Target Spot Extraction Method and Device

Abstract

The present invention provides a method and device for extracting a target light spot. It relates to the technical field of image processing and includes: when a target light spot is detected in the first image frame captured by the capture tracker in the search state, the capture tracker switches from the search state to the protection state and acquires N consecutive second image frames after the first image frame; when the target light spot is detected in all of the N second image frames, the position fluctuation distance between the target light spot in each of the second image frames and the target light spot in the first image frame is acquired; where N is a positive integer; when each of the position fluctuation distances is less than a first preset threshold, the capture tracker switches from the protection state to the locking state, and the capture tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a method and device for extracting a target light spot. Background Art

[0002] With the increase in the number of satellites, the downlink volume also increases. The variable coding modulation (VCM) technology of microwave communication can achieve a downlink rate of 2.0 Gbps, but still cannot meet the demand for the rapidly growing downlink rate. Laser communication can achieve 10 Gbps for a single channel, and the downlink rate can reach 100 Gbps through optical path multiplexing. Due to the narrow beam range of laser communication, it poses a challenge to the automation tracking of the rack.

[0003] Stable and accurate extraction of laser light spots in laser communication is one of the key technologies to ensure the stability of the communication link, and its accuracy and precision directly determine the efficiency of laser communication and the stability of the link. When the communication distance becomes longer and the atmospheric turbulence effect increases, the laser beam passing through the atmospheric channel drifts and expands, and the different intensities of turbulence in the atmosphere cause the wavefront phase reaching the receiving end to also show random fluctuations, resulting in different degrees of distortion of the image. On the other hand, the interference of factors such as the high-speed movement of the satellite, platform jitter, and complex space environment also increases the difficulty of accurately extracting the light spot target.

[0004] Therefore, how to accurately extract the light spot target has become an urgent problem to be solved in the industry. Summary of the Invention

[0005] The present invention provides a method and device for extracting a target light spot to solve the defect of how to accurately extract the light spot target in the prior art.

[0006] The present invention provides a method for extracting a target light spot, including:

[0007] When a target light spot is detected in the first image frame captured by the acquisition tracker in the search state, the acquisition tracker switches from the search state to the protection state and acquires N consecutive second image frames after the first image frame;

[0008] When the target light spot is detected in all of the N second image frames, obtain the position fluctuation distance between the target light spot in each second image frame and the target light spot in the first image frame; where N is a positive integer;

[0009] When all of the position fluctuation distances are less than a first preset threshold, the acquisition tracker switches from the protection state to the locking state, and the acquisition tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot.

[0010] According to a method for extracting a target light spot provided by the present invention, after the step of the capture tracker keeping the target light spot at the center of the field of view and continuously tracking the target light spot, the method further includes:

[0011] When the position fluctuation distance between the target light spots in M consecutive third image frames and the target light spot in the first image frame exceeds a second preset threshold, the capture tracker switches from the locked state to the search state.

[0012] According to a method for extracting a target light spot provided by the present invention, the detection method of the target light spot specifically includes:

[0013] Performing image preprocessing on the captured image frame to obtain a preprocessed image frame;

[0014] Using a threshold segmentation algorithm to divide the pixels in the preprocessed image frame into multiple different pixel regions;

[0015] Generating a binary image based on the different pixel regions, and determining the target light spot based on the non-zero pixels in the binary image.

[0016] According to a method for extracting a target light spot provided by the present invention, the positioning method of the target light spot in the first image frame includes:

[0017] Calculating the gradient of the binary threshold segmentation image corresponding to the first image frame to obtain the gradient amplitude and direction of each pixel point in the binary threshold segmentation image;

[0018] Performing threshold segmentation on the gradient amplitude of the pixel points in the binary threshold segmentation image to obtain the binary image corresponding to the first image frame;

[0019] Based on the connectivity of the image, searching for the connected components of non-zero pixels in the binary image to determine the contour points of the target light spot;

[0020] After converting the contour points into two-dimensional coordinates, using the least squares method to fit the two-dimensional coordinates into an ellipse to obtain ellipse parameters, so as to determine the first positioning information of the target light spot in the first image frame according to the ellipse parameters.

