Method for quickly retrieving data containing meteor spectral information

Abstract

The invention relates to a method for quickly retrieving data containing meteor spectrum information. The method comprises the following steps: step 1, reading a meteor spectrum video; step 2, background noise is removed; step 3, extracting meteor signals; step 4, filtering isolated noise points by using a convolution denoising method; step 5, morphological restoration of the meteor signal image; step 6, de-noising by a local ring filtering method; step 7, meteor signal enhancement based on Hough transform; 8, realizing efficient noise filtering and signal enhancement by a connected region marking method; and step 9, detecting meteor spectral characteristics. According to the method, a local ring filtering method is combined with an image processing and data analysis algorithm, and weak meteor spectral characteristics are accurately identified, so that automatic screening of spectral signals in meteor data is realized, and the data screening accuracy and the data processing efficiency are improved.

Description

Technical Field

[0001] The invention belongs to the field of data information processing, and is particularly suitable for screening meteor spectral information data. Background Art

[0002] Traditional meteor data processing methods, especially for the extraction and analysis of meteor spectral information, have a series of problems, mainly reflected in low recognition efficiency, complex processing, large errors, and heavy repetitive work. These problems seriously restrict the efficient use of meteor spectral information, especially when a large amount of video data and spectral signals are scarce, which further highlights the shortcomings of traditional methods.

[0003] The processing of meteor spectra mainly involves selecting meteor events with spectral information from a large amount of video data. However, since there are relatively few meteor data containing spectral information, most of the data only contain the location information or brightness information of the light point, but no spectral characteristics. Therefore, when processing such data, it takes a lot of time to screen and preprocess, and the amount of repetitive work is huge. In addition, manual selection is susceptible to fatigue, and it is easy to produce misjudgments and missed judgments, especially when detecting weak spectral signals. The error rate is high, which affects the accuracy of subsequent analysis. At present, the common method of meteor video screening still relies on manual selection, and the average screening time for a video is about 1 minute. In the case of a large amount of video data (such as 1,500 videos), manual selection takes 3 to 4 days. Moreover, manual selection is not only inefficient, but also has a certain risk of error, especially when the spectral signal is weak, it is easy to miss key data, resulting in impaired data quality. These problems have brought many challenges to the subsequent processing and in-depth analysis of the data.

[0004] Therefore, it is particularly important to develop a data screening method that can quickly and accurately identify meteor spectral information. This method should be able to automatically and efficiently identify meteor events containing spectral information from a large amount of video data, reduce manual intervention, and improve the accuracy and efficiency of screening. This is of great significance for scientific research such as meteor detection, trajectory analysis, spectral characteristics research, and physical property analysis of meteoroids. Especially in large-scale meteor monitoring systems, its automated screening capabilities will greatly improve the efficiency of research and the accuracy of results. Summary of the invention

[0005] In view of the shortcomings of the existing technology, a method for quickly retrieving data containing meteor spectral information is proposed, which can automatically detect and optimize a large number of weak meteor spectral signals. By combining background subtraction, inter-frame difference, morphological processing, connected region analysis and adaptive noise suppression, the present invention significantly improves the sensitivity, signal quality and noise resistance of meteor spectral detection, while effectively reducing the false detection rate. These optimization methods not only improve the efficiency and accuracy of data processing, but also ensure the integrity of meteor trajectories, providing a reliable data basis for subsequent material chemical composition analysis.

[0006] The key to the present invention is to use a series of optimization steps for weak meteor spectral signals, including convolution denoising, local ring filtering, Hough transform enhancement and other technologies, combined with advanced image processing methods, to automatically and efficiently extract data containing spectral information from a large number of meteor videos. This method greatly improves the screening efficiency of meteor data and can effectively avoid misjudgments and missed judgments in the traditional manual selection process.

[0007] In a first aspect, the present invention provides a method for quickly retrieving data containing meteor spectral information, comprising the following steps: Step 1. Read the meteor spectrum video Read the meteor spectrum video to be retrieved.

[0008] Step 2. Remove background noise 2.1 Convert the RGB image to a grayscale image; 2.2 Using formula (I) exponential moving average method, calculate the image background: (I) Among them, α is the weighting coefficient, α=0.5; Pic 0 is the image sequence of the meteor video, with size M * N * n , n is the number of video frames, M is the video height, and N is the video width.

[0009] In one embodiment of the present invention, M =576, N =720.

[0010] 2.3 Use formula (II) to deduct background noise: (II) 2.4 Use the median filter to smooth each frame of the image; 2.5 Remove the time and lens information, that is, the pixel values ​​of the 550th to 576th rows of each frame image are set to 0; 2.6 Extract the maximum value of each pixel in all frames to obtain a meteor trail image with complete trails after removing background noise.

[0011] Step 3. Extract meteor signals 3.1 Perform frame-by-frame differential detection on the video signal in step 2, and calculate the difference between each frame (except the first frame) and the previous frame.

[0012] 3.2 Determine the preset threshold ( T diff ) Set the initial threshold to 0.01, gradually increase the threshold, process the image frames without meteors, and gradually remove the pixel values ​​below the current threshold until the number of non-zero pixels in the image is reduced to 50, and get the preset threshold ( T diff ).

