A method for positioning and calibration of near-seafloor video data based on ultra-short baseline

Abstract

The present invention discloses a method for positioning and correcting near-seabed video data based on an ultra-short baseline, including Step 1: obtaining ultra-short baseline positioning data; Step 2: removing abnormal data in the ultra-short baseline positioning data, establishing a four-dimensional removal model, and removing abnormal data in the three directions of X, Y, and Z; Step 3: modeling the correction method for the reorganized ultra-short baseline positioning data after removing abnormal data; Step 4: simulating and obtaining camera tow body positioning data with specified accuracy. The present invention realizes the positioning and correction of video data under existing conditions and established operation modes. By fusing and using various survey data, it removes, simulates, and corrects the error data generated by the ultra-short baseline data for video positioning over time in the course direction and other directions, achieving a good effect for near-bottom video data positioning.

Description

Technical Field

[0001] The present invention relates to the technical field of undersea mineral resource exploration, and particularly relates to a method for positioning and correcting near-seabed video data based on an ultra-short baseline. Background Art

[0002] With the mature application of deep-sea high-definition camera and photography technologies, near-seabed camera surveys have become an intuitive detection means for deep-sea mineral resource surveys. The most typical deep-sea detection platform equipped with deep-sea camera and photography equipment is currently the towed detection platform. For the undersea camera equipment carried on the towed detection platform, underwater positioning is required. Its accurate positioning is the working foundation for further in-depth resource surveys and scientific research.

[0003] Currently, the underwater positioning of the towed detection platform is generally achieved by means of an ultra-short baseline fixed on the platform. Using the ultra-short baseline for underwater positioning has become a conventional means for deep-sea scientific research in recent years. It has a fast positioning speed and high accuracy. However, due to reasons such as the measurement environment, it will cause different degrees of false data, incorrect data, and even missing data, which brings troubles to the underwater positioning of the deep-sea camera equipment. Therefore, effective processing and correction of the ultra-short baseline positioning data are the premise for ensuring that the near-bottom video data can be further exploited and utilized. It is necessary to effectively process and correct the ultra-short baseline positioning data so that the camera data can obtain more accurate positioning and be better used for deep-sea prospecting work.

[0004] Most of the existing technologies for processing the ultra-short baseline are to eliminate outliers in space, without eliminating the incorrect data generated by the change of the heading with the travel time, and fail to achieve a good correction effect. Moreover, most of them only analyze and process the ultra-short baseline positioning data itself. If there are many incorrect points in the ultra-short baseline data, there are limitations, which affect the accuracy of the correction result. With the continuous improvement of the detection technology level, generally, multiple detection methods can be used simultaneously to conduct surveys on the same target and obtain multiple relevant data. However, there is currently no specific correction method for comprehensively analyzing multiple data. Therefore, the present invention proposes a method for positioning and correcting near-seabed video data based on an ultra-short baseline to solve the problems existing in the prior art. Summary of the Invention

[0005] Aiming at the above problems, the purpose of the present invention is to propose a method for positioning and correcting near-seabed video data based on an ultra-short baseline. This method for positioning and correcting near-seabed video data based on an ultra-short baseline solves the problems of unsatisfactory and inaccurate correction effects of the correction methods for ultra-short baseline positioning data in the prior art.

[0006] To achieve the purpose of the present invention, the present invention is realized through the following technical solutions: A method for positioning and correcting near-seabed video data based on an ultra-short baseline, comprising the following steps:

[0007] Step 1: Obtain ultra-short baseline positioning data. Use an ultra-short baseline positioning system to obtain ultra-short baseline positioning data including longitude information, latitude information, and the first water depth value.

[0008] Step 2: Establish a four-dimensional rejection model to reject abnormal data in the ultra-short baseline positioning data, and obtain processed and reorganized ultra-short baseline positioning data.

[0009] Step 3: Simulate and model the processed and reorganized ultra-short baseline positioning data from the perspective of time series.

[0010] Step 4: Use the calibration model in Step 3 to simulate and interpolate and predict the calibrated and reorganized ultra-short baseline positioning data to obtain near-seabed video positioning.

[0011] The further improvement lies in: The rejection of abnormal data in Step 2 specifically includes

[0012] A1. From the perspectives of time and space series, establish a four-dimensional rejection model using the time, longitude, latitude, and water depth information in the ultra-short baseline positioning data.

[0013] A2. Determine whether the first water depth value is abnormal according to the four-dimensional rejection model. When it is abnormal, reject the ultra-short baseline positioning data with abnormal first water depth value, and vice versa, retain the ultra-short baseline positioning data.

