high-speed cameras and image recognition algorithms. The high frame rate of high-speed cameras ensures the continuity of the monitoring process and can capture changes in light intensity that may occur in navigation lights within a short period of time. Combined with image recognition algorithms, the system can automatically analyze the image data captured by the camera, quickly identify abnormal states of the lights, and provide real-time feedback on the monitoring results. The introduction of image recognition algorithms not only improves monitoring efficiency but also reduces reliance on manual intervention and minimizes misjudgments caused by human factors.

[0043] In summary, this invention provides an efficient, flexible, and reliable solution for airport navigation light monitoring through high-speed cameras, image recognition algorithms, and a modular device design. The flexibility of the modular device and its modular design for various application scenarios bring significant advantages to this invention. When installed on a vehicle, the device can flexibly schedule monitoring tasks during airport operations. When used as a UAV pod, the device can quickly cover various areas of the airport, performing particularly well in monitoring complex terrain or remote areas inaccessible by vehicles. This flexibility allows the invention to adapt to the monitoring needs of different airport scenarios, providing comprehensive support for airport operations. Attached Figure Description

[0044] Figure 1 This is a flowchart of the light detection method of the present invention;

[0045] Figure 2 This is a schematic diagram illustrating the pixel-to-lamp position coordinate matching process in the method of the present invention.

[0046] Figure 3 This is a schematic diagram illustrating the selection of the imaging range of the lamps in the method of the present invention;

[0047] Figure 4 This is a schematic diagram of the light emission imaging of the lamp in the method of the present invention;

[0048] Figure 5 This is a schematic diagram of the spatial light intensity distribution of the lamps in the method of the present invention.

[0049] Figure 6 This refers to Figure A.4 in the "Technical Requirements for Navigation Lighting for Runways and Taxiways";

[0050] Figure 7 This refers to Figure A.7 in the "Technical Requirements for Navigation Lighting for Runways and Taxiways";

[0051] Figure 8 This refers to Figure A.9 in the "Technical Requirements for Navigation Lighting for Runways and Taxiways";

[0052] Figure 9 This is a schematic diagram showing the position of the high-speed camera when the vehicle of this invention is a ground vehicle;

[0053] Figure 10 This is a schematic diagram showing the position of the high-speed camera when the carrier of this invention is an unmanned aerial vehicle (UAV). Detailed Implementation

[0054] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0055] like Figures 1 to 10 As shown, an airport navigation light detection method of the present invention includes the following steps:

[0056] S1. Input the location coordinates of the navigation lights into the data processing module, and generate twin information of the lights in the data processing module, including number, type and location;

[0057] S2. Install a high-speed camera with a positioning module on the vehicle. As the vehicle moves along the center line of the runway, the high-speed camera continuously images and acquires continuous video images of each light point, and sends the video images to the data processing module.

[0058] S3. The data processing module uses image recognition and position comparison to obtain the illuminance value E of the light points of multiple rows of lights passing through the same lane at different positions.

[0059] S4. Based on the illuminance value, position, and light coefficient F of each light point, generate a luminous intensity test report for the luminaire, including the minimum luminous intensity at a specific point of each beam and the average luminous intensity value of the main beam.

[0060] In some specific embodiments, the vehicle is a drone or a ground vehicle; the high-speed camera is a single-lens camera, an array camera, a time-of-flight camera, or a binocular camera.

[0061] The system involved in this invention includes a high-speed camera, a vehicle, a positioning module, and a data processing module. The high-speed camera is mounted at an appropriate position on the vehicle, which can be a vehicle or a drone. The positioning module uses satellite navigation + inertial navigation + RTK differential positioning to achieve centimeter-level accuracy in positioning the high-speed camera. The data processing module includes data acquisition and data processing. First, the position coordinates of the navigation lights are entered into the data processing module, which generates twin information for the lights, including their number, type, and location. As the vehicle moves along the centerline, video images from the high-speed camera are sent to the data processing module. The data processing module uses image recognition and position comparison to obtain the illuminance values ​​of multiple rows of lights passing sequentially within the same lane at different positions, including centerline lights, side lights, grounding zone lights, and fast exit taxiway indicator lights.

[0062] When using drones as vehicles to approach and land along the glide path of manned aircraft, the illuminance values ​​of various approach lights, such as centerline lights, crossbeam lights, side lights, runway threshold lights, wingplate lights, and sequential flashing lights, can be obtained at different locations. Vehicles are generally not allowed to pass through the areas where airport approach lights are located; this method of using drones as vehicles effectively solves this problem.

