An airport runway area water and dark ice multi-spectral-infrared thermal imaging fusion identification method

Abstract

The application discloses a kind of airport runway area water and dark ice multispectral-infrared thermal imaging fusion identification method, it is related to airport runway surface operation safety monitoring and photoelectric detection technical field, including to the airport runway surface to be detected is divided into area and establishes partition background benchmark library;Using multispectral scanning head is non-contact arc scanning to target area, obtains echo intensity and forms multispectral spectral feature point set;Using infrared thermal imager synchronously collects infrared thermal image temperature matrix and extracts infrared thermal response characteristic;Multispectral spectral feature point set and infrared thermal response characteristic are unified to same airport runway surface plane coordinate system and generate fusion grid;Based on fusion grid, calculate water confidence and dark ice confidence, and according to discriminant threshold, output water and dark ice identification result.The application can reduce the misjudgment caused by sign and marking line, joint, gum mark and strong reflection area, improve the stability of airport runway area water and dark ice identification.

Description

Technical Field

[0001] This invention relates to the field of airport pavement operation safety monitoring and photoelectric detection technology, and in particular to a multispectral-infrared thermal imaging fusion identification method for airport pavement water and dark ice accumulation. Background Technology

[0002] Airport pavement operation safety monitoring is a crucial link in ensuring the safety of aircraft takeoff, landing, and taxiing. Under complex weather conditions, pavement water accumulation, thin ice, and dark ice can significantly reduce the pavement friction coefficient, posing a serious threat to aircraft operations. In recent years, multispectral detection and infrared thermal imaging technologies have been widely used in the field of pavement condition identification. Multispectral detection, based on the differences in reflectivity of different materials at different wavelengths, can distinguish between dry, water-accumulated, and iced conditions; infrared thermal imaging technology, by acquiring the surface temperature field distribution, identifies potential icing risks by utilizing low-temperature anomalies. However, existing multispectral detection technologies mostly use single-point scanning, making it difficult to generate spatial distribution results; while infrared thermal imaging has a large coverage area, it can only reflect temperature anomalies and cannot directly characterize the spectral differences and film thickness of water films and ice films.

[0003] The main shortcomings of existing technologies lie in the fact that airport pavements are divided into various zones, such as ordinary pavement areas, marking and sign areas, seam areas, adhesive residue areas, and high-reflectivity areas. These different zones exhibit significant differences in pavement materials, surface textures, and reflective properties. If a uniform background threshold is used for discrimination, reflective features such as markings, seam shadows, adhesive residue reflections, or localized high reflections are easily misidentified as water accumulation or icing, leading to a high false positive rate. Furthermore, multispectral detection alone is insufficient to generate a spatially continuous pollution distribution map, and infrared thermal imaging alone cannot distinguish the spectral differences between water films and ice films. Without spatial coordinate unification and feature fusion, it is difficult to achieve stable identification and spatialized output of water accumulation and dark ice. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways, which solves the problems of high misjudgment rate caused by uniform threshold discrimination in complex runway backgrounds in existing technologies, and the lack of spatial uniformity and feature fusion of multispectral and infrared data, which makes it difficult to stably identify and spatially output water accumulation and dark ice.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a multispectral-infrared thermal imaging fusion identification method for water and dark ice on airport pavement, which includes: S1, dividing the airport pavement to be detected into regions, and collecting multispectral reflectance reference data and infrared temperature reference data of each region under dry conditions to form an airport pavement region background reference library. S2. A multi-spectral scanning head is used to perform non-contact scanning of the target area of ​​the airport pavement to obtain the echo intensity, and the echo intensity is corrected and calculated according to the background reference library of the airport pavement partition to form a multi-spectral feature point set. S3. Use an infrared thermal imager to synchronously acquire the infrared thermal image temperature matrix of the target area, and process the infrared thermal image temperature matrix to extract infrared thermal response features. S4. Based on the spatial location parameters of the multi-spectral feature point set and the infrared thermal response feature, unify the two into the same airport pavement plane coordinate system and generate a fused mesh; S5. Based on the feature data within the fused grid, calculate the confidence level of water accumulation and the confidence level of dark ice for each grid cell. S6. Based on the comparison results of the water accumulation confidence and the dark ice confidence with their respective discrimination thresholds, determine the state of the target fused grid and output the water accumulation and dark ice identification results.