[0014] A3. Establish a buffer zone with a range of 0.3% of the average cable length of each survey line along the direction where the ultra-short baseline positioning data has aggregation.

[0015] A4. Reject the ultra-short baseline positioning data outside the buffer zone to obtain processed ultra-short baseline positioning data.

[0016] A5. The bathymetric data integrated with the processed ultra-short baseline positioning data is the second water depth value obtained by a thermosalinograph, and the processed ultra-short baseline positioning data and the second water depth value are corresponded one by one according to time to obtain reorganized ultra-short baseline positioning data.

[0017] A6. Use the value extraction to point function in the ArcToolbox toolbox in ArcGIS, and extract the water depth value in the AUV bathymetric data according to the reorganized ultra-short baseline positioning data. Perform cross-sectional comparison and analysis on the water depth values formed by the AUV water depth value and the reorganized ultra-short baseline positioning data, and reject the points with significantly abnormal trends to make the terrain of the ultra-short baseline positioning data and the AUV bathymetric data in the same area maintain a relatively consistent trend.

[0018] A7. In terms of the course, with the help of the on-board GPS positioning data, the changing positions of the time and the ultra-short baseline positioning data in the course are determined, supplemented by the course and speed data, the change frequency of the two sets of positioning data is monitored, and the abnormal data of the ultra-short baseline in the X direction is eliminated.

[0019] A further improvement lies in that: the method for judging the abnormality of the first water depth value in A2 is to arrange the ultra-short baseline positioning data of a sounding line in chronological order in the direction of the first water depth value of the four-dimensional elimination model, obtain the points with abnormal water depth, and eliminate the ultra-short baseline positioning data of the abnormal points.

[0020] A further improvement lies in that: the sounding data integrated with the processed ultra-short baseline positioning data in the second step is the second water depth value obtained by the CTD sensor, and the processed ultra-short baseline positioning data is corresponded to the second water depth value one by one according to the time to obtain the reorganized ultra-short baseline positioning data.

[0021] A further improvement lies in that: the elimination operation in the second step is specifically to use the function of extracting values to points in the ArcToolbox toolbox in ArcGIS, and use the reorganized ultra-short baseline positioning data to extract the water depth value in the AUV sounding data, and conduct a cross-sectional comparison analysis of the AUV water depth value and the water depth value formed by the reorganized ultra-short baseline positioning data, eliminate the points with obvious abnormal trends, and make the ultra-short baseline positioning data have relatively consistent undulations with the terrain of the AUV sounding data in the same area.

[0022] A further improvement lies in that: the elimination operation in the second step is specifically to rely on the on-board GPS positioning data, determine the changing positions of the time and the ultra-short baseline positioning data in the course, supplemented by the course and speed data, monitor the change frequency of the two sets of positioning data, and eliminate the abnormal data of the ultra-short baseline in the X direction.

[0023] A further improvement lies in that: the calibration method in the third step specifically includes

[0024] B1. First, use the least squares fitting method of the cubic polynomial to calculate the calibration coefficients of the longitude and latitude after eliminating the abnormal points respectively, and determine the fitting formula according to the calibration coefficients;

[0025] B2. Use the fitting formula to fit the processed reorganized ultra-short baseline positioning data, and further determine the calibration model through the fitting effect test of the reorganized ultra-short baseline positioning data after eliminating the abnormalities.

[0026] A further improvement lies in that: before determining the calibration coefficients in B1, the data is first subjected to centering processing and standardization processing, and the determined fitting formula is expressed as

[0027] f(x) = p 1 x 3 +p2 x 2 +p 3 x + p 4

[0028] where p 1 、p 2 、p 3 are calibration coefficients, and x is the number of all points.

[0029] A further improvement lies in that: in the B2, the fitting formula is used to correct the error data generated by the fitted recombinant ultra-short baseline positioning data along the ultra-short baseline course with the change of travel time, and the ultra-short baseline positioning data after spatial fitting with time as the constraint is simulated, that is, the video positioning data.

[0030] The beneficial effects of the present invention are as follows: Based on the ultra-short baseline positioning data, the present invention combines AUV high-resolution bathymetric data, pressure sensor bathymetric data, shipborne GPS positioning data, etc. for comprehensive analysis, integrates various data, establishes a four-dimensional abnormal data rejection model, rejects the error data generated by the ultra-short baseline positioning data in the time series, course, vertical course and longitudinal direction, and establishes a cubic polynomial least squares correction model to carry out simulation correction, realizing the positioning correction of video data under the existing conditions and established operation modes, and achieving very good results. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a flowchart of the method of the present invention.