[0063] In one embodiment of the present invention, a visible light high-speed camera is installed at the top front center of the vehicle. When the vehicle travels along the center line of the flight path, the high-speed camera can continuously image all the navigation lights on the same path. In the video, a single navigation light is a light spot that moves continuously and changes in brightness continuously.

[0064] The vehicle is equipped with a satellite navigation + inertial navigation + differential positioning system, enabling it to pinpoint its real-time location to the centimeter level. It also features a data acquisition and processing system. Based on a pre-loaded data processing module showing the installation locations of airport navigation lights, the system uses AI algorithms such as YOLO to perform multimodal matching of light spots and lights, automatically calculating the light fixture corresponding to a single light spot and obtaining the real-time three-dimensional relative position between the high-speed camera and the light fixture.

[0065] In some embodiments, in step S3, the data processing module uses image recognition and position comparison methods as follows: based on the position map, the real-time position of the vehicle and the video signal, it uses AI algorithms to perform multimodal matching of light spots and lamps, automatically calculates the lamp corresponding to a single light spot, and obtains the real-time three-dimensional relative position of the high-speed camera and the lamp.

[0066] In some preferred embodiments, in step S3, the data processing module utilizes image recognition and position comparison methods, including matching the coordinates of pixels in the video image with the coordinates of lamp positions within a selected range; the specific process of coordinate matching is as follows: Figure 2 As shown, this represents the world coordinates of the lighting fixture. pass Transformed into camera coordinates, then processed through a projection matrix. Convert coordinates to image coordinates Then, the image coordinates are transformed using a quadratic transformation matrix S. The coordinates of the pixel Using image drawing functions, the coordinate information of the lamps can be added as text labels to images or videos.

[0067] In this embodiment, to achieve a correspondence between the image light spot position and the actual lamp position, the camera's intrinsic and extrinsic parameters are obtained through camera calibration. Next, images within the field of view are acquired using a high-speed camera. Machine vision algorithms are then used to perform noise reduction, segmentation, and edge detection on the images, and the position of the light spot's emission center within a pixel is extracted. Then, the coordinates of the lamp directly in front of the monitoring device are selected based on positioning and orientation parameters synchronized with the camera. Finally, through coordinate transformation, the pixel positions of the light spots are compared, and the algorithm is used to annotate the lamp coordinates in the acquired video, thus achieving a match between the image light spot and the lamp position.

[0068] In other embodiments, before performing the coordinate matching, the image is denoised and filtered, the image is segmented using an edge detection algorithm, the maximum contour of the light source image is extracted, and then the center pixel is calculated according to the moments of the contour to obtain the pixel position coordinates corresponding to the light spot.

[0069] In this embodiment, considering that the light emission image is a bright spot, occupying a large number of pixels in the image and affecting the light spot matching effect, the image is first processed in grayscale and then subjected to noise reduction filtering before coordinate matching is performed. An edge detection algorithm is used to segment the image, and then the maximum contour of the light emission image is extracted. The center pixel is then calculated based on the contour's moments, thus obtaining the pixel position coordinates corresponding to the light spot. These coordinates are then clearly marked. Although the types of light fixtures differ, the center of the imaged light spot can be identified and marked using the above method. When monitoring the luminous chromaticity value of the light fixture is required, grayscale processing is not performed on the image; noise reduction filtering is applied directly.

[0070] To ensure the accuracy of lamp labeling and improve labeling and matching efficiency, the camera's positioning, height, and orientation information are used to first calculate the distance and viewing angle of the coordinates appearing within the camera's field of view. Combined with the monitor's travel speed, the area where the lamp coordinates appear in the pixel coordinate system is divided. From... Figure 3 As shown, when the camera takes pictures on the pavement, the area that can be imaged is limited by the camera's numerical aperture (2α) and the camera's mounting height (H). Therefore, when selecting the coordinate range of the lamps, it is necessary to determine whether the lamps fall within the numerical aperture range. The position of the camera is represented by C, and the distance between the camera and each lamp can be obtained.

[0071] Therefore, in some embodiments, the method for selecting the positions of the lamps involved in the matching is as follows:

[0072] With the camera's numerical aperture as 2α, the camera's mounting height as H, the camera's position denoted by C, and the distance L between the camera and each light fixture:

[0073]

[0074] in, This refers to the sequential numbering of the lights along the direction the camera is traveling. This refers to the number of the lamps in the direction perpendicular to the direction of travel; this formula can be used to determine the distance between the lamps and the camera.