[0007] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport pavement water and black ice described in this invention, wherein: S1, the area division includes at least a general pavement area, a marking area, and Joint areas, adhesive residue areas, and highly reflective areas; the airport pavement zoning background reference library is represented as follows: ; In the formula, Indicates the pavement zoning category, Indicates multiple spectral bands. Indicates the first Channel-like partitioning at wavelength The dry reflection reference below, This indicates the drying temperature reference for this zone. Indicates a fixed angle of incidence. Indicates the installation height of the scanning head. Indicates the detection distance.

[0008] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport runway surface water and dark ice described in this invention, wherein: S2, the multispectral scanning head performs circular arc scanning at a preset fixed incident angle and records the scanning angle corresponding to each sampling point; the formation process of the multispectral spectral feature point set includes calculating the normalized reflectance coefficient, wherein the first... Each scan point at wavelength Normalized reflectance coefficient The calculation formula is: In the formula, Indicates the first Each scan point at wavelength The measured echo intensity below, Indicates the first The pavement zone to which each scan point belongs The corresponding dry reflectance reference.

[0009] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport runway surface water and black ice described in this invention, wherein: S2 further includes calculating any two bands and Band ratio between The calculation formula is: ; In the formula, and They represent the first Each scan point at wavelength and The normalized reflectance coefficient is given below.

[0010] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport runway surface water and dark ice described in this invention, wherein: S2 further includes calculating the... Water film thickness at each scanning point The thickness measurement band is selected as 1310nm, and the calculation formula is as follows: ; In the formula, Indicates the first The detection light intensity of the pavement section under dry conditions. Indicates the first Measured echo intensity of each scanning point in the thickness band and These are the model parameters obtained by calibration for this partition under a fixed incident angle and detection distance.

[0011] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport runway surface water and dark ice described in this invention, wherein: in S3, the extraction of the infrared thermal response features includes calculating the relative background temperature difference, for the first... The relative background temperature difference of the area corresponding to each scan point The calculation formula is: ; In the formula, This indicates the local average temperature of the area. This indicates the dry temperature baseline for the zone to which this area belongs.

[0012] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport runway surface water and dark ice described in this invention, wherein: S3 further includes calculating the first... Temperature gradient of the region corresponding to each scanning point The calculation formula is: ; In the formula, and These represent the rate of temperature change of the temperature matrix in two directions within the airport pavement plane coordinate system.

[0013] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport pavement surface water and dark ice described in this invention, in step S4, an airport pavement planar coordinate system is established based on the multispectral scanning angle, detection distance, scanning head installation height, fixed incident angle, and infrared thermal imager field of view parameters. The multispectral scanning point set and the infrared thermal image temperature matrix are mapped to the same coordinate system, and a fusion grid is generated by meshing according to a preset grid size, so that each fusion grid unit is associated with multispectral spectral features, water film thickness features, and infrared thermal response features.

[0014] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport runway surface water and dark ice described in this invention, wherein: in S5, for each fusion grid cell... Spectral absorption features, band ratio features, water film thickness features, low temperature anomaly features, temperature gradient features, thermal boundary continuity features, and adjacent grid continuity features were extracted, and the water accumulation confidence level was calculated using a weighted fusion method. And Dark Ice Confidence .

[0015] The confidence level of water accumulation The calculation formula is: ; In the formula, to These are the weighting coefficients. It is a spectral absorption characteristic. The band ratio characteristic, Characteristics of water film thickness This is a characteristic of low temperature anomalies.

[0016] The confidence level of dark ice The calculation formula is: ; In the formula, to These are the weighting coefficients. The spectral response characteristics of ice film It is a low temperature anomaly characteristic. It is a temperature gradient feature. This is a characteristic of thermal boundary continuity.