[0032] Figure 2 is a four-dimensional model diagram of the original ultra-short baseline positioning data in the second embodiment of the present invention.

[0033] Figure 3 is a buffer rejection model diagram in the second embodiment of the present invention.

[0034] Figure 4 is a cross-sectional comparison diagram in the second embodiment of the present invention.

[0035] Figure 5 is a water depth comparison diagram before and after correction in the second embodiment of the present invention.

[0036] Figure 6 is a four-dimensional model diagram of the position of the camera towed body after correction in the second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0037] To deepen the understanding of the present invention, the present invention will be further described in detail below in conjunction with embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation to the protection scope of the present invention.

[0038] Embodiment 1

[0039] According toFigure 1 As shown in the figure, this embodiment provides a method for positioning and correcting near-seabed video data based on ultra-short baseline, including the following steps:

[0040] Step 1: Obtain ultra-short baseline positioning data. Use an ultra-short baseline positioning system to obtain ultra-short baseline positioning data including longitude information, latitude information, and the first water depth value.

[0041] Step 2: Establish a four-dimensional rejection model to reject abnormal data in the ultra-short baseline positioning data and obtain processed reorganized ultra-short baseline positioning data, specifically including

[0042] A1. From the perspectives of time and space sequences, establish a four-dimensional rejection model using the time, longitude, latitude, and water depth information in the ultra-short baseline positioning data.

[0043] A2. Determine whether the first water depth value is abnormal according to the four-dimensional rejection model. When it is abnormal, reject the ultra-short baseline positioning data with abnormal first water depth value; otherwise, retain the ultra-short baseline positioning data.

[0044] When making the determination, arrange the ultra-short baseline positioning data of a survey line in chronological order in the direction of the first water depth value of the four-dimensional rejection model, obtain the points with abnormal water depth, and reject the ultra-short baseline positioning data of the abnormal points.

[0045] A3. Establish a buffer zone with a range of 0.3% of the average cable length of each survey line along the direction where the ultra-short baseline positioning data has aggregation.

[0046] A4. Reject the ultra-short baseline positioning data outside the buffer zone to obtain processed ultra-short baseline positioning data.

[0047] A5. The sounding data integrated with the processed ultra-short baseline positioning data is the second water depth value obtained by a thermosalinograph, and the processed ultra-short baseline positioning data and the second water depth value are corresponded one by one according to time to obtain reorganized ultra-short baseline positioning data.

[0048] A6. Use the value extraction to point function in the ArcToolbox toolbox of ArcGIS, and extract the water depth value in the AUV sounding data according to the reorganized ultra-short baseline positioning data. Conduct cross-sectional comparison and analysis of the AUV water depth value and the water depth value formed by the reorganized ultra-short baseline positioning data, and reject the points with significantly abnormal trends to make the terrain of the ultra-short baseline positioning data and the AUV sounding data in the same area maintain a relatively consistent trend.

[0049] A7. In the navigation direction, with the help of the shipborne GPS positioning data, monitor the change frequency of the two sets of positioning data by the changing positions of time and ultra-short baseline positioning data in the navigation direction, supplemented by the navigation direction and speed data, and reject the abnormal data of the ultra-short baseline in the X direction (navigation direction).

[0050] Step Five: Simulate and model the processed recombinant ultra-short baseline positioning data from the perspective of time series;

[0051] B1. First, use the cubic polynomial least squares fitting method to calculate the correction coefficients of longitude and latitude after removing abnormal points respectively, and determine the fitting formula according to the correction coefficients, expressed as

[0052] f(x) = p 1 x 3 + p 2 x 2 + p 3 x + p 4

[0053] where p 1 , p 2 , p 3 are the correction coefficients, and x is the number of all points;

[0054] B2. Use the fitting formula to fit the processed recombinant ultra-short baseline positioning data, and further determine the correction model through the fitting effect test of the recombinant ultra-short baseline positioning data after removing abnormalities;

[0055] Step Four: Use the correction model in Step Three to simulate and interpolate and predict the corrected recombinant ultra-short baseline positioning data to obtain near-seabed video positioning.