[0075] At the same time, by continuously collecting the camera's position coordinates and The motion direction vector of the camera can be obtained as follows:

[0076] ;

[0077] For any given light fixture, its direction vector relative to the camera is:

[0078] ;

[0079] Therefore, the angle β between the direction of motion of the lamp and the camera can be calculated as follows:

[0080] ;

[0081] in, ,

[0082]

[0083] By comparison, if Greater than The luminaire can be imaged by the camera, and the position coordinates of the luminaire are selected to match the light spot; otherwise, it cannot be imaged and will not participate in the matching.

[0084] In some embodiments, in step S4, the method for calculating the light coefficient F is as follows: select a certain number of lamps of a certain type, measure their actual average light intensity value of the main beam, use them as positive samples to train the calculation model as necessary, and compare the obtained measurement value to calculate the light coefficient F.

[0085] In the embodiments, it is also preferable that the high-speed camera deviates from the centerline by a certain distance to obtain the light coefficient F of different rows of lights; in actual measurement, the high-speed camera travels along different routes to obtain complete light intensity and light distribution data.

[0086] In some embodiments, in step S4, the method for calculating the luminous intensity is as follows: a high-speed camera moves from far to near along the centerline of the runway, passing above a certain light fixture and capturing video images. The brightness and positional changes of the light fixture in the video images are continuously collected to obtain a set of luminous intensity and positional distribution data for the light fixture. The relative positions of the high-speed camera and the light fixture in the horizontal (x, y) and vertical (z) directions are known, and R is calculated using simple trigonometric functions based on the relative position data.

[0087] According to the formula I=E*R, the luminous intensity I of a point light source is related to the illuminance E. 2 The luminous intensity I of the light spot of the lamp can be obtained by .

[0088] The basic monitoring principle is to obtain the mapping relationship between light intensity and pixel value through pre-static calibration, and to draw the light intensity distribution curve of the lamp by using the movement trajectory of the light spot and the brightness change, so as to realize the judgment of the lamp's luminous state.

[0089] The camera images the light emitted by the lamps, such as... Figure 4 As shown, the light emitted by lighting fixtures typically spreads out in the form of spherical waves, but airport lighting fixtures usually have a certain directionality, emitting light only within a specific spatial angle range. Let's assume a distance... At that location, the luminous intensity is Corresponding illuminance value That is, the inverse square law of illuminance, where... For camera lens normal and The angle between directions. Usually Very small, Therefore, we can obtain .

[0090] Along the monitoring direction, for a specific light fixture, the distance The angle between the light-emitting part of the camera and the ground changes from long to short. The illuminance gradually increases, therefore the outline of the imaged light spot also gradually increases. By utilizing the mapping relationship between calibrated illuminance and pixel values, and by integrating the pixels within the imaged light spot area, the corresponding value can be obtained. The illuminance value under the given conditions. Therefore, by continuously sampling and calculating, the illuminance value can be obtained. Distribution of varying light emission patterns from different lighting fixtures.

[0091] As a vehicle approaches from a distance along the runway centerline and passes over a particular light fixture or spot, the continuous acquisition of information on the brightness and position changes of that spot in the video represents the system collecting a set of luminous intensity and light distribution data for that light fixture. Based on the approximate formula for the relationship between luminous intensity and illuminance of a point light source: I=E*R 2 Given the real-time horizontal (x, y) and vertical (z) distances between the high-speed camera and the lighting fixture, the distance R is calculated using simple trigonometric functions, and E is obtained from the output of the high-speed camera. Thus, the following is derived: Figure 6 The data shown is a set of light intensity and light distribution data, that is, the values ​​of each point on the middle vertical line. The middle vertical line light corresponds to a vertical line, while the light that deviates from the middle line corresponds to a diagonal line with a certain small angle (either left or right, depending on the installation position).

[0092] Light fixtures emit light within a spatial angle range; therefore, depending on the spatial angle, the projection of the luminous intensity in the vertical plane facing the direction of light emission is as follows: Figure 5 As shown in the figure, a camera moving along the centerline of the road and directly facing the lamps measures the light intensity distribution of the lamps as it moves, as indicated by ▲ in the figure. This distribution is a line in a vertical plane. By comparing this value with a reference value (which needs to be calibrated according to the specific type and illumination state of the lamps), the quality of the lamps' illumination state can be determined. Lamps within the camera's field of view are not necessarily directly in front of the camera; therefore, the measured intensity distribution will vary slightly depending on the angle of the lamps. However, the measured intensity distribution value can be corrected based on the positional coordinates of the camera and the lamps, thus allowing for the determination of the lamps' illumination state.