[0017] As a preferred embodiment of the multispectral-infrared thermal imaging fusion identification method for airport runway water and dark ice described in this invention, in step S6, when the confidence level of water accumulation in the target fusion grid is greater than the water accumulation discrimination threshold, and the water film spectral response is enhanced, the water film thickness reaches the water accumulation discrimination range, and the infrared temperature field shows a continuous wet and cold response, the fusion grid is determined to be a water accumulation area; when the confidence level of dark ice in the target fusion grid is greater than the dark ice discrimination threshold, and the ice film or refreezing spectral response is met, the film thickness is thin, the reflection boundary is weak or the thermal boundary is indistinct, the infrared temperature is lower than that of the adjacent dry runway surface and the low temperature is abnormally continuous, the fusion grid is determined to be a dark ice area.

[0018] The beneficial effects of this invention are as follows: By establishing a background benchmark library for airport pavement zones, the intensity of multispectral scanning echoes is calculated and corrected for zones. Combined with the temperature matrix synchronously acquired by the infrared thermal imager, the multispectral spectral feature point set and infrared thermal response features are unified into the same airport pavement plane coordinate system to generate a fused grid. Then, based on the spectrum, water film thickness and thermal response features within the grid, the confidence level of water accumulation and black ice is calculated. The state of the target area is determined by comparing the confidence level with the discrimination threshold. This solves the problem of misjudgment caused by a unified threshold under complex pavement backgrounds and achieves high-precision and stable identification and spatialized output of water accumulation and black ice. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart of a multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways provided in an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram illustrating the fusion and recognition of multispectral circular arc scanning point sets and infrared thermal image temperature matrix provided in an embodiment of the present invention.

[0022] Figure 3 A schematic diagram of a laboratory scaled-down experimental platform provided in an embodiment of the present invention.

[0023] Among them, 1 is the airport pavement, 2 is the ordinary pavement area, 3 is the marking and sign area, 4 is the joint area, 5 is the adhesive residue area, 6 is the strong reflection area, 7 is the multi-spectral scanning head, 8 is the circular arc scanning trajectory, 9 is the lateral measurement point sequence, 10 is the infrared thermal imager, 11 is the infrared thermal image temperature matrix, 12 is the fused grid, 13 is the water accumulation area, 14 is the dark ice area, 15 is the laboratory scaled-down test platform, 16 is the synchronous control and data acquisition box, 17 is the low temperature control platform, and 18 is the simulated airport pavement sample plate. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-3 As one embodiment of the present invention, this embodiment provides a multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways, comprising the following steps: like Figure 1 As shown, this embodiment provides a multispectral-infrared thermal imaging fusion identification method for water accumulation and black ice on airport pavements. The method first divides the airport pavement 1 into regions and establishes a regional background reference library; then, it acquires a multispectral circular arc scanning point set of the target region through a multispectral scanning head 7, and simultaneously acquires an infrared thermal image temperature matrix 11 through an infrared thermal imager 10; subsequently, it unifies the multispectral scanning point set and the infrared thermal image temperature matrix 11 to the same airport pavement plane coordinate system, generating a fusion grid 12; finally, based on the spectral characteristics, film thickness characteristics, and thermal response characteristics in the fusion grid 12, it determines the water accumulation region 13 and the black ice region 14, and outputs the spatial distribution results of the airport pavement pollution state.

[0028] In this embodiment, the airport pavement 1 is divided into a general pavement area 2, a marking and sign area 3, a seam area 4, an adhesive residue area 5, and a high-reflectivity area 6. Different zones have different surface materials, texture roughness, reflectivity, and thermal response characteristics. If a uniform background threshold is used for judgment, it is easy to misjudge marking reflections, seam shadows, adhesive residue reflections, or localized high reflections as water accumulation or ice. Therefore, before formal identification, multispectral reflectance reference data and infrared temperature reference data for each zone are collected under dry conditions.

[0029] For the Similar to the path partition, in wavelength Acquire echo intensity under dry conditions Simultaneously, the drying infrared temperature of this zone was collected. Record the scanner head installation height during data acquisition. Detection range and fixed incident angle This creates a background reference library corresponding to pavement zones, optical parameters, and temperature references. This background reference library is used for subsequent zone-based correction of multispectral reflectance intensity and infrared temperature response, so that the reflectance and temperature changes of the measured point can be compared with the dry reference of its region, rather than with a uniform threshold across the entire pavement.