[0021] According to a method for extracting a target light spot provided by the present invention, obtaining the position fluctuation distance between the target light spots in each second image frame and the target light spot in the first image frame includes:

[0022] Obtaining the second positioning information of the target light spot in the second image frame;

[0023] Determine the position fluctuation distance based on the first spot center coordinates in the first positioning information and the second spot center coordinates in the second positioning information.

[0024] According to a target spot extraction method provided by the present invention, the preprocessing the captured image frame to obtain a preprocessed image frame includes:

[0025] After performing color gamut transformation on the image frame, perform Gaussian filtering to obtain the filtered image frame;

[0026] Adjust the gray level distribution of the filtered image frame by histogram equalization to obtain the preprocessed image frame.

[0027] The present invention also provides a target extraction device, including:

[0028] A switching module, configured to, when a target spot is detected in a first image frame captured by a tracker in a search state, switch the tracker from the search state to a protection state, and obtain N consecutive second image frames after the first image frame;

[0029] An acquisition module, configured to, when a target spot is detected in all of the N second image frames, obtain the position fluctuation distance between the target spots in each of the second image frames and the target spot in the first image frame; where N is a positive integer;

[0030] A tracking module, configured to, when each of the position fluctuation distances is less than a first preset threshold, switch the tracker from the protection state to a locking state, and the tracker keeps the target spot at the center of the field of view and continuously tracks the target spot.

[0031] According to a target extraction device provided by the present invention, the device is further configured to:

[0032] When the position fluctuation distances between the target spots in M consecutive third image frames and the target spot in the first image frame all exceed a second preset threshold, switch the tracker from the locking state to the search state.

[0033] According to a target extraction device provided by the present invention, the device is further configured to:

[0034] Preprocess the captured image frame to obtain a preprocessed image frame;

[0035] Use a threshold segmentation algorithm to divide the pixels in the preprocessed image frame into multiple different pixel regions;

[0036] Generate a binary image based on the different pixel regions, and determine the target light spot based on the non-zero pixels in the binary image.

[0037] According to a target extraction device provided by the present invention, the device is further configured to:

[0038] Calculate the gradient of the binary threshold segmentation image corresponding to the first image frame to obtain the gradient magnitude and direction of each pixel point in the binary threshold segmentation image;

[0039] Perform threshold segmentation on the gradient magnitude of the pixel points in the binary threshold segmentation image to obtain the binary image corresponding to the first image frame;

[0040] Based on the connectivity of the image, find the connected components of non-zero pixels in the binary image to determine the contour points of the target light spot;

[0041] After converting the contour points into two-dimensional coordinates, use the least squares method to fit the two-dimensional coordinates into an ellipse, obtain the ellipse parameters, and determine the first positioning information of the target light spot in the first image frame according to the ellipse parameters.

[0042] According to a target extraction device provided by the present invention, the device is further configured to:

[0043] Obtain the second positioning information of the target light spot in the second image frame;

[0044] Based on the first light spot center coordinates in the first positioning information and the second light spot center coordinates in the second positioning information, determine the position fluctuation distance.

[0045] According to a target extraction device provided by the present invention, the device is further configured to:

[0046] After performing color gamut transformation on the image frame, perform Gaussian filtering to obtain the filtered image frame;

[0047] Adjust the gray level distribution of the filtered image frame by using histogram equalization to obtain the preprocessed image frame.

[0048] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method for extracting the target light spot as described in any one of the above is implemented.

[0049] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for extracting the target light spot as described in any one of the above is implemented.

[0050] The present invention also provides a computer program product, including a computer program which, when executed by a processor, implements the target spot extraction method as described in any one of the above.

