A method for quickly locating target pixel positions in batches based on SAR images

Abstract

The application relates to a method for quickly and batchly positioning target pixel positions based on a SAR image, which comprises the following steps: reading a SAR data file, judging whether the SAR image contains a target through a vector outer product method, converting a target point and a vertex of the SAR image from an earth coordinate system to an orthogonal coordinate system, calculating a target position relative to a SAR image width and an azimuth width through a geometric method, so as to obtain a rough value of a pixel position of the target point, and calculating the target pixel position through a bilinear interpolation method. The method can quickly and batchly position the target pixel positions, and the efficiency is greatly improved.

Description

Technical Field

[0001] This invention belongs to the field of airborne radar imaging SAR image recognition, specifically a method for rapidly locating target positions during SAR image applications. Background Technology

[0002] Airborne radar imaging utilizes synthetic aperture imaging (SAR), a technique that uses a small antenna to move along the trajectory of a long array and radiate coherent signals. The received signals are recorded and processed appropriately to obtain azimuth resolution equivalent to that of the long array. The main performance indicators of radar imaging include resolution, positioning accuracy, and image coverage, which are also the primary research areas in radar imaging.

[0003] However, there are still some problems with the application of SAR image recognition. There are generally two methods for target localization in SAR images. One is manual search and interpretation. The other is using specialized software to open the image and display the latitude and longitude information at the cursor position in real time by moving the mouse. Method one is extremely inefficient and prone to errors. Method two requires that the presence of a target point in the current image is known. Even if the target point is known to be present, since changes in latitude and longitude are not as intuitive as changes in pixel position, this method is also relatively inefficient.

[0004] For example, to find 20 targets with known latitude and longitude coordinates in 100 SAR images, Method 1 (manual interpretation) would take at least 2 days, while Method 2 (obtaining via mouse) would take at least 2 hours. Summary of the Invention

[0005] Technical problems to be solved

[0006] To overcome the shortcomings of existing technologies, and considering that a single troop operational test mission using a certain domestic radar model generates hundreds of SAR images, it is difficult to efficiently and accurately locate targets or objects of interest. This invention provides a method for efficiently locating the pixel positions of targets with known latitude and longitude within multiple SAR images.

[0007] Technical solution

[0008] A method for rapidly locating target pixels in batches based on SAR images, characterized by the following steps:

[0009] Step 1: Read in a SAR image and obtain the latitude and longitude of four fixed points A, B, C, and D in the SAR image;

[0010] Step 2: Determine if the current image contains the target to be located: If the current image contains the target O, then the four vectors formed by the four vertices of the image in a clockwise direction are... with vector The right-hand screw rule must be satisfied, and the cross products of vectors must have the same sign, i.e. If the condition is met, proceed to step 3; otherwise, return to step 1 and select a new SAR image.

[0011] Step 3: Convert the latitude and longitude coordinates of A, B, and O into the Earth rectangular coordinate system using Equation 1, where A(X) = A(B ... A ,Y A Z A ), B(X) A ,Y A, Z A ), O(X O ,Y O Z O ).

[0012]

[0013] In equation (1), e1 = (a 2 -b 2 ) / a 2 For the first eccentricity, a represents the Earth's major axis, b represents the Earth's minor axis, a = 6378.137 km, b = 6356.752 km; H, B, and L represent altitude, latitude, and longitude, respectively.

[0014] The length of AB is:

[0015]

[0016] Similarly, the lengths l of BO and AO can be obtained. BO l OA Then the perimeter of the triangle is p = l. AB +l BO +l OA .

[0017] Using Heron's formula, calculate the area S of triangle ABO:

[0018]

[0019] Then the height EO of the triangle is H. EO =2S / l AB Similarly, the height H of triangle ADO can be obtained. FO AE length l AE =l FO AF length l AF =l BG =l EO ;

[0020] Step 4: Let W be the width of the SAR image and the number of pixels. N The number of height pixels is H N Using geometric proportions, the position of target pixel O is then... Because SAR images are geometrically corrected, they have slight shape distortions compared to electronic maps. It is a rough pixel position, denoted as O'(x', y');

[0021] Step 5: Calculate the latitude and longitude of each pixel in a 100-pixel square neighborhood of point O'(x', y'). Using formulas 1 and 2, calculate the pixel position O(x, y) closest to O(lat, lon). This pixel position is the precise pixel position of target O in the current image. The bilinear interpolation method is as follows:

