A method and system for detecting the definition of weak feature targets of an aerial camera

Abstract

The present application relates to the technical field of aerial remote sensing definition detection, and particularly relates to a kind of aerial camera weak feature target definition detection method and system;The present application is first calibrated camera parameter, using image processing method obtains the definition detection result of optimal feature area, and corrects strict imaging model, and under the condition that the definition detection result precision of optimal feature target area is higher, weak feature target area definition detection result can be quickly obtained, and it can be suitable for the definition detection of aerial remote sensing camera.

Description

Technical Field

[0001] This invention relates to the field of aerial remote sensing sharpness detection technology, and in particular to a method and system for detecting the sharpness of weak feature targets in aerial cameras. Background Technology

[0002] Aerial cameras typically operate at altitudes of 200m to 20km. Changes in the aircraft's flight altitude and ambient temperature can cause defocusing during aerial imaging, resulting in decreased image clarity and resolution. Therefore, automatic sharpness detection technology is crucial. It typically involves a sharpness detection device that detects the current amount of defocus and feeds it back to the aerial remote sensing camera. The camera then uses a control device to adjust the focal plane position to achieve sharpness detection.

[0003] Traditional sharpness detection methods are inadequate for aerial cameras due to their poor accuracy and scene adaptability, failing to meet the requirements. In contrast, image processing-based sharpness detection methods are gaining increasing popularity due to their high intelligence, strong environmental adaptability, and simple and compact structure.

[0004] Image processing-based automatic sharpness detection methods often use the contrast between the target and the background as a sharpness criterion. That is, ground targets contain more high-frequency information, making them easier to detect using image-based automatic sharpness detection. However, regions such as oceans, deserts, grasslands, and forests, which cover most of the Earth's surface, have small gradient changes in their contour edges and less high-frequency information, making them weak feature targets. In the process of image sharpness detection, weak feature target regions do not satisfy random processes in time and space and cannot be approximated as Gaussian or exponential distributions. Using conventional automatic sharpness detection algorithms is very likely to lead to detection failure. Summary of the Invention

[0005] The main technical problem solved by this invention is to provide a method for detecting the sharpness of weak feature targets in aerial cameras. This method is based on geographic information assistance and solves the problem of large detection errors in the sharpness of weak feature target images. It has the characteristics of accurate detection and simple implementation. This invention also provides a system for detecting the sharpness of weak feature targets in aerial cameras.

[0006] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is: to provide a method for detecting the sharpness of weak feature targets in aerial cameras, comprising the following steps:

[0007] Step S1: Set the parameters of the aerial remote sensing camera;

[0008] Step S2: Establish a rigorous imaging model;

[0009] Step S3: Correct the rigorous imaging model using the Monte Carlo method;

[0010] Step S4: Obtain aerial remote sensing images by scanning ground objects;

[0011] Step S5: Perform image processing on the aerial remote sensing image to obtain its sharpness detection results;

[0012] Step S6: Based on the preset image overlap rate, search for feature points in the overlapping area between two aerial remote sensing images to obtain the feature points in the overlapping area.

[0013] Step S7: Use geographic information to obtain the matching results of two aerial remote sensing images of overlapping area feature points;

[0014] Step S8: Based on the geographic information of feature points in the overlapping area, and using the high-precision attitude information from the first aerial remote sensing image, a corrected rigorous imaging model is introduced to obtain the sharpness detection result of the second aerial remote sensing image.

[0015] As an improvement of the present invention, in step S1, the parameters of the aerial remote sensing camera are set using a precision angle measurement method, including the calibration focal length, principal point coordinates, and distortion coefficient.

[0016] As a further improvement of the present invention, in step S4, the aerial remote sensing camera follows the aircraft and performs single-strip sweeping tilt imaging along the direction perpendicular to the aircraft's flight direction.

[0017] As a further improvement of the present invention, in step S4, during the scanning imaging, the aerial remote sensing camera takes pictures of the ground object, acquires two aerial remote sensing images, and also acquires the overlapping part between the two aerial remote sensing images.

[0018] As a further improvement of the present invention, in step S5, after the first aerial remote sensing image is captured by the aerial remote sensing camera, the sharpness detection result is obtained and recorded using an image processing method.

[0019] As a further improvement of the present invention, in step S7, the high-precision attitude information at the moment of exposure when the first aerial remote sensing image is captured is obtained using the aircraft's airborne orientation and positioning system. Then, the object point coordinates are determined using the geographic information and the corrected rigorous imaging model. After that, the image point coordinates of the corresponding image in the second aerial remote sensing image are calculated using the corrected rigorous imaging model, thereby obtaining the matching result of the two aerial remote sensing images of the overlapping area feature points.