[0051] The target spot extraction method and device provided by the present invention detect the target spot in the first image frame in the search state and analyze the consecutive N second image frames after the first image frame, ensuring the continuity and authenticity of the target, reducing the possibility of misjudgment. Once the position fluctuations of the target spot in the consecutive N image frames are within the preset threshold, the system quickly switches to the locked state, can quickly respond and lock the target, providing guarantee for the stability of the laser communication link. In the locked state, the system keeps the target spot at the center of the field of view and continuously tracks the target spot, which ensures high-precision alignment during the laser communication process, thus improving the quality and efficiency of communication. Through effective state management and quick locking, unnecessary resource consumption in the search state is reduced, and the overall performance of the system is improved. The solution of the embodiment of the present application realizes high-precision extraction and stable tracking of the laser spot in a complex environment, and has significant technical effects on improving the overall performance of the laser communication system. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0053] Figure 1 It is a schematic flowchart of the target spot extraction method provided by the embodiment of the present application;

[0054] Figure 2 It is a schematic flowchart of the laser communication spot positioning provided by the embodiment of the present application;

[0055] Figure 3 It is a schematic structural diagram of the target spot extraction device provided by the embodiment of the present application;

[0056] Figure 4 It is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0058] Figure 1 It is a schematic flowchart of the target spot extraction method provided by an embodiment of the present application. As Figure 1 shown, it includes:

[0059] Step 110, when a target spot is detected in the first image frame captured by the acquisition tracker in the search state, the acquisition tracker switches from the search state to the protection state and obtains N consecutive second image frames after the first image frame;

[0060] In the embodiment of the present application, the acquisition tracker is a key component in the laser communication system, and its main function is to ensure that the laser signal can be accurately aligned and stably maintained on the receiving device.

[0061] In the embodiment of the present application, in the initial search state, the system searches for and attempts to identify the target spot. Once the target spot is detected in the search state, the system enters the protection state, starts tracking the spot, and prepares to further confirm the stability of the spot. If the spot is successfully detected in N consecutive image frames and the position fluctuation of the spot is within the preset threshold, the system considers that the spot has been stably tracked and enters the locked state.

[0062] The target spot in the laser communication system refers to the spot formed by the beam emitted by the laser and transmitted through the atmosphere or other media to the receiving end. The acquisition tracker uses image processing techniques, such as threshold segmentation, edge detection, morphological operations, etc., to identify and locate the target spot. Once the target spot is detected, the acquisition tracker will enter the protection state and use various algorithms and techniques, such as Kalman filtering, prediction algorithms, etc., to continuously track the spot.

[0063] The N consecutive second image frames in the embodiment of the present application specifically refer to the image frames continuously obtained after the first image frame is obtained.

[0064] Step 120, when the target spot is detected in all of the N second image frames, obtain the position fluctuation distance between the target spot in each of the second image frames and the target spot in the first image frame; where N is a positive integer;

[0065] In an embodiment of the present application, in the protected state of the acquisition tracker, if the target light spot is successfully detected in N consecutive second image frames, it indicates that the system has continuously and stably tracked the target light spot.

[0066] The system will measure and analyze the position fluctuation of the target light spot relative to the position in the first frame image in these N frames of images. This is accomplished by comparing the center positions of the light spots in consecutive frames.

[0067] Step 130, when each of the position fluctuation distances is less than a first preset threshold, the acquisition tracker switches from the protected state to the locked state, and the acquisition tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot.

[0068] If the position fluctuations of the light spot in all N frames of images are within the first preset threshold, it indicates that the position of the light spot is stable and the system has successfully tracked the target light spot continuously. After confirming the stability of the light spot, the acquisition tracker will switch from the protected state to the locked state. The locked state means that the system is highly confident that the target light spot is stably tracked and the tracking algorithm can be further optimized to maintain this state.

[0069] In the locked state, the system will continuously track the target light spot and can maintain high-precision tracking even in the face of complex environmental factors such as atmospheric turbulence, satellite movement, and platform jitter.

[0070] In the locked state, the system will strive to maintain the position of the light spot at the center of the field of view to ensure the stability of the laser communication link. The system may fine-tune the pointing of the optical system according to the deviation between the actual position and the expected position of the light spot to counteract external disturbances.

