Hyperspectral image aerial aircraft inversion method based on wake cloud

Abstract

The invention belongs to the technical field of remote sensing image processing and target inversion, and particularly relates to a hyperspectral image aerial aircraft height and spectral reflectivity inversion method based on wake clouds and shadows of the wake clouds. In order to solve the technical problems of low aerial aircraft height inversion precision and inaccurate spectral reflectivity inversion in the prior art, firstly, wake cloud and shadow detection, matching and geometric modeling are performed on a hyperspectral image, and the aircraft flight height is inverted by using a stable geometric relationship between the wake cloud and the shadow; high-precision height extraction is realized; and then combining flight height information to establish a radiation transmission model for coupling the aerial aircraft and the background environment, and inverting the spectral reflectivity of the model. Through integrated processing of wake cloud detection, shadow matching, height inversion and spectral reflectivity inversion, the accuracy and applicability of aerial aircraft characteristic inversion are significantly improved, and an effective technical support is provided for application of hyperspectral remote sensing in aerial aircraft monitoring and identification.

Description

Technical Field

[0001] This invention belongs to the field of remote sensing image processing and target inversion technology, specifically relating to an inversion method for the altitude and spectral reflectance of an aircraft in the air based on hyperspectral images of contrails and their shadows. Background Technology

[0002] Space-based optical remote sensing provides new technological means for the detection, localization, and identification of aircraft. Currently, research on aircraft detection based on high-resolution optical remote sensing images mainly focuses on target detection using geometric, spectral, or thermal infrared features. However, in medium-resolution hyperspectral images of 10 to 60 m, aircraft typically appear as small targets of 1×1 to 3×3 pixels, making them difficult to identify directly using geometric features due to interference from complex backgrounds and clouds. Identification based on temperature and spectral characteristics requires first solving the problem of altitude inversion. In this regard, methods for aircraft altitude inversion based on thermal infrared remote sensing images have been reported, but their accuracy is typically in the 500 to 1000 m range, and they require high precision in radiative transfer modeling and image calibration. In the field of hyperspectral remote sensing, the lack of effective means for extracting aircraft altitude makes accurate inversion of spectral reflectance difficult, hindering research on aircraft identification based on spectral features.

[0003] Contrails are linear cirrus clouds formed by the condensation of water vapor plumes emitted by jet aircraft and cold air at high altitudes. There is a stable geometric relationship between the shadow cast by the contrail on the underlying surface and the contrail itself. This geometric relationship can be used to infer the flight altitude of aircraft. Currently, contrail detection and cloud shadow detection are mostly studied independently, and there is no integrated method that utilizes the geometric relationship between the two to infer aircraft altitude and, consequently, spectral reflectance. Existing methods also do not fully consider the radiative transfer characteristics of the coupling between the aircraft, the underlying surface, and the atmosphere, leading to significant deviations in the inversion results. Summary of the Invention

[0004] To address the lack of methods for inverting aircraft altitude in hyperspectral remote sensing, which leads to inaccurate spectral reflectance inversion, this invention proposes a method for inverting aircraft altitude and spectral reflectance from hyperspectral images based on contrails and their shadows. First, this invention utilizes the geometric relationships of contrails and their shadows in the image to construct an altitude inversion model, achieving high-precision aircraft flight inversion. Then, combining flight altitude information, a radiative transfer model coupled with the background environment is established to invert the spectral reflectance of the aircraft.

[0005] This invention significantly improves the accuracy and practicality of aircraft characteristic inversion by integrating contrail cloud detection, shadow matching, altitude inversion, and spectral reflectance inversion, providing technical support for the application of hyperspectral remote sensing in aircraft monitoring and identification.

[0006] A method for inverting aircraft altitude and spectral reflectance from hyperspectral images of contrails, characterized by the following steps:

[0007] Step 1: Acquire hyperspectral remote sensing radiance images and extract imaging time, geographical location, and observation geometric parameters;

[0008] Step 2: Perform coarse extraction of contrail clouds from the hyperspectral image;

[0009] Step 3: Perform coarse extraction of contrail shadows from the hyperspectral image;

[0010] Step 4: Perform morphological filtering on the contrails and their shadows to remove nonlinear interference;

[0011] Step 5: Perform edge detection and Hough transform line detection on the contrail and shadow regions, extract the longest line and match it;

[0012] Step 6: Based on the geometric model of the contrails and their shadows, the height of the contrails is inverted using the distance between the contrails and their shadows in the image, spatial resolution, solar zenith angle, observed zenith angle, solar azimuth angle, and observed azimuth angle. The calculation formula is as follows:

[0013]

[0014] in, The height of the contrail. The distance between the contrail shadow and the contrail projection. Let be the angle between the x-axis and this vector in the Hough transform. The zenith angle of the sun. To observe the zenith angle, The azimuth of the sun. To observe the azimuth angle;

[0015] Step 7: Determine the superpixel containing the aircraft based on the location of the contrail cloud, and calculate the area ratio of the aircraft in the superpixel.