[0093] For the sideline lights located far from the runway centerline, the measured luminous intensity distribution values ​​fall within... Figure 7 The diagonal line shown corresponds to the actual installation position of the light fixture, whether it is slightly to the left or right.

[0094] On the other hand, the present invention provides an airport navigation lighting detection system, comprising:

[0095] High-speed camera 101, mounted at the top front center of the vehicle, is used to capture the emission signals of airport navigation lights. As the vehicle travels along the pavement centerline, the high-speed camera covers all navigation lights along the pavement centerline and on both sides. Each navigation light appears as a continuously moving point of light with continuously changing brightness in the video. By capturing the emission signals of the navigation lights in real time, continuous video images are generated for subsequent light intensity distribution and colorimetric monitoring. The type of camera can be selected according to specific requirements, including single-lens cameras, array cameras, time-of-flight cameras, or binocular cameras.

[0096] The positioning module 102 is used to acquire the position, altitude, and orientation information of the monitoring device on the airport runway in real time, providing spatial reference for matching the light spot with the lighting fixture. It preferably employs a high-precision positioning device, whose main function is to acquire the precise position information of the monitoring device in real time. Through the high-precision positioning device, supplemented by altitude and compass sensors, it can acquire the position, altitude, and orientation information of the monitoring device on the airport runway in real time, providing spatial reference for matching the light spot with the lighting fixture. It mainly consists of satellite navigation (GNSS), inertial navigation (INS), and differential positioning (RTK) devices. It can be integrated with the camera on a single unit or a single integrated board.

[0097] The data processing module 103 is used to receive, process, and analyze data from the high-speed camera and the positioning module. The data processing module is the core of the entire monitoring scheme. It is responsible for receiving, processing, and analyzing data from the camera and the positioning device. It needs to have a high-performance processor, a large-capacity storage device, and a high-speed data transmission interface. The software of the data processing module is developed based on a real-time operating system to ensure the efficiency and real-time performance of data processing.

[0098] The data processing module 103 is equipped with a lamp position coordinate input module, a light spot and lamp matching module, a lamp light intensity and light distribution monitoring module, and a data fusion and result output module.

[0099] This invention integrates the above three modules into a flexible mountable device, which can be mounted on the front of inspection vehicles or under drones, adapting to the deployment of lighting monitoring solutions in different airport environments.

[0100] In data processing module 103, the luminaire location coordinate input module is used to input the installation location coordinates of airport navigation lights into the data processing module, forming a luminaire layout database. The data includes luminaire type, number, installation direction, coordinate location (longitude, latitude, elevation), and standard luminous intensity value. The input function is implemented through GIS software import or manual input. Its purpose is to provide a retrieval database for the identification and matching of luminaires during inspections. The installation locations of airport navigation lights are determined by the actual distribution of the airport.

[0101] The light spot and lamp matching module works by obtaining the camera's intrinsic and extrinsic parameters through camera calibration; acquiring images within the field of view using a high-speed camera; performing noise reduction, segmentation, and edge detection on the images based on machine vision algorithms; extracting the position of the light spot's emission center within a pixel; filtering the coordinates of the lamp directly in front of the monitoring device based on positioning and orientation parameters synchronized with the camera; and finally, through coordinate transformation, comparing the pixel positions of the light spot, and using algorithms to annotate the lamp's coordinates in the acquired video, thus achieving the matching of the image light spot with the lamp's position.

[0102] The function of the lamp luminous intensity and light distribution monitoring module is to continuously calculate the light spots imaged by the lamp, thereby realizing the monitoring of the luminous intensity distribution of the lamp.

[0103] The data fusion and result output module integrates the camera's light spot data and the light emission monitoring results of the lamps to generate a monitoring report that includes the lamp number, light intensity value, light distribution curve, abnormal lamp location, and suggested maintenance measures.

[0104] According to the relevant technical specifications of civil aviation, the calculation scope of the average light intensity of the main beam (denoted as A value) is as follows: Figure 7 The value is the arithmetic mean of the light intensity data corresponding to all integer coordinate points within the elliptical region shown; however, the calculation method used in this case only averages the light intensity data of each sampling point on the green line trajectory, and the result is denoted as the B value. Due to the fundamental difference in the spatial range of the data sampling, the calculation results of the two methods will inevitably have discrepancies.