[0013] 3.3 Determine the location where the meteor signal changes significantly and generate a binary signal image: compare the difference between the current frame and the previous frame frame by frame. If the difference exceeds the preset threshold ( T diff ), the signal at that position is considered to have changed significantly, and the pixel value at that position is set to 1, otherwise it is set to 0, and a complete binary trajectory signal graph of the meteor is obtained.

[0014] Step 4. Use convolution denoising to filter out isolated noise points In the process of extracting meteor tracks, video frame difference may introduce a large number of isolated noise points. The present invention adopts a convolution denoising method to effectively remove these noises while retaining weak meteor signals.

[0015] Specifically, a convolution kernel of size 20×20 is used to calculate the total number of non-zero pixels in the neighborhood of each pixel position of the complete meteor trail image obtained in step 3. Subsequently, only pixels with a non-zero number of pixels greater than 3 in the neighborhood are retained to filter out isolated noise points, thereby ensuring that the retained signal area has sufficient significance and meets the conditions of the number of neighborhood pixels.

[0016] Step 5. Morphological restoration of meteor signal images In order to improve the quality of meteor signal images, morphological closing operation is used to repair meteor signal images. Morphological closing operation is an image processing technology that combines dilation and erosion. The operation sequence is dilation first and then erosion, filling small holes and connecting adjacent objects. The dilation refers to restoring the target area and dilating the image using structural elements. The erosion refers to eliminating small noise points and corroding the image using structural elements.

[0017] In the specific repair process, the use of disc-shaped structural elements can effectively smooth the image boundaries and fill small gaps and broken areas. After this processing, a meteor signal image is generated. This step significantly improves the image quality, making the meteor signal, especially the weak signal, more continuous and complete, and effectively solves the technical problems of signal loss, distortion, and high error caused by inaccurate extraction of some weak signals. This step uses morphological closing operations mainly to eliminate small noise points, fill small gaps, and maintain the overall shape of the meteor signal, making the edge smoother and reducing the impact of nonlinear transformation on subsequent processing. This step does not introduce the Hough transform too early, which reduces the loss of local features and avoids the neglect of shorter or weaker meteor signals. In addition, directly applying the Hough transform in the case of more noise may detect false straight lines, affecting the effectiveness of subsequent enhancement operations. Therefore, performing morphological repair first helps to retain signal integrity and provide more accurate input for the Hough transform in the subsequent step 7.

[0018] Step 6: Local ring filter denoising In order to effectively remove noise while retaining weak meteor signals, the present invention proposes a local annular filtering method, which detects signals in the annular area to determine whether to remove the noise in the core area, and applies it to the denoising process of the repaired image in step 5 to achieve accurate denoising.

[0019] The local annular filtering method includes: generating grid coordinates in a Cartesian coordinate system, creating annular and core area masks, and signal detection and core area update: 6.1 Generate grid coordinates in Cartesian coordinate system X , Y Set the core area radius r 1 and the outer radius of the annular area r 2. In the grid range from - r 2 to r 2 in the discrete Cartesian coordinate system, the horizontal and vertical coordinates x i and y j As shown in formula (III-a): (III-a) Grid X , Y Each element of represents the coordinates of the corresponding position: (III-b).

[0020] 6.2 Creating Ring Mask and Core Region Mask The Euclidean distance d from each point to the center is calculated using formula (IV): (IV) Create a ring mask using formula (V) Mask ring and core area mask Mask core : (V).

[0021] 6.3 Signal Monitoring and Update Define a local window for each pixel of the image obtained in step 5 W ( x , y ), the range is ( x,y ) is the center and the radius is r 2 Area, window W ( x , y ) by the ring mask Mask ring and core area mask Mask core By matching the local window with the mask, for any window, execute condition (VI): (VI) By detecting the ring mask around each pixel Mask ring If the signal strength is less than 20, determine whether to mask the core area where the pixel is located. Mask core Set to zero, that is Mask ring If the signal strength is less than 20, Mask core Set to zero to remove noise.

[0022] The denoised image generated by the local annular filtering process of the present invention can not only effectively remove noise, but also retain the weak meteor spectrum signal to the maximum extent, thereby avoiding the possible mistaken deletion phenomenon in the traditional method. The advantage of the local annular filtering method is that it specifically processes the signal changes in the local area, rather than simply uniformly processing the entire image.

[0023] By creating a local annular area and performing signal detection on it, the present invention can accurately identify and retain weak meteor signals in the image, while effectively filtering out irrelevant interference noise. Compared with traditional image denoising methods (such as mean filtering and median filtering), the annular filtering method pays more attention to the signal changes of local features of the image and can process signals more finely. In mean filtering, all pixels in the image are averaged, which easily leads to the loss of weak signals; while median filtering can better remove isolated noise, it is less sensitive to continuously changing weak signals. In contrast, the annular filtering method ensures the integrity and accuracy of the signal by dynamically adjusting the processing method of the local area, and is particularly suitable for processing images containing weak signals.

[0024] Therefore, the local annular filtering method of the present invention can significantly improve the retention effect of weak meteor signals while denoising, avoids the accidental deletion of key signals in traditional methods, improves the accuracy and reliability of signal extraction, and has important application value for the accurate analysis of meteor spectral signals.