[0020] As a further improvement of the present invention, in step S8, the high-precision attitude information at the moment of exposure when the first aerial remote sensing image is captured is obtained using the aircraft-based airborne orientation and positioning system, and the sharpness detection result of the second aerial remote sensing image is obtained by introducing the corrected rigorous imaging model.

[0021] A system for detecting the sharpness of weak feature targets in an aerial camera includes an aerial remote sensing camera and an optical imaging system installed inside the aerial remote sensing camera; the aerial remote sensing camera is equipped with a cover, a control board, a control motor and an optical lens.

[0022] The beneficial effects of this invention are: compared with the prior art, this invention first calibrates the camera parameters, uses image processing methods to obtain the sharpness detection results of the superior feature region, and corrects the strict imaging model. With the high accuracy of the sharpness detection results of the superior feature target region, it can quickly obtain the sharpness detection results of the weak feature target region, which can be applied to the sharpness detection of aerial remote sensing cameras. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the steps of the aerial camera weak feature target sharpness detection method of the present invention;

[0024] Figure 2 This is a schematic diagram of a rigorous imaging model;

[0025] Figure 3 This is the basic operating mode of an aerial camera;

[0026] Figure 4 This is a schematic diagram of the selection of image points with the same name and the correspondence between the object and the image;

[0027] Figure 5 This is a schematic diagram of the coordinate system for influencing factors. Figure 1 ;

[0028] Figure 6 This is a schematic diagram of the coordinate system for influencing factors. Figure 2 ;

[0029] Figure 7 This is a schematic diagram of the control board installation of the aerial camera weak feature target sharpness detection system of the present invention;

[0030] Figure 8 This is a schematic diagram of changes related to sharpness detection;

[0031] Figure 9 This is a schematic diagram of the implementation process of the present invention;

[0032] Attached reference numerals: 1-box cover, 2-control motor, 3-control board, 4-optical structure, 5-lens. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0034] Please refer to Figures 1 to 9 The present invention provides a method for detecting the sharpness of weak feature targets in an aerial camera, comprising the following steps:

[0035] Step S1: Set the parameters of the aerial remote sensing camera;

[0036] Step S2: Establish a rigorous imaging model;

[0037] Step S3: Correct the rigorous imaging model using the Monte Carlo method;

[0038] Step S4: Obtain aerial remote sensing images by scanning ground objects;

[0039] Step S5: Perform image processing on the aerial remote sensing image to obtain its sharpness detection results;

[0040] Step S6: Based on the preset image overlap rate, search for feature points in the overlapping area between two aerial remote sensing images to obtain the feature points in the overlapping area.

[0041] Step S7: Use geographic information to obtain the matching results of two aerial remote sensing images of overlapping area feature points;

[0042] Step S8: Based on the geographic information of feature points in the overlapping area, and using the high-precision attitude information from the first aerial remote sensing image, a corrected rigorous imaging model is introduced to obtain the sharpness detection result of the second aerial remote sensing image.

[0043] In step S2, a rigorous imaging model is established, such as... Figure 2 As shown, the rigorous imaging model is a rigorous imaging model based on the Earth ellipsoid model established with the aid of DEM, in order to eliminate the influence of ground elevation, that is:

[0044]

[0045] In the formula, (x0, y0) are the coordinates of the principal point on the image, f is the camera principal distance, and together with x0 and y0, they are collectively referred to as the camera interior orientation elements. S ,Y S Z S () represents the coordinates of the photography center in the ground-based photographic coordinate system. To use the Y-axis as the main axis The transformation matrix of the ω, κ rotation system, X S Y S Z S , ω and κ are collectively referred to as camera exterior orientation elements.

[0046] To analyze the impact of various influencing factors on the sharpness detection results, equation (1) is expanded into a series of influencing factors:

[0047] dx=dx1+dx2+dx3+dx4+dx5+…(2)

[0048] dy=dy1+dy2+dy3+dy4+dy5+…(3)

[0049] in:

[0050] dx1=Δx p (4)

[0051] dy1=Δy p (5)

[0052]

[0053]

[0054] dx3=(k1r 2 +k2r 4 (x-x0) (8)

[0055] dy3=(k1r 2 +k2r 4 (y-y0) (9)

[0056] dx4=P1(x-x0)(y-y0)+2P2(r 2 +2(x-x0)2 ) (10)

[0057] dy4=P1(r 2 +2(y-y0) 2 )+2P2(x-x0)(y-y0) (11)

[0058] dx5=sinθ(y-y0) (12)

[0059] dy5=(1-cosθ)(y-y0) (13)

[0060] In the formula, dx and dy are the combined error of the photographic center point, and Δx P Δy P The image point offset error is denoted by Δf, the principal distance error is denoted by x0 and y0, the principal point coordinates are denoted by k1 and k2, the radial distortion coefficients are denoted by P1 and P2, and the eccentric distortion coefficient is denoted by dp. x dp y θ is the pixel size change rate, r is the distance from the image point to the principal image point, θ is the rotation angle of the CCD array in the focal plane, and f is the camera principal distance.