[0105] In practical engineering applications, by batch collecting the A-values ​​(standard values) and B-values ​​(simplified measured values) of multiple similar luminaires, a statistical deviation model (such as deviation coefficients and linear regression relationships) can be established. Subsequently, in the data fusion stage, the F-coefficient of the B-value is systematically corrected based on this model, achieving a precise mapping from the B-value to the A-value. This correction logic highly aligns with the visual perception characteristics of aircraft pilots (the comprehensive response law of the human eye to the spatial distribution of light beams), effectively ensuring the accuracy of light intensity parameter representation and consistency with actual visual experience.

[0106] Figures 6 to 8The original images are from Figures A.4 / A.7 / A.9 in the "Technical Requirements for Runway and Taxiway Navigation Lighting" issued by the Civil Aviation Administration of China. Specifically, A.4 is the isointense diagram of the runway centerline lights (white light) and the fast exit taxiway (yellow light) with a longitudinal spacing of 30m; A.7 is the isointense diagram of the runway edge (white light) with a runway width of 45m; and A.9 is the network of points used to calculate the average light intensity of the runway.

[0107] contrast Figure 6 , Figure 7 and Figure 8 The first two images only provide light intensity data for points along the green line, so calculations are not possible. Figure 4 The average value of the main beam required in (A.9).

[0108] During normal takeoff, landing, and taxiing, the pilot cannot see light from all directions emitted by the lights. In fact, what the pilot sees is precisely the light recorded by the high-speed camera in this case—a light that perfectly simulates what the pilot sees.

[0109] Figure 8 The arithmetic mean is used to calculate the average luminous intensity of the main beam. Therefore, for lamps of the same brand, type, and batch, the arithmetic mean is used. Figure 6 and Figure 7 The average light intensity value obtained by the arithmetic mean of the vertical line should have an approximately fixed proportional relationship with the actual average light intensity value of the main beam. In practice, a certain number of lamps can be selected, and their actual average light intensity values ​​of the main beam can be measured. Then, the measured values ​​obtained by applying the method in this example can be compared to calculate the light coefficient F. Different types of lamps have different light coefficient F values. To improve the accuracy and usability of the light coefficient F value, the high-speed camera can be deviated from the centerline by a certain distance to obtain different light coefficient F values. These can be used as positive samples to train the calculation model. In subsequent measurements, the high-speed camera can travel along different routes to obtain more complete light intensity and light distribution data.

[0110] Based on actual test data and corrections to the light coefficient F value, the system automatically generates a test report for the luminaire, including the minimum light intensity at a specific point of each beam and the average light intensity of the main beam.

[0111] After collecting the location and illuminance information of the lamps, the data processing module obtains the data through internal processing. Figure 2 and Figure 3 The light intensity values ​​at each coordinate on the central vertical line are weighted by the light coefficient F value to obtain the average light intensity value and light distribution map of each lamp.

[0112] If a high-speed imaging luminance and colorimeter is used instead of a high-speed camera, colorimetric data can be obtained simultaneously with the luminous intensity data of the lamps. Figure 8The existing calculation of the average value of the main beam shown uses integer coordinate values. The data processing module of this invention can collect denser non-integer coordinate data and use AI to train it to obtain a better light coefficient F value. Different lamps can have different light coefficient F values.

[0113] Preferably, dual high-speed cameras can be installed with a horizontal rotation of 180°, which can acquire bidirectional data of the lamps in a single inspection.

[0114] Preferably, after appropriate conversion during the calculation process of the data processing module, the present invention can accurately detect the ground illuminance value of the area requiring lighting in the airport, and can also accurately detect the brightness and chromaticity values ​​of the signboards and gate signs, without requiring F-value calibration.

[0115] The specific embodiments described in this invention are merely illustrative of the invention and are not intended to limit it. Those skilled in the art can make modifications to these embodiments without contributing any inventive step after reading this specification, but such modifications are protected by patent law as long as they fall within the scope of the claims of this invention.

Claims

1. A method for detecting airport navigation lights, characterized in that, The method includes the following steps: S1. Input the location coordinates of the navigation lights into the data processing module, and generate twin information of the lights in the data processing module, including number, type and location; S2. Install a high-speed camera with a positioning module on the vehicle. As the vehicle moves along the center line of the runway, the high-speed camera continuously images and acquires continuous video images of each light point, and sends the video images to the data processing module. S3. The data processing module uses image recognition and position comparison to obtain the illuminance value E of the light points of multiple rows of lights passing through the same lane at different positions. S4. Based on the illuminance value, position, and light coefficient F of each light point, generate a luminous intensity test report for the luminaire, including the minimum luminous intensity at a specific point of each beam and the average luminous intensity value of the main beam.