[0071] In the locked state, the system can further optimize the tracking algorithm, such as by adjusting filter parameters, improving prediction models, etc., to improve the tracking accuracy and response speed.

[0072] In the embodiment of the present application, by detecting the target light spot in the first image frame in the search state and analyzing the continuous N second image frames after the first image frame, the continuity and authenticity of the target are ensured, the possibility of misjudgment is reduced. Once the position fluctuations of the target light spot in the continuous N image frames are all within the preset threshold, the system quickly switches to the locked state, can quickly respond and lock the target, providing guarantee for the stability of the laser communication link. In the locked state, the system keeps the target light spot at the center of the field of view and continuously tracks the target light spot, which ensures high-precision alignment during the laser communication process, thereby improving the quality and efficiency of communication. Through effective state management and quick locking, unnecessary resource consumption in the search state is reduced, and the overall performance of the system is improved. The solution of the embodiment of the present application realizes high-precision extraction and stable tracking of the laser light spot in a complex environment, and has significant technical effects on improving the overall performance of the laser communication system.

[0073] Optionally, after the step where the acquisition tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot, the method further includes:

[0074] In the case where the position fluctuation distances between the target light spots in the continuous M third image frames and the target light spot in the first image frame all exceed the second preset threshold, the acquisition tracker switches from the locked state to the search state.

[0075] In the embodiment of the present application, in the locked state, the system continues to monitor the position fluctuations of the target light spot in the continuous M third image frames. The system measures and records the position fluctuation distances of the target light spot relative to the first image frame in these M image frames.

[0076] The system compares these measured position fluctuation distances with a second preset threshold. This threshold is used to determine whether the light spot significantly deviates from its expected position at the center of the field of view. If in the continuous M frames, the position fluctuation distances of the target light spot all exceed the second preset threshold, it indicates that the stability of the light spot no longer meets the requirements of the system, which may be due to external disturbances, internal system problems, or the movement of the target itself.

[0077] In the embodiment of the present application, once it is confirmed that the stability of the light spot is lost, the acquisition tracker will switch back from the locked state to the search state, and be ready to re-capture and track the target light spot.

[0078] In the search state, the system will re-scan the field of view and use image processing and pattern recognition technologies to re-detect and identify the target light spot.

[0079] This state transition constitutes a feedback loop, allowing the system to automatically recover when losing tracking, enhancing the robustness of the system.

[0080] Optionally, the detection method of the target light spot specifically includes:

[0081] Perform image preprocessing on the captured image frame to obtain a preprocessed image frame;

[0082] Use a threshold segmentation algorithm to divide the pixels in the preprocessed image frame into multiple different pixel regions;

[0083] Generate a binary image based on the different pixel regions, and determine the target light spot based on the non-zero pixels in the binary image.

[0084] Optionally, the performing image preprocessing on the captured image frame to obtain a preprocessed image frame includes:

[0085] After performing color gamut transformation on the image frame, perform Gaussian filtering to obtain the filtered image frame;

[0086] Use histogram equalization to adjust the gray level distribution of the filtered image frame to obtain a preprocessed image frame.

[0087] In the embodiments of the present application, image preprocessing includes normalization, image enhancement, etc. By normalization, the image gray level is controlled within the range of 0-255, which is convenient for subsequent image processing; adopt an image enhancement algorithm to reduce noise and improve image clarity, thereby improving the interpretation accuracy. Image enhancement algorithms include algorithms such as filtering, smoothing, sharpening, contrast enhancement, and edge enhancement. According to the needs of the actual image, select a suitable image enhancement algorithm to improve the clarity of the image, increase the resolution of the image, and the performance of the extraction algorithm.

[0088] Specifically, in the embodiments of the present application, color gamut transformation is to convert a color image (represented by three channels of Blue, Green, and Red) into a grayscale image (only one channel represents the gray level). The color gamut conversion formula is implemented by the weighted average method, where the weights of different channels reflect the contribution degree of different colors to the final gray value. Usually, the weight of green is higher because the human eye is more sensitive to green. The transformation formula is:

[0089] Gray = 0.299 * Blue + 0.587 * Green + 0.114 * Red

[0090] More specifically, in the embodiments of the present application, the basic idea of Gaussian filtering is to use a two-dimensional Gaussian function as a convolution kernel to perform a convolution operation on the image. The shape of this convolution kernel is determined by the Gaussian function. Specifically, the values of the convolution kernel are calculated from the Gaussian distribution function.

