Target fusion positioning method and device for maritime emergency rescue

Abstract

The invention discloses a target fusion positioning method and device for maritime emergency rescue. The method comprises the following steps: acquiring a multi-source target position information set; preprocessing the multi-source target position information set to obtain a registration position information set; and performing information fusion processing on the registration position information set to obtain target fusion positioning result information of the maritime emergency rescue area. The invention discloses a target fusion positioning method and device for maritime emergency rescue, and aims to solve the problem of target search in maritime emergency rescue, and improve the tracking precision and tracking range of maritime distress targets by organically fusing different types of detection methods.

Description

Technical Field

[0001] This invention relates to the field of maritime emergency rescue, and specifically to a target fusion positioning method and apparatus for maritime emergency rescue. Background Technology

[0002] In maritime emergency rescue, existing rescue platforms mostly use optical tracking, AIS tracking, or radar tracking methods to detect distressed targets, lacking the integration of these three methods.

[0003] Optical tracking methods are less effective for targets with low infrared temperature differences or in nighttime environments. AIS tracking methods require the target to transmit signals, and if the target equipment is damaged, it will be unable to transmit signals. Additionally, it cannot locate personnel targets. Radar tracking methods are limited by adverse factors such as sea clutter and echoes, and are less effective for tracking "low, slow, and small" targets.

[0004] How to address the target search problem in maritime emergency rescue and organically integrate different types of detection methods to improve the tracking accuracy and range of distressed targets at sea is an urgent problem that needs to be solved. Summary of the Invention

[0005] This invention primarily addresses the target search problem in maritime emergency rescue by organically integrating different types of detection methods to improve the tracking accuracy and range of distressed targets at sea. This invention discloses a target fusion positioning method and device for maritime emergency rescue.

[0006] In a first aspect, the present invention discloses a target fusion positioning method for maritime emergency rescue, comprising:

[0007] S1, Obtain a set of multi-source target location information;

[0008] S2, preprocess the multi-source target location information set to obtain a registration location information set;

[0009] S3, perform information fusion processing on the registered location information set to obtain target fusion positioning result information of the maritime emergency rescue area.

[0010] The acquisition of the multi-source target location information set includes:

[0011] S11, acquire target optical images of the maritime emergency rescue area;

[0012] S12 performs detection processing on the target optical image to obtain a target location set; the target location set includes first location information and confidence information at several time points;

[0013] S13, perform radar detection processing on the maritime emergency rescue area to obtain a target positioning information set; the target positioning information set includes second position information at several times and the corresponding detection signal-to-noise ratio and correct detection probability;

[0014] S14, Receive the target AIS signal, perform detection and demodulation processing on the target AIS signal to obtain a target location transmission information set; the target location transmission information set includes third location information at several times and the corresponding received signal-to-noise ratio;

[0015] S15, using the target location set, target positioning information set and target location transmission information set, a multi-source target location information set is constructed.

[0016] The expression for calculating the confidence level information is:

[0017] z = T2(δ1) + L2(WNR),

[0018] Where T2() represents the second-order polynomial of the first-kind Chebyshev polynomial, L2() represents the second-order Laguerre polynomial, z is the confidence information, δ1 is the deviation between the pixel coordinate of the maximum gray value of the target region in the target optical image acquired at a certain time and the first position information at that time, and WNR is the ratio of the average gray value of the target region in the target optical image acquired at a certain time to the average gray value of the non-target region.

[0019] The preprocessing of the multi-source target location information set to obtain the registration location information set includes:

[0020] S21, perform data cleaning processing on the multi-source target location information set to obtain the first dataset;

[0021] S22, The first dataset is processed to obtain the second dataset;

[0022] S23, perform data reduction processing on the second dataset to obtain a set of registration location information.