[0025] Step 7. Meteor signal enhancement based on Hough transform The present invention uses Hough transform to detect and locate straight lines in the image, combines morphological operations to repair breaks or defects along the straight line direction, and uses morphological closing operations to repair and enhance meteor signals, further improving the coherence and integrity of meteor signals. The specific steps are as follows: 7.1 Detect straight lines: Perform Hough transform on the image obtained in step 6 to detect and locate the straight lines therein, including their position, direction and length; 7.2 Constructing structural elements: According to the detected direction and length of the straight line, construct the structural elements of the corresponding morphological dilation operation, so that the dilation operation is performed along the straight line direction to enhance the continuity of the straight line; 7.3 Morphological Closing Operation: Use disk-shaped structure elements to perform morphological closing operations to further repair and enhance meteor signals; 7.4 Repeat processing: To ensure the continuity and repair effect of the straight line, repeat the above expansion and closing operation steps 3-10 times until the desired effect is achieved.

[0026] The closed operation based on Hough transform in the present invention can repair the directional characteristics of meteor signals in a targeted manner, enhance the connectivity along the direction of the meteor trajectory, effectively enhance the coherence of meteor signals, repair the breaks or defects along the straight line direction, and improve the accuracy of signal detection. At the same time, it removes noise inconsistent with the meteor trajectory.

[0027] Step 8. Efficient noise filtering and signal enhancement by connected component labeling In the signal enhancement process of step 7, the noise area may be accidentally enlarged, so further denoising is required to improve the accuracy of the result. The present invention processes the image of step 7 by the connected region labeling method, and the specific steps are as follows: 8.1 Identify all connected regions in the image through the connected region labeling method and analyze their characteristics one by one; 8.2 For each connected region, calculate the sum of its pixel intensities. If the sum of intensities is lower than the specified threshold ( T area =50), then all pixel values ​​in the area are set to 0 to remove the areas that do not meet the conditions; 8.3 Retain the area where the total pixel intensity is higher than the threshold, so as to ensure that only the valid area with large signal intensity is retained. After this step of processing, the reliability and effectiveness of the signal are further improved.

[0028] Step 9. Detect whether there are meteor spectral features: including Hough transform detection straight line, extraction of meteor signal coordinates, linear fitting steps, identification of the number of connected regions through the connected region marking method, combined with connected region analysis, linear fitting parameters, fitting residuals and errors, to determine the meteor spectral signal. After completing all denoising and signal enhancement processing, it is possible to detect whether the video contains meteor spectral signals. The specific process is as follows: 9.1 Retrieve the positions with pixel value 1 in the image obtained in step 8 and count the number of these positions. If the number of detected pixels is less than or equal to 100, it means that no meteor signal is detected; if the number is greater than 100, continue with the subsequent processing; 9.2 Use Hough transform to detect straight lines (i.e. meteor trails), and crop the image according to the vertical position of the straight line to generate a sub-image containing the meteor trail area to improve the accuracy of subsequent processing; 9.3 Extract the coordinates of the pixel value 1 in the sub-image containing the meteor track area ( x,y ), as the coordinates of the meteor signal, where x and y They are the horizontal and vertical coordinates of the image respectively; 9.4 Use the least squares method to calculate the signal pixel coordinates ( x,y ) for linear fitting y = ax + b , get the fitting parameters and correlation coefficients r , fitting residual R and error δ ; The fitting parameters include a and b, and the fitting theoretical value of each data point is calculated through the fitting parameters, and the fitting error / residual is calculated based on this.

[0029] 9.5 Use the connected component labeling method to identify the number of connected components in the cropped image in step 9.2 N ; 9.6 Define the objective function as shown in formula (VII): (VII) in, , , and σ are the fitting residuals R The mean and standard deviation of , X is the design matrix, X=[1,1,…1;x1,x2,…,x n ] T , x i Represents the horizontal coordinates of the signal pixel, and diag() means taking the diagonal elements of the matrix.

[0030] In one embodiment of the present invention, the parameters are set N min =2, N max =20, r max =0.7,σ min =50, δ min =50.

[0031] Calculate the objective function J , determine whether the video data contains meteor spectral features.

[0032] Only when the objective function J =0 or J =1, the video data contains the spectral characteristics of meteors.

[0033] Furthermore, when J = 0, indicating that the video data contains meteor spectral features, and the following conditions are met at the same time: ① The number of connected regions 2≤N<20; ② The correlation coefficient r <0.7; ③Standard deviation of fitting residual σ≥10; ④Error δ ≥10, the video data may contain zero-order spectral features.

[0034] The zero-order spectrum is the luminous trajectory of the meteor itself. Generally speaking, the spectrum obtained by grating spectrometry is far away from the meteor trajectory. The zero-order spectrum can be used to determine the direction of the meteor spectrum, facilitating the subsequent extraction of meteor spectrum information.

[0035] when J =1, indicating the number of connected regions N= 1, although the video data still contains meteor spectral features, it does not contain zero-order spectral features. The specification limit on the number of connected regions is to control the interference of noise or non-meteor spectra.

[0036] In the process of identifying the spectral characteristics of meteors, the present invention constructs an independently designed objective function J, which comprehensively considers factors such as connected area analysis, linear fitting coefficients, fitting residuals and errors. By combining a variety of image processing techniques such as Hough transform, connected area marking, least squares fitting and statistical characteristic analysis, the judgment mechanism can efficiently detect spectral signals, including weak signals. This allows for effective and accurate detection of data containing weak spectral signals, solving the technical problem of large errors in weak signal identification.