[0091] The two-dimensional convolution operation formula of Gaussian filtering is as follows:

[0092]

[0093] Among them, G(x, y) is the value of the two-dimensional Gaussian function at the coordinates (x, y), and σ is the standard deviation of the Gaussian function.

[0094] The formula for the convolution operation is as follows:

[0095]

[0096] Among them, I(x, y) is the pixel value of the original image at the coordinates (x, y), G(i, j) is the value of the Gaussian convolution kernel at the coordinates (i, j), and k is the radius of the convolution kernel.

[0097] In the embodiments of the present application, image enhancement specifically adjusts the gray-level distribution of the image by histogram equalization to make the brightness of the image more uniform, linearizes the cumulative distribution function (CDF) of the image, so that the pixel values of the image are more evenly distributed within the entire gray range, thereby enhancing the visual effect of the image.

[0098]

[0099]

[0100] Among them, histogram(j) is the number of pixels with the gray level j in the image, CDF min is the minimum value of the CDF, and pixels_number is the total number of pixels in the image.

[0101] First, calculate the histogram of the image: count the number of pixels for each gray level in the image.

[0102] Then, calculate the cumulative distribution function (CDF) of the histogram: accumulate each entry in the histogram to obtain a cumulative distribution function.

[0103] Next, perform normalization: map the values in the CDF to the gray-level range (usually 0 to 255).

[0104] Finally, perform mapping: use the normalized CDF values to map each pixel value of the image to a new brightness value to obtain the preprocessed image frame.

[0105] In the embodiments of the present application, the threshold segmentation algorithm segments the image according to the detection target type and image characteristics of laser communication. The threshold method is simple to calculate and has good versatility, and can divide the region with the same features. Selecting an appropriate threshold can better filter out the noise and other disturbances caused by system noise, reduce the calculation time, and improve the positioning accuracy at the same time.

[0106] Threshold segmentation is used to divide the pixels of an image into two or more different regions to simplify the image or emphasize the features of interest. The core idea of threshold segmentation is to compare the gray value of the pixel with the set threshold to divide the pixels into different categories.

[0107]

[0108] Among them: Output(x,y) is the pixel value of the output image. Input(x,y) is the pixel value of the input image. Threshold is the set threshold. MaxValue is the maximum pixel value of the output image, usually 255.

[0109] The steps are as follows:

[0110] Select a threshold: Select an appropriate threshold, usually determined by histogram analysis, trial and error method or other image processing techniques.

[0111] Pixel classification: Compare each pixel in the image with the threshold. If the pixel value is greater than the threshold, it is assigned to one category; if the pixel value is less than or equal to the threshold, it is assigned to another category.

[0112] Generate a binary image: Generate a binary image according to the classification result of the pixels, where the pixels above the threshold are set to white (255), and the pixels below the threshold are set to black (0).

[0113] In the embodiments of the present application, the target light spot can be determined according to the connected regions in the binary image.

[0114] Optionally, the method for positioning the target light spot in the first image frame includes:

[0115] Calculate the gradient of the binary threshold segmentation image corresponding to the first image frame to obtain the gradient magnitude and direction of each pixel point in the binary threshold segmentation image;

[0116] Perform threshold segmentation on the gradient magnitude of the pixel points in the binary threshold segmentation image to obtain the binary image corresponding to the first image frame;

[0117] Based on the connectivity of the image, find the connected components of non-zero pixels in the binary image to determine the contour points of the target light spot;

[0118] After converting the contour points into two-dimensional coordinates, the least squares method is used to fit the two-dimensional coordinates into an ellipse to obtain ellipse parameters, so as to determine the first positioning information of the target light spot in the first image frame according to the ellipse parameters.