[0023] The data reduction process performed on the second dataset to obtain a set of registration location information includes:

[0024] S231, for each type of data attribute in the second dataset, using the data collection information of the data as the known independent variable and the data value of the data as the known dependent variable, the curve to be approximated is constructed using the known independent variable and the known dependent variable;

[0025] S232, use the function approximation method to perform curve fitting on the curve to be approximated, and obtain the best uniform approximation polynomial of the class data attribute;

[0026] S233, using the best uniform approximation polynomial, the known independent variable is calculated and processed to obtain the approximate dependent variable;

[0027] S234, determine whether the absolute value of the difference between the approximate dependent variable and the corresponding known dependent variable is greater than a set first regression discrimination threshold; if it is greater than the first regression discrimination threshold, delete the data from the second dataset; if it is less than or equal to the first regression discrimination threshold, do not process the data.

[0028] S235, perform fusion processing on all data in the second dataset after executing S231 to S234 to obtain a set of registration location information.

[0029] The information fusion processing of the registered location information set to obtain target fusion positioning result information for the maritime emergency rescue area includes:

[0030] S31, perform clustering processing on all location information in the target fusion positioning result information to obtain several location category information and the included location information;

[0031] S32, perform credibility judgment processing on all location category information to obtain a set of feasible location category information;

[0032] S33, perform fusion positioning processing on each location category information in the feasible location category information set to obtain the corresponding reliable location information;

[0033] S34. Using all reliable location information, the target fusion positioning result information of the maritime emergency rescue area is constructed.

[0034] The process of performing credibility determination on all location category information to obtain a set of feasible location category information includes:

[0035] For each location category, the corresponding category confidence information is calculated; the expression for calculating the category confidence information is:

[0036]

[0037] Where L represents category confidence information, m1, m2, and m3 are the number of first, second, and third location information items contained in a location category, respectively, and z i b i c i and d iLet be the confidence level of the i-th first location information contained in the location category information, the signal-to-noise ratio of the detected signal of the i-th second location information contained in the location category information, the correct detection probability of the i-th second location information contained in the location category information, and the signal-to-noise ratio of the received signal of the i-th third location information contained in the location category information, respectively; q is a preset bias factor; z0 is the mean of the confidence level of all first location information contained in the location category information; and d max The maximum value of the received signal-to-noise ratio of all third location information contained in the location category information.

[0038] According to a second aspect of the present invention, a target fusion positioning device for maritime emergency rescue is disclosed, the device comprising:

[0039] Memory containing executable program code;

[0040] A processor coupled to the memory;

[0041] The processor calls the executable program code stored in the memory to execute the target fusion positioning method for maritime emergency rescue.

[0042] In a third aspect, the present invention discloses a computer-storable medium storing computer instructions, which, when invoked by a computer, are used to execute the target fusion positioning method for maritime emergency rescue.

[0043] In a fourth aspect, the present invention discloses an information data processing terminal, which is used to implement the target fusion positioning method for maritime emergency rescue.

[0044] The beneficial effects of this invention are as follows:

[0045] This invention discloses a target fusion positioning method and device for maritime emergency rescue. The target search problem in maritime emergency rescue is solved by organically integrating different types of detection methods, thereby improving the tracking accuracy and tracking range of distressed targets at sea.

[0046] When performing multi-source information fusion, this invention first clusters the target location information to obtain the location information of multiple targets. By establishing a feasibility discrimination model, the credibility-related information of all data in each category, such as detection signal-to-noise ratio and detection correctness probability, is comprehensively evaluated and processed. Target categories with low credibility are deleted to ensure the accuracy of the estimation results.

[0047] After obtaining a target category with high credibility, this invention performs fusion calculation on the location information of multiple target categories, ensuring the accuracy of location fusion estimation. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0049] To better understand the content of this invention, an embodiment is provided here.

[0050] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.