2. The airport navigation light detection method according to claim 1, characterized in that, The vehicle is a drone or a ground vehicle; the high-speed camera is a single-lens camera, an array camera, a time-of-flight camera, or a binocular camera.

3. The airport navigation light detection method according to claim 1, characterized in that, In step S3, the data processing module uses image recognition and position comparison methods as follows: based on the position map, the real-time position of the vehicle and the video signal, it uses AI algorithms to perform multimodal matching of light spots and lamps, automatically calculates the lamp corresponding to a single light spot, and obtains the real-time three-dimensional relative position of the high-speed camera and the lamp.

4. The airport navigation light detection method according to claim 1, characterized in that, In step S3, the data processing module utilizes image recognition and position comparison methods, including matching the coordinates of pixels in the video image with the coordinates of lamp positions within a selected range; the specific process of coordinate matching is: converting the world coordinates of the lamps... pass Transformed into camera coordinates, then processed through a projection matrix. Convert coordinates to image coordinates Then, the image coordinates are transformed using a quadratic transformation matrix S. The coordinates of the pixel Using image drawing functions, the coordinate information of the lamps can be added as text labels to images or videos.

5. The airport navigation light detection method according to claim 4, characterized in that, Before performing the coordinate matching, the image is denoised and filtered, the image is segmented using an edge detection algorithm, the maximum contour of the light source image is extracted, and then the center pixel is calculated according to the moment of the contour, thereby obtaining the pixel position coordinates corresponding to the light spot.

6. The airport navigation light detection method according to claim 4, characterized in that, Method for selecting the locations of the lighting fixtures to be matched: With the camera's numerical aperture as 2α, its mounting height as H, and its position denoted by C, the distance L between the camera and each light fixture is: ， in, This refers to the sequential numbering of the lights along the direction the camera is traveling. This refers to the number of the lamps in the direction perpendicular to the direction of travel; By continuously acquiring the camera's position coordinates and The motion direction vector of the camera can be obtained as follows: ； For any given light fixture, its direction vector relative to the camera is: ； Therefore, the angle β between the direction of motion of the lamp and the camera can be calculated as follows: ； in, , ， By comparison, if It is greater than The luminaire can be imaged by the camera, and the position coordinates of the luminaire are selected to match the light spot; otherwise, it cannot be imaged and will not participate in the matching.

7. The airport navigation light detection method according to claim 1, characterized in that, In step S4, the calculation method of the light coefficient F is as follows: select a certain number of lamps of a certain type, measure their actual average light intensity value of the main beam, use them as positive samples to train the calculation model as necessary, and compare the obtained measurement values ​​to calculate the coefficient F.

8. The airport navigation light detection method according to claim 4, characterized in that, A high-speed camera is deviated from the centerline by a certain distance to obtain the light coefficient F of different rows of lights; in actual measurement, the high-speed camera travels along different routes to obtain complete light intensity and light distribution data.

9. The airport navigation lights detection method according to claim 1, characterized in that, In step S4, the calculation method for the luminous intensity is as follows: a high-speed camera passes over a light fixture from far to near along the runway centerline, capturing video images. The camera continuously collects information on the brightness and positional changes of the light fixture in the video images, obtaining a set of luminous intensity and positional distribution data for the light fixture. Given the relative positions of the high-speed camera and the light fixture in the horizontal (x, y) and vertical (z) directions, the distance R is calculated using simple trigonometric functions based on the relative position data. The luminous intensity I of the point light source is then calculated using the formula I=E*R. 2 The luminous intensity I of the light spot of the lamp can be obtained by .

10. An airport navigation light detection system, characterized in that, include: A high-speed camera is mounted on the top front center of the vehicle to capture the light emission signals of the airport navigation lights. By capturing the light emission signals of the navigation lights in real time, it generates continuous video images for subsequent light intensity distribution and colorimetric monitoring. The positioning module is used to acquire the position, altitude, and orientation information of the monitoring device on the airport flight area pavement in real time, providing spatial reference for matching the light spot with the lamp; The data processing module is used to receive, process, and analyze data from the high-speed camera and the positioning module; It is equipped with a lamp position coordinate input module, a light spot and lamp matching module, a lamp light intensity and light distribution monitoring module, and a data fusion and result output module.

Images