[0119] In the embodiments of the present application, the gradient calculation can specifically be performed by calculating the gradient of the binary threshold segmentation image, and the gradient amplitude and direction of each pixel point in the image can be obtained.

[0120]

[0121]

[0122] Among them, S x and S y are the weights of the Sobel operator; G x and G y respectively represent the gradients of the image in the x and y directions.

[0123] In the embodiments of the present application, performing threshold segmentation on the gradient amplitude of the pixel points in the binary threshold segmentation image specifically refers to performing threshold segmentation on the gradient amplitude to obtain a binary image. The purpose of this step is to emphasize the regions with large gradient changes, that is, the regions that may contain the object contour.

[0124]

[0125] In the embodiments of the present application, after obtaining the binary image, based on the connectivity of the image, the contour is determined by finding the connected components of non-zero pixels in the binary image.

[0126] Convert the contour points into a set of two-dimensional coordinates (x, y), and use the least squares method to fit an ellipse. Find the ellipse parameters that minimize the fitting error. After the fitting is completed, obtain the parameters of the ellipse, the center coordinates (cx, cy), the major axis length a, and the minor axis length b.

[0127]

[0128] In the embodiments of the present application, in each frame of image, the system uses image processing techniques (such as threshold segmentation, edge detection, morphological operations, etc.) to locate the center position of the light spot. Set thresholds (ΔX, ΔY) for the position fluctuations in the X and Y directions, and these thresholds are configurable and used to judge the stability of the light spot position.

[0129] Measure the fluctuations of the spot center position in N consecutive frames of images and compare them with ΔX and ΔY. That is, calculate the average off-target amount of the spot center position in N frames of images, which is the average deviation between the actual position of the spot and the expected position (usually the center of the field of view), and determine the position fluctuation distance. If the position fluctuation distance of the spot center position in N consecutive frames of images is lower than the set first preset threshold (ΔX and ΔY), it is considered that the spot capture is successful.

[0130] Figure 2 Schematic diagram of the laser communication spot positioning process provided by an embodiment of this application, as Figure 2 shown, including:

[0131] Scan and capture, process each frame of the captured image, such as performing color gamut conversion, Gaussian filtering, image enhancement, threshold segmentation, contour fitting, and spot positioning. If the number of frames detected for spot positioning reaches the target number of frames, calculate the off-target amount.

[0132] When it is confirmed that a spot is detected, position the spot to the center and then maintain the spot.

[0133] If the spot is detected to be lost, end the process. If the spot is detected again, restart the scan and capture stage.

[0134] The target spot extraction device provided by the present invention will be described below. The target spot extraction device described below can be mutually referred to the target spot extraction method described above.

[0135] Figure 3 Schematic diagram of the structure of the target spot extraction device provided by an embodiment of this application, as Figure 3 shown, including:

[0136] The switching module 310 is used to switch the capture tracker from the search state to the protection state and obtain N consecutive second image frames after the first image frame when a target spot is detected in the first image frame captured by the capture tracker in the search state;

[0137] The acquisition module 320 is used to obtain the position fluctuation distance between the target spots in each of the second image frames and the target spot in the first image frame when target spots are detected in all of the N second image frames; where N is a positive integer;

[0138] The tracking module 330 is used to switch the capture tracker from the protection state to the locked state when each of the position fluctuation distances is less than the first preset threshold, and the capture tracker keeps the target spot at the center of the field of view and continuously tracks the target spot.

[0139] According to a target extraction device provided by the present invention, the device is further used for:

[0140] When the position fluctuation distance between the target light spot in the continuous M-frame third image frame and the target light spot in the first image frame exceeds the second preset threshold, the capture tracker switches from the locked state to the search state.

[0141] According to a target extraction device provided by the present invention, the device is further configured to:

[0142] Perform image preprocessing on the captured image frame to obtain a preprocessed image frame;

[0143] Use a threshold segmentation algorithm to segment the pixels in the preprocessed image frame into multiple different pixel regions;

[0144] Generate a binary image based on the different pixel regions, and determine the target light spot based on the non-zero pixels in the binary image.