[0037] In a second aspect, the method provided by the present invention for rapidly retrieving data containing meteor spectral information is used in the fields of meteor monitoring, meteor shower research, solar system formation and evolution research, etc.

[0038] Beneficial effects of the present invention: 1. The present invention combines the local annular filtering method with image processing and data analysis algorithms. Through a series of efficient data processing algorithms, it automatically screens out meteor data with spectral information and accurately identifies the spectral characteristics of meteors. This method realizes the automatic screening of spectral signals in meteor data, which not only improves the accuracy of data screening, but also greatly reduces the time and workload of manual processing. The entire retrieval process takes only 2-6s, which significantly improves the data processing efficiency.

[0039] 2. The innovative local annular filtering method of the present invention can accurately denoise the core area of ​​the image and effectively retain weak meteor signals. The advantage of this method is that it can flexibly adapt to the changes in local features in the image, avoiding the problem of weak signal loss due to excessive smoothing in traditional mean filtering and median filtering methods; in addition, the straight line features in the image are detected by combining the Hough transform, and the missing or broken parts along the straight line are repaired through morphological dilation and closing operations, which significantly improves the integrity and continuity of the meteor signal, thereby ensuring the accurate extraction of weak signals. This technological breakthrough is not only of great significance in meteor monitoring, but also provides a reliable technical means for the extraction and analysis of weak spectral signals.

[0040] 3. The objective function J is constructed in the process of identifying the spectral characteristics of meteors, taking into account factors such as connected region analysis, linear fitting coefficient, fitting residual and error. By combining a variety of image processing techniques such as Hough transform, connected region labeling, least squares fitting and statistical characteristic analysis, an efficient determination mechanism is formed, which helps to accurately detect spectral signals.

[0041] 4. The identification method of the present invention uses a small amount of multiple denoising to retain weak spectrum / meteor signals and avoid the loss and accidental deletion of key signals. The method has strong versatility for weak signal data and can be applied to meteor data of different types and sources after appropriate adjustment and optimization; at the same time, it has a high degree of automation and small error, and is suitable for wide promotion and application.

[0042] 5. The present invention can be widely used in the fields of meteor monitoring, meteor shower research, solar system formation and evolution research, etc., providing more efficient and accurate data support for related scientific research, and promoting the scientific community to further explore meteors and their spectral characteristics. In particular, in large-scale meteor monitoring systems and meteor spectrum analysis projects, the application of this technology can significantly improve research efficiency and promote the development of related disciplines such as meteorology and space physics. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 The original image of the meteor sample of Example 1, wherein (a) the original image containing the complete meteor track; (b) the image containing the complete track after removing the background noise; Figure 2 The binary image containing the meteor signal extracted in Example 1; Figure 3 This is the meteor trajectory diagram after filtering out isolated noise points in Example 1 ( FilteredImage ); Figure 4 The meteor track image after morphological restoration in Example 1 ( RepairedSignalMap ); Figure 5 The image after denoising by the annular local annular filtering method in Example 1 ( FilteredSignalMap ); Figure 6 is the enhanced meteor signal image of Example 1 ( EnhancedSignalMap ); Figure 7 The image after the connected region marking and denoising in Example 1 ( CleanedSignalMap ); Figure 8 The linear fitting result of the meteor signal area in Example 1; Fig. 9The figure is a diagram of the spectral information recognition process and results of the meteor (20191220_210801) in Example 2, wherein (a) is the original image, (b) is the image after removing the background noise, (c) is the meteor signal extraction image, (d) is the noise point filtered image, (e) is the morphological restoration image, (f) is the local filtering denoised image, (g) is the enhanced meteor signal image, (h) is the connected area marking denoised image, and (i) is the linear fitting result of the meteor signal area; Fig.10 The figure is a diagram of the spectral information recognition process and results of the meteor (20210211_165334) in Example 3, where (a) is the original image, (b) is the image after removing the background noise, (c) is the meteor signal extraction image, (d) is the noise point filtered image, (e) is the morphological restoration image, (f) is the local filtering denoising image, (g) is the enhanced meteor signal image, (h) is the connected area marking denoising image, and (i) is the linear fitting result of the meteor signal area; Fig.11 The figure is a diagram of the spectral information recognition process and results of the meteor (20230129_173228) in Example 4, wherein (a) is the original image, (b) is the image after removing the background noise, (c) is the meteor signal extraction image, (d) is the noise point filtered image, (e) is the morphological restoration image, (f) is the local filtering denoised image, (g) is the enhanced meteor signal image, (h) is the connected area marking denoised image, and (i) is the linear fitting result of the meteor signal area; Fig.12 The spectral information recognition process and result diagram of the meteor (20210912_173638) in Example 5 are shown, where (a) is the original image, (b) is the image after background noise removal, (c) is the meteor signal extraction image, (d) is the noise point filtered image, (e) is the morphological restoration image, (f) is the local filtering denoised image, (g) is the enhanced meteor signal image, (h) is the connected area marked denoised image, and (i) is the linear fitting result of the meteor signal area; Fig.13 The figure is a diagram of the spectral information recognition process and results of the meteor (20220102_153646) in Example 6, wherein (a) is the original image, (b) is the image after removing the background noise, (c) is the meteor signal extraction image, (d) is the noise point filtered image, (e) is the morphological restoration image, (f) is the local filtering denoised image, (g) is the enhanced meteor signal image, (h) is the connected area marking denoised image, and (i) is the linear fitting result of the meteor signal area; Fig.14The figure shows the spectral information recognition process and results of the meteor (20231010_204157) in Example 7, where (a) is the original image, (b) is the image after removing background noise, (c) is the meteor signal extraction image, (d) is the noise point filtered image, (e) is the morphological restoration image, (f) is the local filtering denoised image, (g) is the enhanced meteor signal image, (h) is the connected area marked denoised image, and (i) is the linear fitting result of the meteor signal area. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0046] Example 1 Step 1. Read video data: Read the meteor spectrum video to be retrieved.

