An underwater target recognition method based on sonar image and laser image information fusion

Abstract

The application provides an underwater target recognition method based on information fusion of sonar images and laser images, comprising the following steps: collecting a suspected target sonar image by using a two-dimensional imaging sonar; performing edge detection on the suspected target sonar image, and performing contour searching on the sonar image after the edge detection to determine a target position; inputting the target position information into a sonar angle calculation formula and a sonar distance conversion formula to calculate target angle information and target distance information; controlling a laser camera to rotate according to the target angle information, so that the laser camera is directed to a direction where the target is located; performing distance gate imaging on the laser camera according to the target distance information to obtain a target laser image, and performing target recognition on the target laser image by using a target recognition algorithm to obtain target information. The application scheme reduces manual intervention in the detection process, makes up for the shortcomings of poor sonar image quality and complicated laser camera parameter setting, and improves the underwater target recognition efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of underwater target recognition technology, and in particular to an underwater target recognition method based on the fusion of sonar image and laser image information. Background Technology

[0002] Sonar and photoelectric imaging are two of the most important technologies for detecting and identifying underwater targets, but currently they are used separately. Imaging sonar images based on the backscattering of sound waves emitted by an array by an object. It has advantages such as long detection range and large field of view underwater, and is widely used in underwater detection and other fields. However, due to its unique imaging mechanism, the acoustic image quality is relatively poor, especially in terms of image detail, with low contrast and blurred edges, failing to meet the requirements for underwater target identification. Laser range-gated imaging technology uses pulsed lasers as the active light source and utilizes the time-of-flight gating principle. The detector only receives the light signal reflected from the target distance, effectively shielding the backscattered signal between the detector and the target. It has advantages such as high signal-to-noise ratio, long detection range, and good imaging effect, and is widely used in underwater target detection and other fields. However, range-gated imaging technology is a fixed-range imaging technology, meaning it requires prior knowledge of the distance from the target to the detector. By adjusting the gating time parameter, it detects at a fixed distance near the target, resulting in low search efficiency and poor real-time detection, posing significant challenges to practical engineering applications.

[0003] Therefore, in order to fully leverage the respective advantages of imaging sonar and range-gated imaging, rapid and accurate identification of underwater targets has become an urgent problem to be solved. Summary of the Invention

[0004] To overcome the shortcomings and deficiencies of existing technologies, this invention provides an underwater target identification method based on the fusion of sonar and laser image information, comprising the following steps:

[0005] Step 1: Mount the two-dimensional imaging sonar and laser camera on the underwater robot with the two-dimensional imaging sonar facing the same direction, and use the two-dimensional imaging sonar to acquire sonar images of suspected targets;

[0006] Step 2: Perform edge detection on the suspected target sonar image, and then perform contour search on the edge-detected sonar image to determine the target location;

[0007] Step 3: Input the target location information into the sonar angle calculation formula and the sonar distance conversion formula for calculation to obtain the target angle information and target distance information;

[0008] Step 4: Send the obtained target angle information to the underwater robot. The underwater robot controls the rotation of the laser camera according to the target angle information. After reaching the target angle, the rotation stops, so that the laser camera is facing the target.

[0009] Step 5: The laser camera performs distance gating imaging based on the target distance information to obtain a high-resolution target laser image, and uses a target recognition algorithm to identify the target information from the high-resolution target laser image.

[0010] Further, step 2 includes:

[0011] The Sobel operator is used to perform edge detection on the sonar image of the suspected target to obtain the image edge information of the target;

[0012] Local adaptive threshold binarization is performed on the sonar image after edge detection, and contour search is performed to obtain the coordinates of the target center position.

[0013] Further, step 3 includes:

[0014] Based on the target's center coordinates, the target's angle information is obtained using the sonar angle calculation formula, as follows:

[0015]

[0016] Where θ represents the angle by which the target deviates from the center line of the sonar field of view, (x,y) represents the coordinates of the target center in the sonar image, and w represents the width of the sonar image;

[0017] Based on the correspondence between the target's center coordinates and the sonar detection range and sonar image altitude, the target distance information is obtained using the sonar distance conversion formula, as follows:

[0018]

[0019] Where D represents the actual distance from the target to the sonar, and L pix The value represents the actual distance value per pixel in the sonar image, (x,y) represents the coordinates of the target center in the sonar image, L represents the sonar detection range, w represents the width of the sonar image, and h represents the height of the sonar image.

[0020] By adopting the above technical solution, the present invention has at least the following beneficial effects:

[0021] 1. Optimize the underwater target detection process. Target orientation parameters and laser distance gating parameters are automatically generated based on image processing results, reducing manual intervention during the detection process.

[0022] 2. By combining sonar imaging and laser imaging technologies, the advantages of wide detection range of imaging sonar and clear and distinguishable laser camera images are fully utilized, making up for the shortcomings of poor sonar image quality and cumbersome laser camera parameter settings, thereby improving the efficiency and accuracy of underwater target identification. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. 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.

[0024] Figure 1 This embodiment provides a flowchart of an underwater target recognition method based on the fusion of sonar and laser image information;

[0025] Figure 2 A schematic diagram illustrating the calculation of target angle and distance information in sonar images;

[0026] Figure 3 This is an underwater target identification map based on the fusion of sonar and laser image information. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] refer to Figure 1 This embodiment provides an underwater target identification method based on the fusion of sonar and laser image information, the method comprising:

[0029] Step 1: Mount the two-dimensional imaging sonar and laser camera on the underwater robot with the two-dimensional imaging sonar facing the same direction, and use the two-dimensional imaging sonar to acquire sonar images of suspected targets;

[0030] In this step, the range of the two-dimensional imaging sonar is set, the suspected target is detected, the acquired video stream is sent to the processor, and image frames are taken out at fixed time intervals for further processing.