[0145] According to a target extraction device provided by the present invention, the device is further configured to:

[0146] Perform gradient calculation on the binary threshold segmentation image corresponding to the first image frame to obtain the gradient magnitude and direction of each pixel point in the binary threshold segmentation image;

[0147] Perform threshold segmentation on the gradient magnitude of the pixel points in the binary threshold segmentation image to obtain the binary image corresponding to the first image frame;

[0148] Based on the connectivity of the image, find the connected components of the non-zero pixels in the binary image to determine the contour points of the target light spot;

[0149] After converting the contour points into two-dimensional coordinates, use the least squares method to fit the two-dimensional coordinates into an ellipse, obtain the ellipse parameters, and determine the first positioning information of the target light spot in the first image frame according to the ellipse parameters.

[0150] According to a target extraction device provided by the present invention, the device is further configured to:

[0151] Obtain the second positioning information of the target light spot in the second image frame;

[0152] Based on the first light spot center coordinates in the first positioning information and the second light spot center coordinates in the second positioning information, determine the position fluctuation distance.

[0153] According to a target extraction device provided by the present invention, the device is further configured to:

[0154] After performing color gamut change on the image frame, perform Gaussian filtering to obtain the filtered image frame;

[0155] The gray level distribution of the filtered image frame is adjusted by histogram equalization to obtain a preprocessed image frame.

[0156] In the embodiment of the present application, by detecting the target light spot in the first image frame in the search state and analyzing the continuous N second image frames after the first image frame, the continuity and authenticity of the target are ensured, the possibility of misjudgment is reduced. Once the position fluctuations of the target light spot in the continuous N image frames are within the preset threshold, the system quickly switches to the locked state, can quickly respond and lock the target, providing guarantee for the stability of the laser communication link. In the locked state, the system keeps the target light spot at the center of the field of view and continuously tracks the target light spot, which ensures high-precision alignment during the laser communication process, thereby improving the quality and efficiency of communication. Through effective state management and quick locking, unnecessary resource consumption in the search state is reduced, and the overall performance of the system is improved. The solution of the embodiment of the present application realizes high-precision extraction and stable tracking of the laser light spot in a complex environment, and has significant technical effects on improving the overall performance of the laser communication system.

[0157] Figure 4 is a schematic structural diagram of an electronic device provided by the present invention, as Figure 4 shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440. Among them, the processor 410, the communication interface 420, and the memory 430 complete mutual communication through the communication bus 440. The processor 410 can call the logical instructions in the memory 430 to execute the target light spot extraction method, and the method includes: when a target light spot is detected in the first image frame captured by the capture tracker in the search state, the capture tracker switches from the search state to the protection state, and obtains the continuous N second image frames after the first image frame;

[0158] When target light spots are detected in all the N second image frames, obtain the position fluctuation distances between the target light spots in each of the second image frames and the target light spot in the first image frame; where N is a positive integer;

[0159] When all the position fluctuation distances are less than the first preset threshold, the capture tracker switches from the protection state to the locked state, and the capture tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot.

[0160] In addition, when the logical instructions in the above-mentioned memory 430 can be implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0161] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the target spot extraction method provided by the above-mentioned various methods. The method includes: when a target spot is detected in the first image frame captured by the capture tracker in the search state, the capture tracker switches from the search state to the protection state and acquires N consecutive second image frames after the first image frame;

[0162] When the target spot is detected in all of the N second image frames, acquire the position fluctuation distance between the target spot in each of the second image frames and the target spot in the first image frame; where N is a positive integer;

[0163] When each of the position fluctuation distances is less than a first preset threshold, the capture tracker switches from the protection state to the locking state, and the capture tracker keeps the target spot at the center of the field of view and continuously tracks the target spot.