[0022] The latitude and longitude coordinates of the four vertices of the image are A(lat,lon), B(lat,lon), C(lat,lon), and D(lat,lon); taking point O'(x', y') as an example, its latitude and longitude in the image are:

[0023]

[0024] A method for calculating radar positioning accuracy is characterized by: opening a SAR image using PS software, locating the position with coordinates O(x, y), where the target point displayed in the SAR image is generally located near O(x, y); reading its pixel coordinates using PS; and then using bilinear interpolation to calculate the latitude and longitude O of the radar-detected target point. 探测 (lat,lon), probe value O 探测 The distance between (lat,lon) and the actual value O(lat,lon) is the radar positioning accuracy.

[0025] A computer system is characterized by comprising: one or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method described above.

[0026] A computer-readable storage medium is characterized by storing computer-executable instructions, which, when executed, are used to implement the above-described method.

[0027] Beneficial effects

[0028] This invention proposes a method for rapid batch location of target pixels based on SAR images. The location of each target pixel is acquired instantaneously, significantly improving efficiency. This method has a reasonable structure and is simple to implement in engineering. Compared with traditional methods, the innovation of this invention lies in:

[0029] 1. It can quickly screen out image files containing targets from multiple SAR image files;

[0030] 2. It can quickly and accurately locate the pixel coordinates of the target in the SAR image. Attached Figure Description

[0031] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0032] Figure 1 Flowchart of the target pixel location method;

[0033] Figure 2 The target is located in the geometric representation of the SAR image;

[0034] Figure 3 The search file required for this embodiment;

[0035] Figure 4 Files containing target points;

[0036] Figure 5 Target points on electronic maps;

[0037] Figure 6 SAR image with target points. Detailed Implementation

[0038] 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. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0039] The hardware platform for this invention is a regular computer with more than 4GB of memory, the software development tool is Qt, and the programming language is C++. This patent provides a method, therefore other programming languages ​​and tools can also be used.

[0040] Step 1:

[0041] SAR images are generally quasi-rectangular (see appendix) Figure 2 Furthermore, the latitude and longitude of the four vertices are known; typically, latitude and longitude information is found in the frame header information of SAR image data. (See attached image.) Figure 2 The latitude and longitude of the four vertices A, B, C, and D of the SAR image are read from the frame header.

[0042] Step 2:

[0043] Determine if the current image contains the target to be located. If the current image contains the target O, then the four vectors formed by the four vertices of the image in a clockwise direction are... with vector The right-hand screw rule must be satisfied, and the cross products of vectors must have the same sign, i.e. If the condition is met, proceed to step 3; otherwise, return to step 1 and select a new SAR image.

[0044] Step 3:

[0045] Convert the latitude and longitude coordinates of A, B, and O to Earth rectangular coordinates using Equation 1, and their coordinates are A(X) and A(X). A ,Y A Z A ), B(X) A ,Y A Z A ), O(X O ,Y O Z O ).

[0046]

[0047] In equation (1), e1 = (a 2 -b 2 ) / a 2 For the first eccentricity, a represents the Earth's major axis, b represents the Earth's minor axis, a = 6378.137 km, b = 6356.752 km. H, B, and L represent altitude, latitude, and longitude, respectively.

[0048] The length of BO is:

[0049]

[0050] Similarly, the lengths l of BO and AO can be obtained. BO l OA Then the perimeter of the triangle is p = l. AB +l BO +l OA .

[0051] Using Heron's formula, calculate the area S of triangle ABO:

[0052]

[0053] Then the height EO of the triangle is H. EO =2S / l AB Similarly, the height H of triangle ADO can be obtained. FO AE length l AE =l FO AF length l AF =l BG =l EO .

[0054] Step 4:

[0055] Let W be the width of the SAR image and the number of pixels. N The number of height pixels is H N Using geometric proportions, the position of target pixel O is then... Because SAR images are geometrically corrected, they have slight shape distortions compared to electronic maps. It is a rough pixel position, denoted as O'(x', y').