[0047] Step 2. Remove background noise interference 2.1 Convert the RGB image to a grayscale image; 2.2 Calculate image background using exponential moving average method: (I) Among them, α is the weighting coefficient, α=0.5; Pic 0 is the image sequence of the meteor video, with size M * N * n ( M =576, N =720, n is the number of frames of the video).

[0048] 2.3 Subtract background noise: (II) 2.4 Use a median filter to smooth each frame of the image with a window size of [2, 2].

[0049] 2.5 Remove the time and lens information, that is, the pixel values ​​of the 550th to 576th rows of each frame are set to 0. By extracting the maximum value of each pixel in all frames, a complete meteor trail image can be obtained, as shown in Figure 1.

[0050] Step 3. Extract meteor signals Perform frame-by-frame differential detection on the video signal and calculate the difference between each frame (except the first frame) and the previous frame. T diff ) and mark it as 1 to indicate the location where the signal changes significantly, thereby generating a binary signal image. T diff The determination is made by gradually increasing the threshold, processing the image frames without meteors (such as the tenth frame of the video), and gradually removing the pixel values ​​below the current threshold until the number of non-zero pixels in the image is reduced to 50. In all the frames processed, the maximum signal value of each pixel position is extracted to generate an image containing the maximum signal intensity. MaxSignalMap ( M × N ),like Figure 2 shown.

[0051] Step 4. Use convolution denoising to filter out isolated noise points In the process of extracting meteor trails, video frame difference may introduce a large number of isolated noise points. In order to effectively remove these noises while retaining weak meteor signals, the convolution denoising method is used: a convolution kernel of size 20×20 is used to calculate the image containing the complete meteor trail MaxSignalMap The total number of non-zero pixels in the neighborhood of each pixel position. Subsequently, only pixels with a non-zero number of pixels in the neighborhood greater than 3 are retained to filter out isolated noise points, thereby ensuring that the retained signal area has sufficient significance and meets the conditions of the number of neighborhood pixels. The image after this denoising process is shown in Figure 1. Figure 3 As shown, it is recorded as FilteredImage .

[0052] Step 5. Morphological restoration of meteor signal images To improve the meteor signal image FilteredImage The quality of the image was improved by using morphological closing operation. Morphological closing operation is an image processing technology that combines dilation and erosion. The operation sequence is dilation first and then erosion. The repair process uses a disk-shaped structure element with a radius of 6, which can effectively smooth the image boundary and fill small pores and broken areas. The repaired image generated by this process significantly improves the image quality and makes the meteor signal, especially the weak signal, more continuous and complete. Figure 4 As shown, it is recorded as RepairedSignalMap .

[0053] Step 6: Local ring filter denoising In order to effectively remove noise while retaining weak meteor signals, a local ring filter method is used and applied to the repaired image. RepairedSignalMapDe-noising. This method detects the signal in the annular area and determines whether to remove the noise in the core area to achieve accurate denoising. The specific steps include generating grid coordinates in the Cartesian coordinate system, creating annular and core area masks, signal detection and core area update, as described below: 6.1 Generate grid coordinates in Cartesian coordinate system X , Y Set the core area radius r 1=2 and the outer radius of the annular area r 2=8, in the grid range from - r 2 to r 2 in the discrete Cartesian coordinate system, the horizontal and vertical coordinates x i and y j for: (III-a) Then, the grid X , Y Each element of represents the coordinates of the corresponding position: (III-b).

[0054] 6.2 Creating Ring Mask and Core Region Mask Compute the Euclidean distance from each point to the center: (IV) Creating a Ring Mask Mask ring and core area mask Mask core : (V).

[0055] 6.3 Signal Monitoring and Update Traversing the image RepairedSignalMap For each pixel, define a local window W ( x , y ), whose range is the area centered at (x, y) with a radius of r2, and the window W ( x , y ) by the ring mask Mask ring and core area mask Mask core By matching the local window with the mask, for any window, execute condition (VI): (VI) This step detects a ring mask around each pixel. Mask ring If the signal strength is less than 20, determine whether to mask the core area where the pixel is located. Mask core After local ring filtering, the generated denoised image is as follows: Figure 5 As shown, it is recorded as FilteredSignalMap This method can not only effectively remove noise, but also retain weak spectral signals to the maximum extent and avoid accidental deletion.