[0031] Step 2: Perform edge detection on the suspected target sonar image, and then perform contour search on the edge-detected sonar image to determine the target location;

[0032] In this step, the Sobel operator is used to perform edge detection on the suspected target sonar image. The image is convolved with the horizontal and vertical kernels to obtain approximate values ​​of the horizontal and vertical brightness differences, respectively. The specific formulas are as follows:

[0033]

[0034]

[0035] Where A represents the denoised image, G x G y These represent the grayscale values ​​of the image after horizontal and vertical edge detection, respectively;

[0036] The gray level of a point is calculated by using the horizontal and vertical gray values ​​of each pixel in the image, thus obtaining the image edge information of the target.

[0037] Local adaptive threshold binarization is performed on the sonar image after edge detection, and contour search is performed to obtain the coordinates of the target center position.

[0038] Step 3: Input the target location information into the sonar angle calculation formula and the sonar distance conversion formula for calculation to obtain the target angle information and target distance information;

[0039] In this step, such as Figure 2 As shown, the present invention is based on Figure 2 The formula for calculating sonar angles is summarized below. Based on the target's center coordinates, the target angle information is obtained using the sonar angle calculation formula, which is as follows:

[0040]

[0041] Where θ represents the angle by which the target deviates from the center line of the sonar field of view, (x,y) represents the coordinates of the target center in the sonar image, and w represents the width of the sonar image;

[0042] Based on the correspondence between the target's center coordinates and the sonar detection range and sonar image altitude, the target distance information is obtained using the sonar distance conversion formula, as follows:

[0043]

[0044] Where D represents the actual distance from the target to the sonar, and L pix The value represents the actual distance value per pixel in the sonar image, (x,y) represents the coordinates of the target center in the sonar image, L represents the sonar detection range, w represents the width of the sonar image, and h represents the height of the sonar image.

[0045] Step 4: Send the obtained target angle information to the underwater robot. The underwater robot controls the rotation of the laser camera according to the target angle information. After reaching the target angle, the rotation stops, so that the laser camera is facing the target.

[0046] In this step, the two-dimensional imaging sonar and laser camera are fixed on the underwater robot with their fields of view aligned. When the target angle information is sent to the underwater robot, the underwater robot begins to rotate. When the underwater robot rotates to the set angle, it stops rotating. At this time, the target is located on the center line of the field of view of the two-dimensional imaging sonar and laser camera.

[0047] Step 5: The laser camera performs distance gating imaging based on the target distance information to obtain a high-resolution target laser image, and uses a target recognition algorithm to identify the target information from the high-resolution target laser image.

[0048] In this step, the underwater robot rotates into position, and the target is located on the center line of the field of view of the two-dimensional imaging sonar and the laser camera. Within the field of view of the laser camera, the laser camera gate distance is set according to the target distance information to image the location, obtain a high-resolution target laser image, and complete the target identification through the target recognition algorithm.

[0049] Target recognition results are as follows Figure 3 As shown, the underwater target recognition method based on the fusion of sonar and laser image information can quickly detect suspected targets at unknown distances and determine their location and distance. Laser gating imaging is then performed, and target recognition is performed on the laser image. This method ensures both the accuracy and efficiency of target recognition.

[0050] Although the present invention has been disclosed above with reference to embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. An underwater target identification method based on the fusion of sonar and laser image information, characterized in that, include: Step 1: Mount the two-dimensional imaging sonar and laser camera on the underwater robot with the two-dimensional imaging sonar facing the same direction, and use the two-dimensional imaging sonar to acquire sonar images of suspected targets; Step 2: Perform edge detection on the suspected target sonar image, and then perform contour search on the edge-detected sonar image to determine the target location; Step 3: Input the target location information into the sonar angle calculation formula and the sonar distance conversion formula for calculation to obtain the target angle information and target distance information; Step 4: Send the obtained target angle information to the underwater robot. The underwater robot controls the rotation of the laser camera according to the target angle information. After reaching the target angle, the rotation stops, so that the laser camera is facing the target. Step 5: The laser camera performs distance gating imaging based on the target distance information to obtain a high-resolution target laser image, and uses a target recognition algorithm to identify the target information from the high-resolution target laser image. Step 3 includes: Based on the target's center coordinates, the target's angle information is obtained using the sonar angle calculation formula, as follows: in, This indicates the angle by which the target deviates from the center line of the sonar field of view. This indicates the coordinates of the target center in the sonar image. Indicates the width of the sonar image; Based on the correspondence between the target's center coordinates and the sonar detection range and sonar image altitude, the target distance information is obtained using the sonar distance conversion formula, as follows: in, This indicates the actual distance from the target to the sonar. This means that each pixel in the sonar image represents the actual distance value. This indicates the coordinates of the target center in the sonar image. Indicates the sonar detection range. Indicates the width of the sonar image. Indicates the height of the sonar image.

2. The underwater target identification method based on the fusion of sonar image and laser image information according to claim 1, characterized in that, Step 2 includes: The Sobel operator is used to perform edge detection on the sonar image of the suspected target to obtain the image edge information of the target; Local adaptive threshold binarization is performed on the sonar image after edge detection, and contour search is performed to obtain the coordinates of the target center position.

Images