[0016] Step 8: Based on the radiative transfer model coupling the aircraft and the background, the aircraft spectral reflectance is inverted using sensor entrance pupil radiance, path radiance, aircraft area ratio, background radiance, atmospheric transmittance, solar irradiance, and sky radiance. The inversion formula is as follows:

[0017]

[0018] in, Indicates the reflectivity of an aircraft in the air. Indicates the percentage of the aircraft's area. This represents the entrance pupil radiance, including aerial aircraft and background signals. This represents the path path radiation from the aircraft's altitude to the top of the atmosphere. This represents the path path radiation from the top of the atmosphere to the aircraft's altitude. The value representing the background entrance pupil radiance is the average of the four neighboring pixels containing the aircraft superpixel. This indicates the radiance of solar radiation at the top of the atmosphere. This indicates the atmospheric transmittance from the top of the atmosphere to the altitude of the aircraft. This indicates the atmospheric transmittance from the aircraft's altitude to the top of the atmosphere.

[0019] Step 9: Output the aircraft's flight altitude and spectral reflectance data.

[0020] The beneficial effects achieved by this invention are as follows:

[0021] (1) By utilizing the stable geometric relationship between contrails and their shadows, high-precision inversion of aircraft flight can be achieved, with an altitude inversion accuracy of up to 100 m for wide-body passenger aircraft.

[0022] (2) By combining flight altitude information, a radiation transfer model suitable for aerial targets is constructed, which significantly improves the accuracy of spectral reflectance inversion;

[0023] (3) The method is applicable to various backgrounds and cloud interference scenarios, and has strong applicability and stability; Attached Figure Description

[0024] Figure 1 This is a flowchart of the method;

[0025] Figure 2 This is an image of the imaging geometric model of a contrail and its shadow.

[0026] Figure 3 A diagram of the radiation transfer model of an aircraft in the air;

[0027] Figure 4 The image shows the process and results of contrail shadow detection.

[0028] Figure 5 The image shows the results of the spectral reflectance inversion of aircraft in the air and a comparison chart. Detailed Implementation

[0029] Step 1: Acquire hyperspectral remote sensing radiance images and extract imaging time, geographical location, and observation geometric parameters;

[0030] Step 2: Perform coarse extraction of contrail clouds from the hyperspectral image;

[0031] Step 3: Perform coarse extraction of contrail shadows from the hyperspectral image;

[0032] Step 4: Perform morphological filtering on the contrails and their shadows to remove nonlinear interference;

[0033] Step 5: Perform edge detection and Hough transform line detection on the contrail and shadow regions, extract the longest line and match it;

[0034] Step 6: Based on the geometric model of the contrails and their shadows, the height of the contrails is inverted using the distance between the contrails and their shadows in the image, spatial resolution, solar zenith angle, observed zenith angle, solar azimuth angle, and observed azimuth angle. The calculation formula is as follows:

[0035]

[0036] in, The height of the contrail. The distance between the contrail shadow and the contrail projection. Let be the angle between the x-axis and this vector in the Hough transform. The zenith angle of the sun. To observe the zenith angle, The azimuth of the sun. To observe the azimuth angle;

[0037] Step 7: Determine the superpixel containing the aircraft based on the location of the contrail cloud, and calculate the area ratio of the aircraft in the superpixel.

[0038] Step 8: Based on the radiative transfer model coupling the aircraft and the background, the aircraft spectral reflectance is inverted using sensor entrance pupil radiance, path radiance, aircraft area ratio, background radiance, atmospheric transmittance, solar irradiance, and sky radiance. The inversion formula is as follows:

[0039]

[0040] in, Indicates the reflectivity of an aircraft in the air. Indicates the percentage of the aircraft's area. This represents the entrance pupil radiance, including aerial aircraft and background signals. This represents the path path radiation from the aircraft's altitude to the top of the atmosphere. This represents the path path radiation from the top of the atmosphere to the aircraft's altitude. The value representing the background entrance pupil radiance is the average of the four neighboring pixels containing the aircraft superpixel. This indicates the radiance of solar radiation at the top of the atmosphere. This indicates the atmospheric transmittance from the top of the atmosphere to the altitude of the aircraft. This indicates the atmospheric transmittance from the aircraft's altitude to the top of the atmosphere.

