Gravity measurement downsampling method and device optimized by edge detection technology, equipment and medium

Abstract

The invention discloses a gravity measurement downsampling method, a gravity measurement downsampling device, gravity measurement downsampling equipment and a medium optimized by adopting an edge detection technology, belongs to the technical field of gravity measurement, and provides a gravity measurement downsampling method optimized by adopting the edge detection technology by improving downsampling processing adopting a mean filter. Evaluating the change degree of the gravity field by using an edge detection technology, and in an area where the gravity field changes severely, reducing the length of a gravity measurement data sequence subjected to mean filtering, and retaining more changed gravity measurement signals; in an area where the gravitational field changes gently, the length of a gravitational measurement data sequence subjected to mean filtering is increased, so that the filtering result is smoother, the data volume after downsampling is smaller, and the subsequent calculation amount is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of gravity measurement technology, and more specifically, relates to a gravity measurement downsampling method, device, equipment and medium optimized by edge detection technology. Background Art

[0002] Sea and air gravimeters can measure gravity information in a large range and with high efficiency under the condition of a dynamic base. Since the 1980s, with the maturity of satellite navigation technology, inertial device technology, and inertial navigation system technology, sea and air gravimeters have made great contributions in resource exploration, gravity mapping, etc. However, due to the limitation of the measurement mechanism, during the gravity measurement process, the original signal contains a large amount of high-frequency noise and external disturbance noise. Therefore, in engineering applications, a low-pass filter is generally used to reduce the influence of noise on the results of gravity measurement. Since the original sensor signal usually has a high-frequency sampling frequency of 50 to 200 Hz, using this high-frequency signal to directly filter and solve will not only seriously consume the processor computing resources, but also fail to correctly reflect the low-frequency gravity information, and will also be unfavorable for the attenuation and suppression of high-frequency disturbances. Therefore, it is necessary to use downsampling technology to process the original data of the sensor, reduce the sampling frequency, reduce the amount of calculation, and reduce high-frequency noise.

[0003] Downsampling is essentially a low-pass filter. While lowering the sampling frequency and reducing high-frequency noise, it also removes some changing gravity measurement signals. Therefore, in some areas where the gravity field changes rapidly, downsampling will distort the gravity measurement signal and reduce the accuracy of gravity measurement. Summary of the invention

[0004] In view of the above defects or improvement needs of the prior art, in order to enable the raw data of gravity measurement to retain the changed gravity value after downsampling processing, the present invention uses edge detection technology to evaluate the degree of change of the gravity field, optimizes the length of the downsampling sequence according to the degree of change of the gravity field, and then retains the changed gravity signal in the area where the gravity field changes drastically, and outputs a smoother downsampling result with a smaller data volume in the area where the gravity field changes gently.

[0005] To achieve the above object, according to one aspect of the present invention, a gravity measurement downsampling method optimized by edge detection technology is provided, comprising:

[0006] The gravity reference image is converted into a grayscale image by using edge detection, and the grayscale image is scanned to calculate the gradient sum of the grayscale image pixels;

[0007] Based on the gradient and the sequence length of the optimized mean filter, the raw data of gravity measurement is downsampled using the optimized mean filter. The larger the gradient and the greater the change in the gravity field, the shorter the corresponding filter sequence.

[0008] In some optional implementations, the step of converting the gravity reference image into a grayscale image by edge detection, scanning the grayscale image and calculating the gradient sum of the grayscale image pixels includes:

[0009] Divide the gravity reference image into a number of evaluation areas, and convert the gravity reference image into a grayscale image;

[0010] The grayscale image is scanned by the Sobel operator template to obtain the gradient sum of the grayscale image pixels in each evaluation area.

[0011] In some optional implementations, after obtaining the gradient sum of the grayscale image pixels in each evaluation area, the method further includes:

[0012] The gradient sum of the grayscale image pixels in each evaluation area is normalized to obtain the normalized gradient sum in each evaluation area.

[0013] In some optional implementations, when the sum of the normalized gradients in the evaluation area corresponding to the current measurement area is greater than a set threshold, it is considered that the gravity field changes more drastically and the optimized filtering sequence is shorter;

[0014] When the sum of the normalized gradients in the corresponding evaluation area in the current measurement area is less than the set threshold, it is considered that the gravity field changes more slowly and the optimized filtering sequence is longer.