[0056] Step 7. Meteor signal enhancement based on Hough transform In order to improve the coherence and integrity of the meteor signal, Hough transform is used to detect the straight lines in the image, and morphological operations are combined to repair the breaks or defects along the straight line. The specific steps are as follows: 7.1 Line Detection: Image FilteredSignalMap Apply Hough transform to detect and locate the straight lines, including their position, direction and length; 7.2 Structural element construction: According to the detected straight line direction and length, the structural element of the corresponding morphological dilation operation is constructed so that the dilation operation is performed along the straight line direction to enhance the continuity of the straight line; 7.3 Morphological closing operation: A disk-shaped structure element with a radius of 8 is used to perform morphological closing operation to further repair and enhance the meteor signal; 7.4 Repeat processing: To ensure the continuity and repair effect of the straight line, repeat the above expansion and closing operation steps until the expected effect is achieved.

[0057] The image after this processing is recorded as EnhancedSignalMap ,like Figure 6 This method effectively enhances the continuity of meteor signals, repairs breaks or defects along the straight line, and improves the accuracy of signal detection.

[0058] Step 8. Efficient noise filtering and signal enhancement by connected component labeling During the signal enhancement process, noisy areas may be accidentally amplified, so further denoising is required to improve the accuracy of the results. EnhancedSignalMap To process, the specific steps are as follows: 8.1 Identify all connected regions in the image through the connected region labeling method and analyze their characteristics one by one; 8.2 For each connected region, calculate the sum of its pixel intensities. If the sum of intensities is lower than the specified threshold ( T area =50), then all pixel values ​​in the area are set to 0 to remove the areas that do not meet the conditions; 8.3 Only the areas where the sum of pixel intensities is higher than the threshold are retained, thereby ensuring that only valid areas with large signal intensity are retained.

[0059] After this step, the resulting denoised image is marked as CleanedSignalMap ,like Figure 7 As shown, the reliability and effectiveness of the signal are further improved.

[0060] Step 9. Check whether there are spectral features After all the denoising and signal enhancement processing is completed, it is possible to detect whether the video contains spectral signals. The specific process is as follows: 9.1 First, retrieve the image CleanedSignalMap The positions where the pixel value is 1 in the image are detected, and the number of these positions is calculated. If the number of detected pixels is less than or equal to 100, it means that no meteor signal is detected; if the number is greater than 100, continue with the subsequent processing; 9.2 Use Hough transform to detect straight lines (i.e. meteor trails), and crop the image according to the vertical position of the straight line to generate a sub-image containing the meteor trail area MeteorTrackRegion , in order to improve the accuracy of subsequent processing; 9.3 Extracting images MeteorTrackRegion The coordinates of the pixels with a value of 1 ( x,y ), as the coordinates of the meteor signal, where x and y They are the horizontal and vertical coordinates of the image respectively; 9.4 Use the least squares method to calculate the signal pixel coordinates ( x,y ) for linear fitting y = ax + b , get the fitting parameters and correlation coefficients r , fitting residual R and error δ ; The fitting parameters include a and b, and the fitting theoretical value of each data point is calculated through the fitting parameters, and the fitting error / residual is calculated based on this.

[0061] 9.5 Image Recognition Using Connected Component Labeling MeteorTrackRegion The number of connected regions N ; 9.6 Defining the Objective Function J , as shown in formula (VII): (VII) in, , , and σ are the fitting residuals R The mean and standard deviation of . , X is the design matrix, X=[1,1,…1;x1,x2,…,x n ] T , x i Indicates the horizontal coordinates of the signal pixel, and diag() means taking the diagonal elements of the matrix. Setting parameters N min =2, N max =20, r max =0.7,σ min =50, δ min =50.

[0062] The fitting results of this example are as follows Figure 8 show J = 0, the video data contains the spectral features of meteors. Manual selection is used for verification, which takes 2-3 days. The results of the two methods are consistent, verifying the accuracy of the spectral data processing method of this embodiment.

[0063] Example 2 According to steps 1-9 of Example 1, the meteor video case 20191220_210801 is processed, and the spectrum recognition process and results are shown in the figure. Fig. 9 As shown, (a) is the original image obtained in step 1, (b) is the image after removing the background noise in step 2, (c) is the meteor signal extraction image in step 3, (d) is the noise point filtered image in step 4, (e) is the morphological restoration image in step 5, (f) is the local filtering denoising image in step 6, (g) is the enhanced meteor signal image in step 7, (h) is the connected area marking denoising image in step 8; (i) is the linear fitting result of the meteor signal area in step 9.

[0064] The fitting results of this example show J =0, the video data contains meteor spectral features and may contain zero-order spectra; Manual selection was used for verification simultaneously, and the manual selection took 2-3 days. The results of the two methods were consistent, which verified the accuracy of the spectral data processing method of this embodiment.

[0065] Example 3 The meteor video case 20210211_165334 is processed according to steps 1-9 of Example 1. The spectrum recognition process and results are shown in the figure below. Fig.10As shown, (a) is the original image obtained in step 1, (b) is the image after removing the background noise in step 2, (c) is the meteor signal extraction image in step 3, (d) is the noise point filtered image in step 4, (e) is the morphological restoration image in step 5, (f) is the local filtering denoising image in step 6, (g) is the enhanced meteor signal image in step 7, (h) is the connected area marking denoising image in step 8; (i) is the linear fitting result of the meteor signal area in step 9.

[0066] The fitting results of this example show J =0, the video data contains meteor spectral features and may contain zero-order spectra.

[0067] Manual selection was used for verification simultaneously, and the manual selection took 2-3 days. The results of the two methods were consistent, which verified the accuracy of the spectral data processing method of this embodiment.