[0061] After analyzing and fitting the effects of each influencing factor, the strict imaging model corresponding to each aerial camera is corrected by equation (1).

[0062] Based on the characteristic that there are overlapping areas between two aerial camera images during operation, it can be known that the same geographic information auxiliary feature point in the overlapping area corresponds to different image points in the two images. It is also known that they satisfy the modified strict imaging model. Therefore, the modified strict imaging model can be used to match the geographic information feature points in the overlapping area between the two images.

[0063] After matching the feature points of the overlapping areas in the two images taken by the aerial camera using the modified rigorous imaging model, the sharpness detection result of the previous image (high feature target) obtained by image processing is used as the initial condition and substituted into equation (1) to solve for the sharpness detection result of the subsequent image (weak feature target).

[0064] In step S1, the parameters of the aerial remote sensing camera are set using a precision angle measurement method, including the calibrated focal length, principal point coordinates, and distortion coefficient.

[0065] In step S4, the aerial remote sensing camera follows the aircraft and performs single-strip sweeping oblique imaging along a direction perpendicular to the aircraft's flight direction. During the sweeping imaging, the aerial remote sensing camera takes pictures of ground objects, acquiring two aerial remote sensing images and also acquiring the overlapping part between the two aerial remote sensing images.

[0066] In step S5, after the first aerial remote sensing image is captured by the aerial remote sensing camera, the sharpness detection result is obtained and recorded using image processing methods.

[0067] In step S7, the high-precision attitude information of the first aerial remote sensing image is obtained at the moment of exposure using the aircraft's airborne orientation and positioning system. Then, the object point coordinates are determined using the geographic information and the corrected rigorous imaging model. After that, the image point coordinates of the second aerial remote sensing image are calculated using the corrected rigorous imaging model, thereby obtaining the matching result of the two aerial remote sensing images of the overlapping area feature points.

[0068] In step S8, the aircraft's airborne orientation and positioning system is used to obtain high-precision attitude information at the moment of exposure when the first aerial remote sensing image is captured. The corrected rigorous imaging model is then introduced to obtain the sharpness detection result of the second aerial remote sensing image.

[0069] Specifically, to analyze various error factors using the Monte Carlo method, the following is established: Figure 2 Coordinate system: Geographic coordinate system OX i Y i Z i O2X, the carrier aircraft navigation coordinate system F YF Z F Establish such as Figure 5 Coordinate systems: Principal point image plane coordinate system oxy, image coordinate system o4JI, image space coordinate system SX G Y G Z G Establish such as Figure 6 Coordinate system: Aircraft attitude coordinate system O2X b Y b Z b Aircraft vibration coordinate system O2X P Y P Z P .

[0070] Using the coordinate systems mentioned above, and comprehensively considering the impact of various influencing factors on the sharpness detection results, the rigorous imaging model is modified, namely:

[0071] dx=dx1+dx2+dx3+dx4+dx5+…(14)

[0072] dy=dy1+dy2+dy3+dy4+dy5+…(15)

[0073]

[0074] like Figure 3 As shown, the aerial remote sensing camera imaging system is fixed to the carrier aircraft and flies forward with the aircraft. During flight, the aerial remote sensing camera performs single-scan tilt imaging along a direction perpendicular to the flight direction of the carrier aircraft, from sweep limit B to sweep limit A. During sweep imaging, the aerial remote sensing camera acquires aerial remote sensing images of ground objects based on the photoelectric imaging principle. There is an overlap between the two acquired aerial remote sensing images, and the width of the overlap area is l. B The system continuously acquires images of ground objects through sweeping. After the aerial remote sensing camera captures the previous image (high-feature area), image processing methods are used to obtain and record its sharpness detection results. Based on the pre-designed image overlap rate, an algorithm is used to search for feature points in the overlapping area. Based on the feature points determined by the algorithm, the aircraft's airborne orientation and positioning system (POS) is used to obtain high-precision attitude information at the moment of the previous image exposure. Combined with geographic information, a modified rigorous imaging model is used to determine the object point coordinates. Then, the modified rigorous imaging model is used to back-calculate the coordinates of the corresponding image points in the next image, realizing a feature matching algorithm. Finally, based on the geographic information of the feature points in the overlapping area, the aircraft's airborne orientation and positioning system (POS) corresponding to the previous image capture is used to obtain its instantaneous high-precision attitude information. Introducing the modified rigorous imaging model yields the sharpness detection results for the next image, thus completing the geographic information-assisted aerial camera weak-feature target sharpness detection acquisition.