[0015] In some optional embodiments, According to the normalized gradient of the evaluation area corresponding to the current gravity measurement area and Get the filter sequence length N t ,in, represents the median of the sum of the gradients in the gravity reference map, is the set threshold, and N0 is the set initial filtering sequence length.

[0016] In some optional implementation schemes, the gravity field comprehensive characteristic parameter screening threshold Set the median of the gradient sum as the upper limit of the increasing length of the filter sequence.

[0017] In some optional embodiments, Downsampling is performed by the mean filtering method, where is the downsampling result, f i is the i-th original gravity measurement value.

[0018] According to another aspect of the present invention, there is provided a gravity measurement downsampling device optimized by edge detection technology, comprising:

[0019] A gradient sum calculation module is used to convert the gravity reference image into a grayscale image by edge detection, and scan the grayscale image to calculate the gradient sum of the grayscale image pixels;

[0020] The sequence length optimization module is used to optimize the sequence length of the mean filter based on the gradient and optimize the mean filter. The original gravity measurement data is downsampled using the optimized mean filter. The more drastic the change of the gravity field, the greater the gradient and the shorter the optimized filter sequence. The more gentle the change of the gravity field, the smaller the gradient and the longer the optimized filter sequence.

[0021] According to another aspect of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0023] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0024] (1) Aiming at the problem that the mean filter used for downsampling in areas with rapid changes in the gravity field tends to filter out valid gravity measurements as noise, a gravity measurement downsampling method optimized based on edge detection technology is proposed. The edge detection technology is used to evaluate the degree of change of the gravity field. In areas where the gravity field changes dramatically, the length of the downsampling sequence is reduced, and more changing gravity measurement signals are retained, which can more effectively retain the details of the gravity measurement.

[0025] (2) In areas where the gravity field changes slowly, the length of the downsampling sequence is increased to make the filtering result smoother, the amount of data after downsampling is smaller, and the amount of subsequent calculations is reduced. This improves the suppression effect of high-frequency noise and motion noise in areas where the gravity field changes slowly, and at the same time reduces the sampling frequency even further, reducing the amount of calculations for subsequent gravity measurement solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic flow chart of a gravity measurement downsampling method optimized by edge detection technology provided by an embodiment of the present invention;

[0027] Figure 2 It is a schematic diagram of a Sobel operator template provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0029] In the examples of the present invention, "first", "second", etc. are used to distinguish different objects rather than to describe a specific order or sequence.

[0030] Aiming at the problem that the downsampling parameter setting of sea and air gravimeter affects the accuracy of gravity measurement, the present invention proposes a gravity measurement downsampling method based on edge detection technology optimization, such as Figure 1 As shown, when gravity measurement is performed, the original gravity measurement data is downsampled by a mean filter. Mean filtering is essentially a low-pass filter, which has the function of filtering out high-frequency noise and motion noise while downsampling. In areas where the gravity field changes rapidly, some of the changed gravity measurement values ​​will also be regarded as noise and removed by the filter, which reduces the accuracy of gravity measurement. The present invention is based on a gravity measurement downsampling method optimized by edge detection technology. First, the edge detection technology is used to convert the gravity reference image into a grayscale image, and the grayscale image is scanned by a Sobel operator template to calculate the gradient sum of the grayscale image pixels. Then, based on the normalized gradient sum, the sequence length of the mean filter is optimized, and the original gravity measurement data is downsampled by the optimized mean filter. In this method, the gradient sum characterizes the degree of change of the gravity field in the area. The more drastic the change of the gravity field, the larger the gradient sum is, the shorter the optimized filter sequence is, and the more the retained change gravity measurement signals are; the more gentle the change of the gravity field, the smaller the gradient sum is, the longer the optimized filter sequence is, the smoother the filtering result is, and the smaller the amount of data after downsampling.

[0031] Furthermore, the gravity reference map saves the global gravity values ​​in matrix form at fixed longitude and latitude intervals. Assuming that the size of the gravity reference map is a p×q matrix grid, the gravity reference map is divided into several evaluation areas with a grid size of m×n. The current actual gravity measurement area corresponds to one of the m×n grid areas, and G represents the gravity value at the grid point in the gravity reference map.

[0032] The gravity reference image preprocessing based on edge detection technology specifically includes the following steps:

[0033] The gravity reference image is converted into an 8-bit grayscale image, and the image texture is analyzed by edge detection technology. The characteristic indicators of the gravity reference image are extracted by analyzing the relationship between the image texture features and pixels. i,jis the gravity value at the grid point (i, j). The gravity reference image is converted into an 8-bit grayscale image according to the following formula, and the grayscale range is 1 to 256.