[0030] like Figure 2 As shown, the multispectral scanning head 7 is positioned at a preset fixed incident angle. Non-contact scanning of airport pavement 1 is performed. The fixed incident angle... This is the angle between the optical axis of the multispectral scanning head 7 and the normal to the airport pavement. In a preferred embodiment, The initial angle is 20°, adjustable within the range of 15° to 30°, and locked after calibration. By scanning with a fixed incident angle, fluctuations in reflection intensity caused by changes in the incident angle during the scanning process can be reduced, improving the comparability between different measurement points.

[0031] Multispectral scanning employs at least two near-infrared or short-wave infrared bands located in the sensitive regions of absorption differences between water and ice, preferably 1310 nm and 1550 nm, with 1430 nm potentially added as a further option. Specifically, 1310 nm can be used for water film thickness measurement, 1550 nm can enhance the discrimination of water and ice state differences, and 1430 nm can serve as a supplementary band for feature correction. The multispectral scanning head 7 oscillates along an arc scanning trajectory 8, scanning at each sampling angle... Obtain echo intensity for each band The angle encoder records the scanning angle corresponding to each scanning point. Furthermore, the scanning angle is bound to the echo intensity, sampling time, and spatial location to form a lateral measurement point sequence with spatial order.

[0032] For the For each scan point, the system calls the corresponding dry reflection reference based on the pavement zone q to which that point belongs. The normalized reflectance coefficient and band ratio of the scanned point are calculated for each band. Normalization is used to reduce the background reflection differences between ordinary pavement, road markings, seams, adhesive residue, and highly reflective areas. The band ratio reflects the differences in absorption of water film, ice film, or ice-water mixtures by different bands.

[0033] In a preferred embodiment, the water film thickness is calculated using a dual-wavelength laser water film thickness model. 1310nm is selected as the preferred thickness measurement wavelength band, and the first wavelength is set as... The probe light intensity of the pavement partition in a dry state is , No. The measured echo intensity of each scanning point in the thickness measurement band is: , and Given the model parameters obtained by calibration for this partition under a fixed incident angle and detection distance, the water film thickness at the i-th scanning point is... It can be calculated using the following formula: ; In the formula, Indicates the first Water film thickness at each scanning point This indicates the echo intensity under the measured condition. Indicates the first similar to a road surface The intensity of the dry echo in the zone, and These are model parameters calibrated using a dry state and at least two sets of samples with known water film thickness. In other embodiments, film thickness characteristics can also be obtained using calibration curves, lookup table models, or regression models.

[0034] For ice films, dark ice, or ice-water mixtures, stability cannot be determined solely by the water film thickness. Therefore, this embodiment further combines the relative reflectance at 1550nm, band ratio, infrared low-temperature anomalies, and thermal boundary continuity for joint identification. In other words, the multispectral scanning point set includes not only water film thickness information but also band absorption differences, reflection boundary changes, and candidate features of the film state.

[0035] like Figure 2As shown, the infrared thermal imager 10 is fixedly installed above or to the side of the airport pavement 1, passively observing the multispectral circular arc scanning coverage area. The infrared thermal imager 10 does not emit a beam of light onto the airport pavement 1, but instead acquires an infrared thermal image temperature matrix 11 formed by the self-radiation of the target area. The infrared thermal image temperature matrix 11 is used to reflect the low-temperature anomalies, local temperature differences, temperature gradients, thermal boundaries, and regional continuity of the target area. For suspected black ice areas, if their surface boundaries are not obvious under visible light, but the temperature is lower than that of the adjacent dry pavement and the low-temperature area is continuous, it can serve as an important basis for black ice identification.

[0036] After acquiring the infrared thermal image temperature matrix 11, radiometric correction, noise suppression, and background temperature compensation are performed on it. For the region corresponding to the multispectral measurement point, the local average temperature is extracted, and the relative temperature difference is calculated in conjunction with the drying temperature benchmark of the zone to which the point belongs; at the same time, the temperature gradient and thermal boundary continuity are extracted based on the temperature matrix. Through the above processing, the infrared thermal response characteristics corresponding to the multispectral scanning points are obtained.