[0068] Example 4 The meteor video case 20230129_173228 is processed according to steps 1-9 of Example 1. The spectrum recognition process and results are shown in the figure below. Fig.11 As shown, (a) is the original image obtained in step 1, (b) is the image after removing the background noise in step 2, (c) is the meteor signal extraction image in step 3, (d) is the noise point filtered image in step 4, (e) is the morphological restoration image in step 5, (f) is the local filtering denoising image in step 6, (g) is the enhanced meteor signal image in step 7, (h) is the connected area marking denoising image in step 8; (i) is the linear fitting result of the meteor signal area in step 9.

[0069] The fitting results of this example show J =1, indicating the number of connected regions N =1, although the video data still contains the meteor spectrum feature, it does not contain the zero-order spectrum feature at this time. Manual selection is used for verification simultaneously, and manual selection takes 2-3 days. The results of the two methods are consistent, which verifies the accuracy of the spectrum data processing method of this embodiment.

[0070] Example 5 The meteor video case 20210912_173638 is processed according to steps 1-9 of Example 1. The spectrum recognition process and results are shown in the figure below. Fig.12As shown, (a) is the original image obtained in step 1, (b) is the image after removing the background noise in step 2, (c) is the meteor signal extraction image in step 3, (d) is the noise point filtered image in step 4, (e) is the morphological restoration image in step 5, (f) is the local filtering denoising image in step 6, (g) is the enhanced meteor signal image in step 7, (h) is the connected area marking denoising image in step 8; (i) is the linear fitting result of the meteor signal area in step 9.

[0071] The fitting result of this embodiment shows J=2.73, and the video data does not contain the spectral features of meteors. Manual selection is used for verification, which takes 2-3 days. The results of the two methods are consistent, which verifies the accuracy of the spectral data processing method of this embodiment.

[0072] Example 6 The meteor video case 20220102_153646 is processed according to steps 1-9 of Example 1. The spectrum recognition process and results are shown in the figure below. Fig.13 As shown, (a) is the original image obtained in step 1, (b) is the image after removing the background noise in step 2, (c) is the meteor signal extraction image in step 3, (d) is the noise point filtered image in step 4, (e) is the morphological restoration image in step 5, (f) is the local filtering denoising image in step 6, (g) is the enhanced meteor signal image in step 7, (h) is the connected area marking denoising image in step 8; (i) is the linear fitting result of the meteor signal area in step 9.

[0073] The fitting result of this embodiment shows J=34.23, and the video data does not contain the spectral characteristics of meteors. Manual selection is used for verification, which takes 2-3 days. The results of the two methods are consistent, which verifies the accuracy of the spectral data processing method of this embodiment.

[0074] Example 7 The meteor video case 20231010_204157 is processed according to steps 1-9 of Example 1. The spectrum recognition process and results are shown in the figure below. Fig.14 As shown, (a) is the original image obtained in step 1, (b) is the image after removing the background noise in step 2, (c) is the meteor signal extraction image in step 3, (d) is the noise point filtered image in step 4, (e) is the morphological restoration image in step 5, (f) is the local filtering denoising image in step 6, (g) is the enhanced meteor signal image in step 7, (h) is the connected area marking denoising image in step 8; (i) is the linear fitting result of the meteor signal area in step 9.

[0075] The fitting result of this embodiment shows J=99.98, and the video data does not contain the spectral features of meteors. Manual selection is used for verification, which takes 2-3 days. The results of the two methods are consistent, verifying the accuracy of the spectral data processing method of this embodiment.

[0076] In summary, the present invention adopts an innovative local annular filtering method, combined with image processing and data analysis algorithms, and with the help of a series of efficient data processing algorithms, automatically screens out meteor data with spectral information, significantly improving the efficiency and accuracy of meteor spectrum screening. The local annular filtering method of the present invention can flexibly respond to local changes in meteor signals, accurately remove interference noise in the image, and retain weak meteor spectral signals. Compared with traditional denoising methods, this method avoids signal loss caused by excessive smoothing while retaining important signals, and has significant advantages in the extraction of weak signals.

[0077] In addition, the present invention combines image processing techniques such as Hough transform, morphological operations and connected region analysis to effectively improve the coherence and integrity of meteor trajectories. These technical means not only enhance the recognizability of meteor signals, but also reduce the problem of data loss caused by signal breakage or loss. Through these optimizations, the present invention can accurately identify meteor signals in complex environments and provide reliable basic data for subsequent scientific analysis.

[0078] It is worth mentioning that the entire retrieval process of the present invention only takes 2-6 seconds, which is much faster than the traditional manual screening method, greatly reducing the complexity and labor intensity of manual operation. At the same time, the automated screening system can eliminate errors in manual operation, greatly improving the accuracy of data screening, especially in large-scale data processing, showing more significant advantages. For scientific research institutions and meteor research projects that need to process a large amount of meteor monitoring video or image data, this technology not only improves work efficiency, but also effectively saves manpower and time costs.

[0079] In general, the application of this invention in the field of meteor spectrum screening not only provides an efficient and accurate solution for meteor monitoring, but also provides strong data support for related scientific research. Its efficient automatic screening, powerful signal retention ability and simple operation process make meteor data processing more intelligent and promote the research progress of meteorology and other astronomical fields.