[0034]

[0035] Among them, H i,j is the gray value at the grid point (i, j), H min The minimum gray value is 1, H max The maximum grayscale value is 256, G max is the maximum gravity value in the gravity data, G min is the minimum gravity value in the gravity data, g i,j is the gravity value at the grid point (i, j).

[0036] Image edge is an important feature of the image, and it is the place where the grayscale value of the pixel in the image changes dramatically. By using image edge detection, the grayscale value change characteristics of the gravity grayscale image can be reflected, and then the change characteristics of gravity can be reflected from the details. The Sobel operator is a common method for image edge detection. Compared with other edge detection methods, the Sobel operator is easy to implement, has a small amount of calculation, and has good noise resistance. This method is used to extract the local change gradient of gravity anomaly in the gravity map. The Sobel operator template is as follows: Figure 2 shown.

[0037] Use the Sobel operator template to scan each pixel of the grayscale image, and calculate the gradient value of the pixel in each direction using the following formula.

[0038]

[0039] Among them, h i is the grayscale value within the scan template range.

[0040] Therefore, the gradient and H of the center point of the scanning template gr (i,j) is:

[0041]

[0042] Among them, H gr (i, j) is the gradient sum of the center point of the scanning template, and the coordinates of the center point of the scanning template in the m×n grid of the evaluation area are (i, j); H gr is the sum of the gradients in the evaluation area. gr If the value is small, it means that the grayscale change between the pixel in the grayscale image and the surrounding pixels is small, that is, the gravity amplitude change of the evaluation area in the gravity reference map is small; on the contrary, H gr The value is large, and the gravity value of the assessment area in the gravity reference map varies greatly.

[0043] Furthermore, the gravity measurement downsampling method specifically includes the following steps:

[0044] During the gravity measurement process, the original signal contains a large amount of high-frequency noise and external disturbance noise. Therefore, in engineering applications, a low-pass filter is generally used to reduce the impact of noise on the gravity measurement results. Since the original sensor signal usually has a high-frequency sampling frequency of 50 to 200 Hz, using this high-frequency signal for direct filtering and solving will not only seriously consume processor computing resources, but also fail to correctly reflect the low-frequency gravity information, and will not be conducive to the attenuation and suppression of high-frequency disturbances. Therefore, it is necessary to use downsampling technology to process the raw data of the sensor. This solution uses the mean filtering method to perform downsampling processing. The calculation formula is as follows:

[0045]

[0046] in, is the downsampling result; f i is the ith original gravity measurement value; N is the length of the filter sequence. After downsampling by mean filtering, not only can the high-frequency noise be directly suppressed, but also the subsequent gravity measurement filtering processing can be facilitated.

[0047] Furthermore, the gradient-based and optimized gravity measurement downsampling parameter setting can be achieved by the following steps:

[0048] (1) Data preprocessing

[0049] In order to facilitate the use of gradient and H gr To design the downsampling parameters, we need to convert H gr Perform normalization:

[0050]

[0051] Among them, H gr Indicates the gradient sum of the current gravity measurement area and the corresponding evaluation area; H gr min It represents the minimum sum of gradients of all evaluation areas in the gravity reference map; H grmax Represents the maximum gradient sum of all evaluation areas in the gravity reference map; the normalized gradient sum The value range is [0,1].

[0052] (2) Gradient-based and optimized mean filter sequence length

[0053] The filter sequence length selected in general engineering applications is the size of the original data collected within 1 second, assuming it is N0. Greater than the set threshold When , it can be considered that the gravity value of the current gravity field changes greatly, and the detailed information of the gravity field is more. In order to retain more detailed features of the measured gravity values, the length of the filter sequence can be reduced so that the smaller changes in the original measured gravity values ​​can be retained. Less than the set threshold When , it can be considered that the gravity value of the current gravity field changes little, and the length of the filter sequence can be increased to reduce the sampling frequency and the amount of subsequent filtering calculations. At the same time, the high-frequency noise and external disturbance noise in the original measured gravity value can be removed more thoroughly. The calculated filter sequence length is:

[0054]

[0055] in, Represents the median of the gradient sum in the gravity reference map, and the screening threshold of the comprehensive characteristic parameters of the gravity field In order to prevent the length of the filter sequence from being infinitely enlarged, the median of the gradient sum is set as the upper limit of the filter sequence length increase, and N0 is the set initial filter sequence length.