[0037] To achieve multi-source data fusion, an airport pavement planar coordinate system is established based on the multi-spectral scanning angle, detection distance, scanning head installation height, fixed incident angle, and the field-of-view parameters of the infrared thermal imager 10. For multi-spectral circular arc scanning points, their positions in the airport pavement planar coordinate system are determined according to the scanning angle and optical path geometric parameters. For pixels in the infrared thermal image temperature matrix 11, they are converted to the same airport pavement planar coordinates using the field-of-view parameters and extrinsic parameter calibration relationship of the infrared thermal imager 10. After the coordinate transformation is completed, the target pavement area is divided according to the preset grid size to generate a fused grid 12.

[0038] Each fusion grid 12 is associated with at least the multispectral normalized reflectance coefficient, band ratio, water film thickness or film candidate features, local temperature, relative temperature difference, temperature gradient, and thermal boundary continuity within the corresponding region. Through the fusion grid 12, point-based multispectral scanning results and area-based infrared temperature results can be fused within the same spatial cell, avoiding misjudgments caused by inconsistencies in the spatial location of single-point spectral data and infrared image data.

[0039] Spectral absorption features were extracted from each fused grid 12. Band ratio characteristics Water film thickness characteristics Low temperature anomaly characteristics Temperature gradient characteristics Thermal boundary continuity characteristics and the continuity features of adjacent grids The system calculates the confidence level of water accumulation through rule-based discrimination or weighted fusion. And Dark Ice Confidence In a preferred embodiment, the confidence scores for water accumulation and dark ice are expressed as follows: ; in, to , to The weighting coefficients are determined through calibration experiments. Specifically, samples in known states—dry, waterlogged, thin ice, dark ice, and ice-water mixtures—are prepared. The spectral absorption characteristics, band ratio characteristics, water film thickness characteristics, low-temperature anomaly characteristics, temperature gradient characteristics, thermal boundary continuity characteristics, and adjacent grid continuity characteristics of each sample are extracted. Using the true state of the samples as labels, the weighting coefficients are determined through least-squares fitting, logistic regression, grid optimization, or methods that maximize classification accuracy.

[0040] When the water accumulation confidence of fused grid 12 If the water accumulation threshold is greater than the water accumulation detection threshold, and the grid simultaneously satisfies the following conditions: enhanced water film spectral response, water film thickness within the water accumulation detection range, and continuous wet and cold response in the infrared temperature field, it is identified as a water accumulation region 13. In a preferred embodiment, preliminary discrimination can be performed by combining the relative reflectance in the 1550nm band and the water film thickness. When the relative reflectance in the 1550nm band is lower than the water accumulation detection threshold and the water film thickness is greater than or equal to the water accumulation thickness threshold, the grid is determined to have water accumulation candidate characteristics; when the relative reflectance in the 1550nm band is lower than the water accumulation detection threshold but the water film thickness is less than the water accumulation thickness threshold, the grid is determined to have wet or thin water film candidate characteristics. The water accumulation detection threshold and water accumulation thickness threshold can be obtained through airport pavement sample calibration.

[0041] When the Dark Ice confidence level of fused grid 12 If the value of the grid exceeds the dark ice discrimination threshold, and the grid satisfies the following conditions: ice film or refreezing spectral response, thin film thickness, weak or indistinct thermal boundary, infrared temperature lower than the adjacent dry road surface, and continuous low temperature anomaly, then the grid is identified as a dark ice region 14. In a preferred embodiment, when the relative reflectance in the 1550nm band is within the ice film candidate range, and the infrared thermal image temperature matrix 11 shows continuous low temperature anomaly, while the reflectance boundary is weaker than the ordinary water accumulation boundary and the thermal boundary shows a continuous low temperature distribution, then the fused grid is identified as a dark ice candidate region.

[0042] When the multispectral characteristics are inconsistent with the infrared thermal response characteristics, the system performs secondary correction by combining the continuity of adjacent fused grids, historical temporal changes, and the regional background benchmark library. For example, if the multispectral reflectance anomaly of a certain grid only appears in marking area 3 or strong reflectance area 6, and the infrared temperature does not show a low temperature anomaly, its dark ice confidence level is reduced; if the infrared temperature of a certain area is consistently lower than that of the adjacent dry area, and the spectral characteristics show an ice film or refreezing response, its dark ice confidence level is increased. Through the above secondary correction, misjudgments caused by pavement background differences and local reflectance anomalies can be reduced.