[0056] Example Two

[0057] According to Figure 1-6 as shown, this example provides a near-seabed video data positioning and correction method based on an ultra-short baseline, including the following steps:

[0058] Step One: Obtain ultra-short baseline positioning data in.txt format, and use the ultra-short baseline positioning system to obtain ultra-short baseline positioning data including longitude information, latitude information, and the first water depth value;

[0059] Step Two: Remove abnormal data from the ultra-short baseline positioning data to obtain the processed ultra-short baseline positioning data, specifically including

[0060] A1. From the perspective of spatial sequence, use the longitude information, latitude information, and the first water depth value in the ultra-short baseline positioning data to establish a four-dimensional removal model. As shown in the attached instructions Figure 2 as shown, in the four-dimensional model, X represents the deviation distance of the target along the bow direction, that is, longitude, Y represents the deviation distance of the target along the port and starboard direction of the mother ship, that is, latitude, Z represents the deviation distance of the target along the vertical direction, that is, the first water depth value, and T represents the time series;

[0061] A2. Determine whether the first water depth value is abnormal according to the four-dimensional rejection model. When it is abnormal, reject the ultra-short baseline positioning data with abnormal first water depth value, and vice versa, retain the ultra-short baseline positioning data;

[0062] During the judgment, in the direction of the first water depth value of the four-dimensional rejection model, arrange the ultra-short baseline positioning data of a survey line in chronological order, obtain the points with abnormal water depth, and reject the ultra-short baseline positioning data of the abnormal points;

[0063] A3. Use the buffer function of spatial analysis in ArcGIS software to establish a buffer with a range of 0.3% of the average cable length of each survey line along the direction of the reliable towed body position data with aggregation;

[0064] A4. Reject the ultra-short baseline positioning data outside the buffer and retain the data inside the buffer. As shown in the attached instructions, obtain the processed ultra-short baseline positioning data; Figure 3 As shown, obtain the processed ultra-short baseline positioning data;

[0065] A5. Integrate the processed ultra-short baseline positioning data with the second water depth value of the bathymetric data obtained by the CTD sensor in.txt format, and correspond the processed ultra-short baseline positioning data to the second water depth value one by one according to time to obtain more reliable reorganized ultra-short baseline positioning data;

[0066] A6. Reject the data with abnormal water depth in the reorganized ultra-short baseline positioning data according to the AUV bathymetric data. Specifically, use the function of extracting values to points in the ArcToolbox of ArcGIS, and use the reorganized ultra-short baseline positioning data to extract the water depth value in the AUV bathymetric data. Compare and analyze the cross-section of the AUV water depth value and the water depth value formed by the reorganized ultra-short baseline positioning data, and reject the points with obvious abnormal trends, so that the ultra-short baseline positioning data and the terrain of the AUV bathymetric data in the same area maintain relatively consistent undulations, and obtain the processed reorganized ultra-short baseline positioning data. As shown in the attached instructions, CTD in the figure represents the water depth value formed by the reorganized ultra-short baseline positioning data; Figure 4 As shown, in the figure, CTD represents the water depth value formed by the reorganized ultra-short baseline positioning data;

[0067] A7. In the navigation direction, with the help of the shipborne GPS positioning data, monitor the change frequency of the two sets of positioning data by the change position of time and the ultra-short baseline positioning data in the navigation direction, supplemented by the navigation direction and speed data, and reject the abnormal data of the ultra-short baseline in the X direction (navigation direction).

[0068] After steps one to two, only the abnormal value points that can be judged manually are rejected. However, in the x direction, the points are disordered in the time series, and the spatial position does not change with time. After being corrected by step three, the points are arranged in sequence, and the geographical position changes in the forward direction with the increase of time. As shown in the attached instructions; Figure 6 As shown;

[0069] Step 3: Simulate and model the processed recombinant ultra-short baseline positioning data from the perspective of time series, specifically including

[0070] B1. Taking the serial number (chronological order) as x, first perform centering processing and standardization processing on the data, then use the least squares fitting method of cubic polynomials to calculate the correction coefficients of longitude and latitude respectively, and determine the fitting formula according to the correction coefficients, expressed as

[0071] f(x) = p 1 x 3 + p 2 x 2 + p 3 x + p 4

[0072] where p 1 , p 2 , p 3 are correction coefficients, and x is the serial number of all points;

[0073] B2. Use the fitting formula to fit the processed recombinant ultra-short baseline positioning data, and further determine the correction model through the fitting effect test of the recombinant ultra-short baseline positioning data after removing anomalies, so as to obtain the corrected recombinant ultra-short baseline positioning data, as shown in the attached drawings of the specification Figure 5 ;

[0074] Step 4: Use the correction model in Step 3 to simulate and interpolate and predict the corrected recombinant ultra-short baseline positioning data to obtain near-seabed video positioning.

[0075] During the acquisition process of the ultra-short baseline, there will be a problem of short-term positioning failure. Coupled with the removal of a part of the positioning abnormal data, the positioning data is discontinuous and incomplete. Therefore, the positioning data at all time points is interpolated using the correction model curve (i.e., the fitting formula) formed in Step 3 to obtain continuous positioning data of the camera tow body. To further verify this method, a set of high-quality ultra-short baseline positioning data is selected not to participate in the fitting and is used to compare with the corresponding positioning data simulated by fitting and interpolation. After verification, the method is very effective.