[0051] In a first aspect, the present invention discloses a target fusion positioning method for maritime emergency rescue, comprising:

[0052] S1, Obtain a set of multi-source target location information;

[0053] S2, preprocess the multi-source target location information set to obtain a registration location information set;

[0054] S3, perform information fusion processing on the registered location information set to obtain target fusion positioning result information of the maritime emergency rescue area;

[0055] The acquisition of the multi-source target location information set includes:

[0056] S11, acquire target optical images of the maritime emergency rescue area;

[0057] S12 performs detection processing on the target optical image to obtain a target location set; the target location set includes first location information and confidence information at several time points;

[0058] S13, perform radar detection processing on the maritime emergency rescue area to obtain a target positioning information set; the target positioning information set includes second position information at several times and the corresponding detection signal-to-noise ratio and correct detection probability;

[0059] S14, Receive the target AIS signal, perform detection and demodulation processing on the target AIS signal to obtain a target location transmission information set; the target location transmission information set includes third location information at several times and the corresponding received signal-to-noise ratio;

[0060] S15, using the target location set, target positioning information set, and target location transmission information set, a multi-source target location information set is constructed;

[0061] The expression for calculating the confidence level information is:

[0062] z = T2(δ1) + L2(WNR),

[0063] Where T2() represents the second-order polynomial of the first-kind Chebyshev polynomial, L2() represents the second-order Laguerre polynomial, z is the confidence information, δ1 is the deviation between the pixel coordinate of the maximum gray value of the target region of the target optical image acquired at a time and the first position information at the time, and WNR is the ratio of the average gray value of the target region of the target optical image acquired at a time to the average gray value of the non-target region.

[0064] The confidence level calculation expression incorporates image pixel deviation δ1 and grayscale ratio WNR, evaluating the reliability of optical image positioning results from two aspects: target positioning accuracy and image feature intensity. For example, a smaller deviation value and higher grayscale contrast result in higher confidence, effectively filtering out optical data that is interfered with or inaccurately positioned. Utilizing the nonlinear characteristics of Chebyshev and Laguerre polynomials, small deviations or grayscale changes are amplified, enhancing the ability to distinguish the reliability of different data and avoiding insufficient information differentiation due to linear calculations. Polynomial functions can smoothly handle data fluctuations and have strong robustness to interference factors such as image noise and illumination changes, ensuring accurate reliability assessment of optical image positioning results even under complex sea conditions.

[0065] The target-free region is the region in the target optical image acquired at a given time where no target was detected; the target region in the target optical image is obtained through image feature detection algorithms such as border detection.

[0066] The preprocessing of the multi-source target location information set to obtain the registration location information set includes:

[0067] S21, perform data cleaning processing on the multi-source target location information set to obtain the first dataset;

[0068] S22, The first dataset is processed to obtain the second dataset;

[0069] S23, perform data reduction processing on the second dataset to obtain a set of registration location information.

[0070] The data reduction process performed on the second dataset to obtain a set of registration location information includes:

[0071] S231, for each type of data attribute in the second dataset, using the data collection information of the data as the known independent variable and the data value of the data as the known dependent variable, the curve to be approximated is constructed using the known independent variable and the known dependent variable;

[0072] S232, use the function approximation method to perform curve fitting on the curve to be approximated, and obtain the best uniform approximation polynomial of the class data attribute;

[0073] S233, using the best uniform approximation polynomial, the known independent variable is calculated and processed to obtain the approximate dependent variable;

[0074] S234, determine whether the absolute value of the difference between the approximate dependent variable and the corresponding known dependent variable is greater than a set first regression discrimination threshold; if it is greater than the first regression discrimination threshold, delete the data from the second dataset; if it is less than or equal to the first regression discrimination threshold, do not process the data.

[0075] S235, perform fusion processing on all data in the second dataset after executing S231 to S234 to obtain a set of registration location information;

[0076] The information fusion processing of the registered location information set to obtain target fusion positioning result information for the maritime emergency rescue area includes:

[0077] S31, perform clustering processing on all location information in the target fusion positioning result information to obtain several location category information and the included location information;

[0078] S32, perform credibility judgment processing on all location category information to obtain a set of feasible location category information;

[0079] S33, perform fusion positioning processing on each location category information in the feasible location category information set to obtain the corresponding reliable location information;

[0080] S34, confirm all reliable location information as target fusion positioning results information for the maritime emergency rescue area.