[0043] like Figure 3 As shown, this embodiment uses a laboratory scaled-down test platform 15 to verify the above method. The simulated airport pavement sample plate 18 has dimensions of 1200mm × 800mm × 50mm, and the surface of the sample plate is equipped with a dry zone, a water film zone, a thin ice zone, a dark ice zone, a marking simulation zone, and a joint simulation zone. A low-temperature control platform 17 is set below the sample plate, with a temperature control range of -10 ℃ to 5 ℃.

[0044] Probe height of multispectral scanning head 7 500mm, fixed incident angle The angle is 20°, the arm radius is 600mm, the sweep range is ±30°, the sampling interval is 5° / point, and each sweep forms 13 lateral measurement points. Multispectral bands are 1310nm, 1550nm, and optional 1430nm. Water film thickness is set to 0.2mm, 0.5mm, 1mm, 3mm, 5mm, and 10mm; ice film thickness is set to 0.2mm, 0.5mm, 1mm, 3mm, and 5mm.

[0045] During the experiment, the synchronous control and data acquisition box 16 controlled the servo motor, angle encoder, multispectral scanning head 7, infrared thermal imager 10, and visible light auxiliary camera to synchronously acquire data, with a synchronization error of no more than 10 ms. After the acquired data was recorded by the host computer, the fusion recognition unit calculated the confidence scores for water accumulation and dark ice, and output the water accumulation and dark ice recognition results corresponding to the lateral measurement point sequence 9.

[0046] The final identification results output by this invention include water accumulation area 13, dark ice area 14, coverage area, maximum film thickness, average film thickness, low temperature anomaly range, and risk level. The identification results can form a spatial distribution map of airport pavement contamination status and can be sent to airport pavement inspection, de-icing operations, or operational support terminals. The specific wavebands, fixed incident angles, sweeping ranges, and test dimensions in the above embodiments are merely preferred embodiments. Equivalent substitutions or modifications made by those skilled in the art without departing from the concept of this invention should fall within the protection scope of this invention.

[0047] In summary, this invention establishes a background benchmark library for airport pavement zones, performs zoned correction calculations on the intensity of multispectral scanning echoes, and combines the temperature matrix synchronously acquired by an infrared thermal imager to unify the multispectral spectral feature point set and infrared thermal response features into the same airport pavement plane coordinate system to generate a fused grid. Then, based on the spectrum, water film thickness, and thermal response features within the grid, the confidence level of water accumulation and black ice is calculated. The state of the target area is determined by comparing the confidence level with the discrimination threshold, thereby solving the problem of misjudgment caused by a unified threshold under complex pavement backgrounds and achieving high-precision and stable identification and spatialized output of water accumulation and black ice.

[0048] 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways, characterized in that: include, S1. Divide the airport pavement to be inspected into regions, and collect multispectral reflectance reference data and infrared temperature reference data for each region under dry conditions to form an airport pavement region background reference library. S2. A multi-spectral scanning head is used to perform non-contact scanning of the target area of ​​the airport pavement to obtain the echo intensity, and the echo intensity is corrected and calculated according to the background reference library of the airport pavement partition to form a multi-spectral feature point set. S3. Use an infrared thermal imager to synchronously acquire the infrared thermal image temperature matrix of the target area, and process the infrared thermal image temperature matrix to extract infrared thermal response features. S4. Based on the spatial location parameters of the multi-spectral feature point set and the infrared thermal response feature, unify the two into the same airport pavement plane coordinate system and generate a fused mesh; S5. Based on the feature data within the fused grid, calculate the confidence level of water accumulation and the confidence level of dark ice for each grid cell. S6. Based on the comparison results of the water accumulation confidence and the dark ice confidence with their respective discrimination thresholds, determine the state of the target fused grid and output the water accumulation and dark ice identification results.

2. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 1, characterized in that: S1, The area division includes at least the ordinary pavement area, the sign and marking area, and the connecting area. The airport pavement zoning background reference library is represented as follows: (The text also mentions areas with adhesive residue and strong reflections, but this is incomplete and lacks context.) ； In the formula, Indicates the pavement zoning category, Indicates multiple spectral bands. Indicates the first Channel-like partitioning at wavelength The dry reflection reference below, This indicates the drying temperature reference for this zone. Indicates a fixed angle of incidence. Indicates the installation height of the scanning head. Indicates the detection distance.

3. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 2, characterized in that: S2. The multi-spectral scanning head performs circular arc scanning at a preset fixed incident angle and records the scanning angle corresponding to each sampling point; the formation process of the multi-spectral feature point set includes calculating the normalized reflectance coefficient, wherein the first... Each scan point at wavelength Normalized reflectance coefficient The calculation formula is: In the formula, Indicates the first Each scan point at wavelength The measured echo intensity below, Indicates the first The pavement zone to which each scan point belongs The corresponding dry reflectance reference.

4. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 3, characterized in that: S2 also includes calculating any two bands and Band ratio between The calculation formula is: ; In the formula, and They represent the first Each scan point at wavelength and The normalized reflectance coefficient is given below.

5. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 4, characterized in that: S2 also includes calculating the first... Water film thickness at each scanning point The thickness measurement band is selected as 1310nm, and the calculation formula is as follows: ; In the formula, Indicates the first The detection light intensity of the pavement section under dry conditions. Indicates the first Measured echo intensity of each scanning point in the thickness band and These are the model parameters obtained by calibration for this partition under a fixed incident angle and detection distance.

6. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 5, characterized in that: In S3, the extraction of the infrared thermal response features includes calculating the relative background temperature difference, for the first... The relative background temperature difference of the area corresponding to each scan point The calculation formula is: ; In the formula, This indicates the local average temperature of the area. This indicates the dry temperature baseline for the zone to which this area belongs.

7. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 6, characterized in that: S3 also includes calculating the first [missing information] based on the infrared thermal image temperature matrix. Temperature gradient of the region corresponding to each scanning point The calculation formula is: ; In the formula, and These represent the rate of temperature change of the temperature matrix in two directions within the airport pavement plane coordinate system.

8. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 7, characterized in that: In S4, an airport pavement plane coordinate system is established based on the multispectral scanning angle, detection distance, scanning head installation height, fixed incident angle, and infrared thermal imager field of view parameters. The multispectral scanning point set and the infrared thermal image temperature matrix are mapped to the same coordinate system, and a fused grid is generated by meshing according to the preset grid size, so that each fused grid unit is associated with multispectral spectral features, water film thickness features, and infrared thermal response features.

9. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 1, characterized in that: In S5, for each merged mesh cell Spectral absorption features, band ratio features, water film thickness features, low temperature anomaly features, temperature gradient features, thermal boundary continuity features, and adjacent grid continuity features were extracted, and the water accumulation confidence level was calculated using a weighted fusion method. And Dark Ice Confidence The confidence level of the accumulated water. The calculation formula is: ; In the formula, to These are the weighting coefficients. It is a spectral absorption characteristic. The band ratio characteristic, Characteristics of water film thickness This is a characteristic of low temperature anomalies; The confidence level of dark ice The calculation formula is: ; In the formula, to These are the weighting coefficients. The spectral response characteristics of ice film It is a low temperature anomaly characteristic. It is a temperature gradient feature. This is a characteristic of thermal boundary continuity.

10. The multispectral-infrared thermal imaging fusion identification method for water accumulation and dark ice on airport runways as described in claim 9, characterized in that: In S6, when the water accumulation confidence of the target fused grid is greater than the water accumulation discrimination threshold, and the water film spectral response is enhanced, the water film thickness reaches the water accumulation discrimination range, and the infrared temperature field shows a continuous wet and cold response, the fused grid is determined to be a water accumulation area. When the confidence level of dark ice in the target fused grid is greater than the dark ice discrimination threshold, and the following conditions are met: ice film or refreezing spectral response, thin film thickness, weak reflection boundary or indistinct thermal boundary, infrared temperature lower than the adjacent dry road surface and low temperature abnormality continuous, the fused grid is determined to be a dark ice region.

Images