[0056] The formula for calculating the length of the gravity measurement downsampling filter sequence based on gradient and optimization is as follows:

[0057]

[0058] The present invention proposes a gravity measurement downsampling method based on edge detection technology optimization through the above technical solution. In view of the change of gravity field, the grayscale image gradient corresponding to the gravity reference map is integrated into the design of gravity measurement downsampling sequence length. By using the grayscale image gradient and adaptively adjusting the sequence length of gravity measurement downsampling, more accurate downsampling results are provided in areas where the gravity field changes faster, and smoother and smaller data volume downsampling results are provided in areas where the gravity field changes more slowly.

[0059] The present invention also provides a gravity measurement downsampling device optimized by edge detection technology, comprising:

[0060] A gradient sum calculation module is used to convert the gravity reference image into a grayscale image by edge detection, and scan the grayscale image to calculate the gradient sum of the grayscale image pixels;

[0061] The sequence length optimization module is used to optimize the sequence length of the mean filter based on the gradient and optimize the mean filter. The original gravity measurement data is downsampled using the optimized mean filter. The more drastic the change of the gravity field, the greater the gradient and the shorter the optimized filter sequence. The more gentle the change of the gravity field, the smaller the gradient and the longer the optimized filter sequence.

[0062] The specific implementation of each module can refer to the description of the above method embodiment, and the embodiment of the present invention will not be repeated.

[0063] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0064] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0065] It should be pointed out that, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, and two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0066] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A gravity measurement downsampling method optimized by edge detection technology, characterized in that: include: The gravity reference image is converted into a grayscale image by using edge detection, and the grayscale image is scanned to calculate the gradient sum of the grayscale image pixels; Based on the gradient and the sequence length of the optimized mean filter, the raw data of gravity measurement is downsampled using the optimized mean filter. The larger the gradient and the greater the change in the gravity field, the shorter the corresponding filter sequence.

2. The method according to claim 1, characterized in that The method of converting the gravity reference image into a grayscale image by edge detection and scanning the grayscale image to calculate the gradient sum of the grayscale image pixels includes: Divide the gravity reference image into a number of evaluation areas, and convert the gravity reference image into a grayscale image; The grayscale image is scanned by the Sobel operator template to obtain the gradient sum of the grayscale image pixels in each evaluation area.

3. The method according to claim 2, characterized in that After obtaining the gradient sum of the grayscale image pixels in each evaluation area, the method further includes: The gradient sum of the grayscale image pixels in each evaluation area is normalized to obtain the normalized gradient sum in each evaluation area.

4. The method according to claim 3, characterized in that When the sum of the normalized gradients in the corresponding evaluation area in the current measurement area is greater than the set threshold, it is considered that the gravity field changes more drastically and the optimized filter sequence is shorter; When the sum of the normalized gradients in the corresponding evaluation area in the current measurement area is less than the set threshold, it is considered that the gravity field changes more slowly and the optimized filtering sequence is longer.

5. The method according to claim 4, characterized in that Depend on According to the normalized gradient of the evaluation area corresponding to the current gravity measurement area and Get the filter sequence length N t ,in, represents the median of the sum of the gradients in the gravity reference map, is the set threshold, and N0 is the set initial filtering sequence length.

6. The method according to claim 5, characterized in that Screening threshold of comprehensive characteristic parameters of gravity field Set the median of the gradient sum as the upper limit of the increasing length of the filter sequence.

7. The method according to claim 5 or 6, characterized in that: Depend on Downsampling is performed by the mean filtering method, where is the downsampling result, f i is the i-th original gravity measurement value.

8. A gravity measurement downsampling device optimized by edge detection technology, characterized in that: include: A gradient sum calculation module is used to convert the gravity reference image into a grayscale image by edge detection, and scan the grayscale image to calculate the gradient sum of the grayscale image pixels; The sequence length optimization module is used to optimize the sequence length of the mean filter based on the gradient and optimize the mean filter. The original gravity measurement data is downsampled using the optimized mean filter. The more drastic the change of the gravity field, the greater the gradient and the shorter the optimized filter sequence. The more gentle the change of the gravity field, the smaller the gradient and the longer the optimized filter sequence.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

Images