[0080] The above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for quickly retrieving data containing meteor spectrum information, characterized in that: The steps include: Step 1. Read the meteor spectrum video; Step 2. Remove background noise; Step 3. Extract meteor signals; Step 4. Use convolution denoising method to filter out isolated noise points; Step 5. Use morphological closing operation to repair the meteor signal image; Step 6. Local annular filtering method for denoising: including generating grid coordinates in a Cartesian coordinate system, creating annular and core area masks, and signal detection and core area update; Step 7. Meteor signal enhancement based on Hough transform; Step 8. Connected region labeling method is used to achieve efficient noise filtering and signal enhancement; Step 9. Detect meteor spectral features: including Hough transform detection straight line, extraction of meteor signal coordinates, linear fitting steps, and identification of the number of connected regions through the connected region marking method, combined with connected region analysis, linear fitting parameters, fitting residuals and errors, to determine the meteor spectral signal.

2. The method according to claim 1, characterized in that Step 2 includes the following steps: 2.1 Convert the RGB image to a grayscale image; 2.2 Using formula (I) exponential moving average method, calculate the image background: （I） 2.3 Use formula (II) to deduct background noise: (II) 2.4 Use the median filter to smooth each frame of the image; 2.5 Remove time and lens information; 2.6 Extract the maximum value of each pixel in all frames to obtain a meteor trail image with complete trails after removing background noise.

3. The method according to claim 1, characterized in that Step 3 includes the following steps: 3.1 Perform frame-by-frame differential detection on the video signal in step 2, and calculate the difference between each frame and the previous frame; 3.2 Determine the preset threshold ( T diff ); 3.3 Determine the location where the meteor signal changes significantly and generate a binary signal image.

4. The method according to claim 1, characterized in that: Step 4 uses a convolution kernel of size 20×20 to calculate the total number of non-zero pixels in the neighborhood of each pixel position in the image obtained in step 3, retaining only pixels with a non-zero pixel number greater than 3 in the neighborhood to filter out isolated noise points.

5. The method according to claim 1, characterized in that Step 5 adopts the operation sequence of dilation followed by erosion, using disk-shaped structural elements to effectively smooth the image boundaries and fill small gaps and broken areas.

6. The method according to claim 1, characterized in that Step 6 includes: 6.1 Generate grid coordinates in Cartesian coordinate system X , Y : Set the core area radius r 1 and the outer radius of the annular area r 2. In the grid range from - r 2 to r 2 in the discrete Cartesian coordinate system, the horizontal and vertical coordinates x i and y j As shown in formula (III-a): (III-a) Each element of the grid represents the coordinates of the corresponding position as shown in formula (III-b): (III-b)。 7. The method according to claim 6, characterized in that Step 6 also includes: 6.2 Create annular mask and core area mask: The Euclidean distance d from each point to the center is calculated using formula (IV): (IV) Create a ring mask using formula (V) Mask ring and core area mask Mask core : （V） 6.3 Signal monitoring and updating: Define a local window for each pixel of the image obtained in step 5 W ( x , y ), the range is ( x,y ) is the center and the radius is r 2 Area, window W ( x , y ) by the ring mask Mask ring and core area mask Mask core Composition; match the local window with the mask, for any window, execute condition (VI): (WE) Detect ring mask around each pixel Mask ring If the signal strength is less than 20, determine whether to mask the core area where the pixel is located. Mask core Set to zero to remove noise.

8. The method according to claim 1, characterized in that Step 7 includes: detecting and locating straight lines in the image by Hough transform, repairing the breaks or defects along the straight lines by combining morphological operations, and repairing and enhancing meteor signals by using morphological closing operations.

9. The method according to claim 1, characterized in that: Step 8 includes: 8.1 Identify all connected regions in the image through the connected region labeling method and analyze them one by one; 8.2 For each connected region, calculate the sum of its pixel intensities; if the sum of intensities is lower than the specified threshold, set all pixel values ​​in the region to 0 and remove the unqualified regions; 8.3 Retain the regions where the sum of pixel intensities is above a specified threshold.

10. The method according to claim 1, characterized in that Step 9 includes: 9.1 Retrieve the positions where the pixel value is 1 in the image obtained in step 8, and count the number of said positions; 9.2 Use Hough transform to detect straight lines, and crop the image according to the vertical position of the straight line to generate a sub-image containing the meteor track area; 9.3 Extract the coordinates of the pixel value 1 in the sub-image containing the meteor track area ( x,y ), as the coordinates of the meteor signal, where x and y They are the horizontal and vertical coordinates of the image respectively; 9.4 Use the least squares method to calculate the signal pixel coordinates ( x,y ) for linear fitting y = ax + b , get the fitting parameters and correlation coefficients r , fitting residual R and error δ ; 9.5 Use the connected component labeling method to identify the number of connected components in the cropped image in step 9.2 N ; 9.6 Calculate the objective function J according to formula (VII) to determine whether the video data contains meteor spectral features: (VII) in, , , and σ are the fitting residuals R The mean and standard deviation of , X is the design matrix, X=[1,1,…1;x1,x2,…,x n ] T , x i Represents the horizontal coordinates of the signal pixel, and diag() means taking the diagonal elements of the matrix; When the objective function J =0 or J =1, the video data contains the spectral characteristics of meteors.

Images