[0041] Step 9: Output the aircraft's flight altitude and spectral reflectance data.

[0042] Experiments show that the method of this invention can stably achieve altitude and spectral reflectance inversion of aircraft in various typical scenarios, including land, sea, and coastal zones. Specifically, the altitude inversion accuracy for wide-body passenger aircraft can reach 100 m, significantly outperforming existing methods. Figure 5 It can be seen that the spectral reflectance inversion results obtained by this method are significantly better than those obtained by the classic surface spectral reflectance inversion algorithm software FLAASH, which verifies the effectiveness and reliability of this method and has high engineering practical value.

Claims

1. A method for inverting aircraft altitude and spectral reflectance from hyperspectral images of contrails, characterized in that, Includes the following steps: Step 1: Acquire hyperspectral remote sensing radiance images and extract imaging time, geographical location, and observation geometric parameters; Step 2: Perform coarse extraction of contrail clouds from the hyperspectral image; Step 3: Perform coarse extraction of contrail shadows from the hyperspectral image; Step 4: Perform morphological filtering on the contrails and their shadows to remove nonlinear interference; Step 5: Perform edge detection and Hough transform line detection on the contrail and shadow regions, extract the longest line and match it; Step 6: Based on the geometric model of the contrails and their shadows, the height of the contrails is inverted using the distance between the contrails and their shadows in the image, spatial resolution, solar zenith angle, observed zenith angle, solar azimuth angle, and observed azimuth angle. Step 7: Determine the superpixel containing the aircraft based on the location of the contrail cloud, and calculate the area ratio of the aircraft in the superpixel. Step 8: Based on the radiative transfer model of the aircraft and the background in the air, the spectral reflectance of the aircraft is inverted by the sensor entrance pupil radiance, path radiation, aircraft area ratio, background radiance, atmospheric transmittance, solar irradiance, and sky radiance. Step 9: Output the aircraft's flight altitude and spectral reflectance data.

2. The method as described in claim 1, characterized in that, The coarse extraction method for contrail clouds in the hyperspectral image described in step 2 is to select the water vapor absorption edge band as the feature band and use the maximum inter-class variance method to extract contrail cloud pixels.

3. The method as described in claim 1, characterized in that, The coarse extraction method for contrail shadows in hyperspectral images described in step 3 involves converting the image to the HSV color space, calculating the normalized saturation-brightness difference index, and performing threshold segmentation to extract shadow pixels.

4. The method as described in claim 1, characterized in that, The morphological filtering described in step 4 includes flooding, opening, and closing operations.

5. The method as described in claim 1, characterized in that, The Hough transform line detection described in step 5 specifically includes performing a Hough transform on the wake cloud and shadow edge images, selecting the line with the highest cumulative count as the detection result, and considering a match as the wake cloud and shadow line whose directions differ by no more than a certain angle.

6. The method as described in claim 1, characterized in that, The formula for calculating the height of the inverted contrail cloud in step 6 is as follows: ； in, The height of the contrail. The distance between the contrail shadow and the contrail projection. Let be the angle between the x-axis and this vector in the Hough transform. The zenith angle of the sun. To observe the zenith angle, The azimuth of the sun. To observe the azimuth angle.

7. The method as described in claim 1, characterized in that, The method for determining the superpixel containing the aircraft in step 7 is as follows: within a certain pixel range along the extension direction of the straight endpoint of the contrail cloud, select several pixels with significantly higher radiance than the background and merge them as the superpixel containing the aircraft.

8. The method as described in claim 1, characterized in that, The spectral reflectance inversion formula described in step 8 is as follows: ； in, Indicates the reflectivity of an aircraft in the air. Indicates the percentage of the aircraft's area. This represents the entrance pupil radiance, including aerial aircraft and background signals. This represents the path path radiation from the aircraft's altitude to the top of the atmosphere. This represents the path path radiation from the top of the atmosphere to the aircraft's altitude. The value representing the background entrance pupil radiance is the average of the four neighboring pixels containing the aircraft superpixel. This indicates the radiance of solar radiation at the top of the atmosphere. This indicates the atmospheric transmittance from the top of the atmosphere to the altitude of the aircraft. It represents the atmospheric transmittance from the aircraft's altitude to the top of the atmosphere.

Images