[0076] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for positioning and calibration of near-seafloor video data based on ultra-short baseline, characterized in that, it includes the following steps: Step 1: Obtain ultra-short baseline positioning data. Use the ultra-short baseline positioning system to obtain ultra-short baseline positioning data including longitude information, latitude information, and the first water depth value; Step 2: Establish a four-dimensional rejection model to reject abnormal data in the ultra-short baseline positioning data, and obtain processed reorganized ultra-short baseline positioning data; The rejection of abnormal data specifically includes A1. From the perspectives of time and space sequences, establish a four-dimensional rejection model using the time, longitude, latitude, and water depth information in the ultra-short baseline positioning data; A2. Judge whether the first water depth value is abnormal according to the four-dimensional rejection model. When it is abnormal, reject the ultra-short baseline positioning data with abnormal first water depth value, otherwise retain the ultra-short baseline positioning data; Among them, the method for judging abnormal first water depth value is to arrange the ultra-short baseline positioning data of a survey line in chronological order in the direction of the first water depth value of the four-dimensional rejection model, obtain the points with abnormal water depth, and reject the ultra-short baseline positioning data of the abnormal points; A3. Establish a buffer zone with a range of 0.3% of the average cable length of each survey line along the direction where the ultra-short baseline positioning data has aggregation; A4. Reject the ultra-short baseline positioning data outside the buffer zone to obtain processed ultra-short baseline positioning data; A5. The bathymetric data integrated with the processed ultra-short baseline positioning data is the second water depth value obtained by a thermosalinograph, and the processed ultra-short baseline positioning data is corresponded to the second water depth value one by one according to time to obtain reorganized ultra-short baseline positioning data; A6. Use the value extraction to point function in the ArcToolbox toolbox of ArcGIS, and extract the water depth value in the AUV bathymetric data according to the reorganized ultra-short baseline positioning data. Conduct a cross-sectional comparison analysis between the AUV water depth value and the water depth value formed by the reorganized ultra-short baseline positioning data, and reject the points with obviously abnormal trends, so that the ultra-short baseline positioning data and the terrain of the AUV bathymetric data in the same area maintain a relatively consistent trend; A7. In the navigation direction, with the help of shipborne GPS positioning data, monitor the change frequency of the two sets of positioning data by the change positions of time and ultra-short baseline positioning data in the navigation direction, supplemented by the navigation direction and speed data, and reject the abnormal data of the ultra-short baseline in the X direction; Step 3: Simulate and model the processed reorganized ultra-short baseline positioning data from the time series perspective; Step 4: Use the calibration model in Step 3 to simulate and interpolate and predict the calibrated reorganized ultra-short baseline positioning data to obtain near-seafloor video positioning.

2. A method for positioning and calibration of near-seafloor video data based on ultra-short baseline according to claim 1, characterized in that: The calibration method in Step 3 specifically includes B1. First, use the cubic polynomial least squares fitting method to calculate the calibration coefficients of longitude and latitude after rejecting abnormal points respectively, and determine the fitting formula according to the calibration coefficients; B2. Use the fitting formula to fit the processed reorganized ultra-short baseline positioning data, and further determine the calibration model through the fitting effect test of the reorganized ultra-short baseline positioning data after rejecting abnormalities.

3. A method for positioning and calibration of near-seabed video data based on ultra-short baseline according to claim 2, characterized in that: before determining the calibration coefficient in B1, the data is first subjected to centering processing and standardization processing, and the determined fitting formula is expressed as f(x) = p 1 x 3 + p 2 x 2 + p 3 x + p 4 where p 1 , p 2 , p 3 are correction factors, and x is the numbering of all points on the time series.

4. A method for positioning and calibration of near-seabed video data based on ultra-short baseline according to claim 2, characterized in that: in B2, the fitting formula is used to correct the error data generated by the fitted recombined ultra-short baseline positioning data along the ultra-short baseline navigation direction with the travel time, and the ultra-short baseline positioning data after spatial fitting with time as the constraint is simulated, that is, the video positioning data.

5. A method for positioning and calibration of near-seabed video data based on ultra-short baseline according to claim 1, characterized in that: in the fourth step, after obtaining the near-seabed video positioning, some ultra-short baseline positioning data with good quality are selected and not involved in the fitting, and are used to compare with the corresponding positioning data predicted by fitting and interpolation to verify the effectiveness of the video positioning calibration method.

Images