[0056] Step 5:

[0057] Using bilinear interpolation, the latitude and longitude of each pixel in a 100-pixel square neighborhood of point O'(x', y') are calculated. Formulas 1 and 2 are then used to calculate the pixel position O(x, y) closest to O(lat, lon). This pixel position is the precise pixel location of the target O in the current image. The bilinear interpolation method is as follows:

[0058] The latitude and longitude coordinates of the four vertices of the image are A(lat, lon), B(lat, lon), C(lat, lon), and D(lat, lon). Taking point O'(x', y') as an example, its latitude and longitude in the image are:

[0059]

[0060] Step 6:

[0061] Open the SAR image with Photoshop and locate the position O(x, y). Generally, the target point displayed in the SAR image is located near O(x, y), as shown in the attached image. Figure 2 Point O' in the image is used to read its pixel coordinates using PS. Then, using bilinear interpolation, the latitude and longitude O of the radar target can be calculated. 探测 (lat,lon), probe value O 探测 The distance between (lat,lon) and the actual value O(lat,lon) is the radar positioning accuracy.

[0062] Use Case Description

[0063] Figure 3 The data is SAR imagery from a certain flight. The goal is to find the target point with latitude and longitude (109.7931132615, 21.2509164681). Figure 4 To find the results, the results are displayed. Figure 3 Of the 86 files, 3 contain the target point, and its pixel position is as follows: Figure 4 As shown.

[0064] See the electronic map of the target point Figure 5According to the latitude and longitude coordinates of this target point, its location in the SAR image is point O. Comparing it with the electronic map, it is clear that point O' near point O is the radar detection point (see...). Figure 6 ).

[0065] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the scope of the technology disclosed in the present invention, and such modifications or substitutions should all be covered within the scope of protection of the present invention.

Claims

1. A method for rapidly locating target pixel positions in batches based on SAR images, characterized in that... The steps are as follows: Step 1: Read in a SAR image and obtain the latitude and longitude of four fixed points A, B, C, and D in the SAR image; Step 2: Determine if the current image contains the target to be located: If the current image contains the target O, then the four vectors formed by the four vertices of the image in a clockwise direction are... with vector The right-hand screw rule must be satisfied, and the cross products of vectors must have the same sign, i.e. If the condition is met, proceed to step 3; otherwise, return to step 1 and select a new SAR image. Step 3: Convert the latitude and longitude coordinates of A, B, and O into the Earth's rectangular coordinate system using Equation 1. The coordinates are A(X, B, O) and O(X, O). A , Y A, Z A ), B (X) A ,Y A, Z A ), O (X) O , Y O, Z O ): （1） In equation (1), For the first eccentricity, , where a is the Earth's major semi-axis and b is the Earth's minor semi-axis, a = 6378.137km, b = 6356.752km; H、 、L These are altitude, latitude, and longitude, respectively. The length of AB is: (2) Similarly, the lengths of BO and AO can be obtained. , Then the perimeter of the triangle is... ; Using Heron's formula, calculate the area S of triangle ABO: (3) Then the altitude EO of the triangle is Similarly, the height H of triangle ADO can be obtained. FO AE length AF length ; Step 4: Let the width of the SAR image and the number of pixels be... W N The number of height pixels is H N Using geometric proportions, the position of target pixel O is then... Because SAR images undergo geometric correction, they exhibit slight shape distortion compared to electronic maps. This is a rough pixel position, denoted as . ; Step 5: Calculate the points one by one using bilinear interpolation. The latitude and longitude of pixels in a square neighborhood with a side length of 100 pixels are used to calculate the distance using Formulas 1 and 2. The nearest point, the pixel location of this point. This represents the precise pixel location of the target O in the current image; the bilinear interpolation method is as follows: The latitude and longitude coordinates of the four vertices of the image are respectively ; With point For example, the latitude and longitude of the image are: （4）。 2. A method for calculating radar positioning accuracy based on the method for rapid batch positioning of target pixels using SAR images as described in claim 1, characterized in that: Open the SAR image with Photoshop and locate the coordinates. The location, the target shown in the SAR image is... Nearby, using PS to read its pixel coordinates, and employing bilinear interpolation, the latitude and longitude of the radar detection target can be obtained. Detection value and actual value The distance between them is the radar positioning accuracy.

3. A computer system, characterized in that... include: One or more processors, a computer-readable storage medium for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method of claim 1.

4. A computer-readable storage medium, characterized in that... The device stores computer-executable instructions, which, when executed, are used to implement the method of claim 1.

Images