[0075]

[0076] like Figure 8 As shown, a modified, rigorous imaging model is used, based on the most fundamental Gaussian imaging formula in optics:

[0077]

[0078] In the formula, H is the flight altitude, f is the camera's main distance, and F is the focal length.

[0079] The effects of external factors such as temperature and pressure, internal factors such as optical distortion, GPS drift error, and eccentricity error are reflected in the minute changes in the camera's principal distance. The accurate position of the focusing lens is obtained through the deformed Gaussian imaging formula, and the focusing lens is moved to this position by a motor.

[0080]

[0081] This invention also provides a system for detecting the sharpness of weak-feature targets in an aerial camera, comprising an aerial remote sensing camera and an optical imaging system disposed within the aerial remote sensing camera; the aerial remote sensing camera includes a cover, a control board, a control motor, and an optical lens; specifically,

[0082] The aerial camera weak feature target sharpness detection system of the present invention mainly includes: a control board 1, the designed algorithm is imported into the control board 3 and installed in the cover 1. After receiving the control signal, the control board 3 will control the optical structure 4 in the lens 5 through the control motor 2 to complete the sharpness detection task.

[0083] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for detecting the sharpness of weak-feature targets in an aerial camera, characterized in that, Includes the following steps: Step S1: Set the parameters of the aerial remote sensing camera; Step S2: Establish a rigorous imaging model; Step S3: Correct the rigorous imaging model using the Monte Carlo method; Step S4: Obtain aerial remote sensing images by scanning ground objects; Step S5: Perform image processing on the aerial remote sensing image to obtain its sharpness detection results; Step S6: Based on the preset image overlap rate, search for feature points in the overlapping area between two aerial remote sensing images to obtain the feature points in the overlapping area; wherein, the two aerial remote sensing images include: a first aerial remote sensing image and a second aerial remote sensing image; the first aerial remote sensing image contains the target with superior features; Step S7: Use geographic information to obtain the matching results of two aerial remote sensing images of overlapping area feature points; Step S8: Based on the geographic information of feature points in the overlapping area, and using the high-precision attitude information from the first aerial remote sensing image, a corrected rigorous imaging model is introduced to obtain the sharpness detection result of the second aerial remote sensing image.

2. The method for detecting the sharpness of weak feature targets in an aerial camera according to claim 1, characterized in that, In step S1, the parameters of the aerial remote sensing camera are set using a precision angle measurement method, including the calibrated focal length, principal point coordinates, and distortion coefficient.

3. The method for detecting the sharpness of weak feature targets in an aerial camera according to claim 2, characterized in that, In step S4, the aerial remote sensing camera follows the aircraft and performs single-strip sweeping tilt imaging along a direction perpendicular to the aircraft's flight direction.

4. The method for detecting the sharpness of weak feature targets in an aerial camera according to claim 3, characterized in that, In step S4, during the scanning imaging, the aerial remote sensing camera takes pictures of the ground object, acquiring two aerial remote sensing images and also acquiring the overlapping part between the two aerial remote sensing images.

5. The method for detecting the sharpness of weak feature targets in an aerial camera according to claim 4, characterized in that, In step S5, after the first aerial remote sensing image is captured by the aerial remote sensing camera, the sharpness detection result is obtained and recorded using image processing methods.

6. The method for detecting the sharpness of weak feature targets in an aerial camera according to claim 5, characterized in that, In step S7, the high-precision attitude information of the first aerial remote sensing image is obtained at the moment of exposure using the aircraft's airborne orientation and positioning system. Then, the object point coordinates are determined using the geographic information and the corrected rigorous imaging model. After that, the image point coordinates of the second aerial remote sensing image are calculated using the corrected rigorous imaging model, thereby obtaining the matching result of the two aerial remote sensing images of the overlapping area feature points.

7. The method for detecting the sharpness of weak feature targets in an aerial camera according to claim 6, characterized in that, In step S8, the aircraft's airborne orientation and positioning system is used to obtain high-precision attitude information at the moment of exposure when the first aerial remote sensing image is captured. The corrected rigorous imaging model is then introduced to obtain the sharpness detection result of the second aerial remote sensing image.

8. A system for detecting the sharpness of weak-feature targets in an aerial camera using the method for detecting the sharpness of weak-feature targets in an aerial camera according to claim 1, characterized in that, It includes an aerial remote sensing camera and an optical imaging system installed inside the aerial remote sensing camera; the aerial remote sensing camera is equipped with a cover, a control board, a control motor and an optical lens.

Images