[0164] In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the target spot extraction method provided by the above-mentioned various methods. The method includes: when a target spot is detected in the first image frame captured by the capture tracker in the search state, the capture tracker switches from the search state to the protection state and acquires N consecutive second image frames after the first image frame;

[0165] When the target spot is detected in all of the N second image frames, acquire the position fluctuation distance between the target spot in each of the second image frames and the target spot in the first image frame; where N is a positive integer;

[0166] When the fluctuation distance at each of the positions is less than a first preset threshold, the capture tracker switches from the protection state to the locked state, and the capture tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot.

[0167] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0168] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware. Based on such an understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0169] Finally, 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 them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or equivalently replace some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for extracting a target light spot, characterized in that Including: When the capture tracker detects a target light spot in the first image frame captured in the search state, the capture tracker switches from the search state to the protection state and acquires N consecutive second image frames after the first image frame; When the target light spot is detected in all of the N second image frames, obtain the position fluctuation distance between the target light spot in each of the second image frames and the target light spot in the first image frame; where N is a positive integer; When each of the position fluctuation distances is less than a first preset threshold, the capture tracker switches from the protection state to the locking state, and the capture tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot; Wherein, after the step that the capture tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot, the method further includes: When the position fluctuation distances between the target light spots in M consecutive third image frames and the target light spot in the first image frame all exceed a second preset threshold, the capture tracker switches from the locking state to the search state.

2. The target light spot extraction method according to claim 1, wherein The detection method of the target light spot specifically includes: Perform image preprocessing on the captured image frame to obtain a preprocessed image frame; Use a threshold segmentation algorithm to segment the pixels in the preprocessed image frame into multiple different pixel regions; Generate a binary image based on the different pixel regions, and determine the target light spot based on the non-zero pixels in the binary image.

3. The target light spot extraction method according to claim 2, characterized in that, The positioning method of the target light spot in the first image frame includes: Perform gradient calculation on the binary threshold segmentation image corresponding to the first image frame to obtain the gradient magnitude and direction of each pixel point in the binary threshold segmentation image; Perform threshold segmentation on the pixel points in the binary threshold segmentation image according to the gradient magnitude to obtain the binary image corresponding to the first image frame; Based on the connectivity of the image, find the connected components of the non-zero pixels in the binary image to determine the contour points of the target light spot; After converting the contour points into two-dimensional coordinates, use the least squares method to fit the two-dimensional coordinates into an ellipse, obtain ellipse parameters, and determine the first positioning information of the target light spot in the first image frame according to the ellipse parameters.

4. The target light spot extraction method according to claim 3, wherein Obtaining the position fluctuation distance between the target light spot in each of the second image frames and the target light spot in the first image frame includes: Obtain the second positioning information of the target light spot in the second image frame; Based on the first light spot center coordinate in the first positioning information and the second light spot center coordinate in the second positioning information, determine the position fluctuation distance.

5. The target spot extraction method according to claim 2, wherein, The performing image preprocessing on the captured image frame to obtain a preprocessed image frame includes: After performing color gamut change on the image frame, perform Gaussian filtering to obtain the filtered image frame; Adjust the gray level distribution of the filtered image frame by using histogram equalization to obtain a preprocessed image frame.

6. A target light spot extraction device, characterized in that, Including: A switching module, configured to, when a target light spot is detected in a first image frame captured by a tracker in a search state, switch the tracker from the search state to a protection state and obtain N consecutive second image frames after the first image frame; An obtaining module, configured to, when the target light spot is detected in all of the N second image frames, obtain a position fluctuation distance between the target light spot in each of the second image frames and the target light spot in the first image frame; where N is a positive integer; A tracking module, configured to, when each of the position fluctuation distances is less than a first preset threshold, switch the tracker from the protection state to a locking state, and the tracker keeps the target light spot at the center of the field of view and continuously tracks the target light spot; Wherein, the device is further configured to: When a position fluctuation distance between the target light spot in M consecutive third image frames and the target light spot in the first image frame exceeds a second preset threshold, switch the tracker from the locking state to the search state.

7. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the target light spot extraction method according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the target light spot extraction method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the target light spot extraction method according to any one of claims 1 to 5.

Images