[0081] The process of performing credibility determination on all location category information to obtain a set of feasible location category information includes:

[0082] For each location category, the corresponding category confidence information is calculated; the expression for calculating the category confidence information is:

[0083]

[0084] Where L represents category confidence information, m1, m2, and m3 are the number of first, second, and third location information items contained in a location category, respectively, and z i b i c i and d iLet be the confidence level of the i-th first location information contained in the location category information, the signal-to-noise ratio of the detected signal of the i-th second location information contained in the location category information, the correct detection probability of the i-th second location information contained in the location category information, and the signal-to-noise ratio of the received signal of the i-th third location information contained in the location category information, respectively. Let q be a preset bias factor, which can take the value 3, z0 be the mean of the confidence levels of all first location information contained in the location category information, and d be the mean of the confidence levels of all first location information contained in the location category information. max The maximum value of the received signal-to-noise ratio of all third location information contained in the location category information.

[0085] The calculation expression for the category credibility information combines the confidence level of optical images, the signal-to-noise ratio and correct detection probability of radar detection, and the received signal-to-noise ratio of AIS signals to comprehensively evaluate the credibility of location categories. The weighted summation of different types of information ensures that the fusion result comprehensively reflects the advantages and reliability of each data source. A logarithmic transformation is applied to optical information to amplify the influence of high-confidence data; a squared term is used for radar information to enhance the weight of high signal-to-noise ratio and high detection probability data; and a sine function is used for AIS information to rapidly increase credibility when the signal-to-noise ratio is close to its maximum value, avoiding interference from low-quality data in decision-making. By presetting parameters such as the bias factor q, the weights of various types of information can be flexibly adjusted according to different rescue scenarios (such as marine environment and equipment performance), improving the model's adaptability to complex scenarios.

[0086] The step of performing fusion positioning processing on each location category information in the feasible location category information set to obtain the corresponding reliable location information includes:

[0087] For each location type information in the feasible location category information set, perform fusion calculation processing on all the location information it contains to obtain the corresponding reliable location information;

[0088] The expression for the fusion calculation process is:

[0089]

[0090] in, For reliable location information, z1 i b1 i and d1 i Each location category contains the i-th first location information p. 1i The confidence information and location category information include the i-th second location information p 2i The detection signal signal-to-noise ratio and the location category information include the i-th third location information p. 3iThe received signal-to-noise ratio, n1, n2, and n3 are the number of first, second, and third location information contained in a location category information in the feasible location category information set, respectively, q1, q2, and q3 are the preset first, second, and third normalization factors, respectively, and d1 max The maximum value of the received signal-to-noise ratio of all third location information contained in the location category information.

[0091] The expression used in the fusion calculation is a weighted average based on the confidence level of each location information (such as optical confidence level, radar signal-to-noise ratio, and AIS signal-to-noise ratio). Data with higher confidence levels have a larger proportion in the fusion result, achieving high-precision target positioning. A normalization factor is used to normalize the weights of different dimensions, ensuring that the weights of each data source are calculated on a uniform scale. This avoids deviations in the fusion result due to differences in dimensions and improves positioning stability. The normalization factor can be adjusted according to the actual rescue scenario. For example, when radar signals are interfered with, the weight of radar data can be reduced, while the role of optical or AIS data can be enhanced. This allows the fusion positioning method to quickly adapt to changes in the maritime environment and ensures the accuracy of rescue target positioning.

[0092] The curve fitting of the curve to be approximated using the function approximation method can employ the best uniform linear approximation method. The best uniform approximation polynomial f(Ix) is expressed as:

[0093] f(Ix)=α P1 (Ix) P1 +α P1-1 (Ix) P1-1 +…+α2(Ix) 2 +α1(Ix)+α0,

[0094] Where P1 is the order of the best uniform approximation polynomial f(Ix), α0, α1, α2, ..., α P1 The coefficients of the best uniform approximation polynomial f(Ix);

[0095] The received signal signal-to-noise ratio is the signal-to-noise ratio of the output signal after the target AIS signal has passed through the matched filter;

[0096] The signal-to-noise ratio of the detected signal is the signal-to-noise ratio of the output signal after the radar received signal has passed through a matched filter;

[0097] The first normalization factor, the second normalization factor, and the third normalization factor can be 2.3, 4.5, and 3.7, respectively.

[0098] The clustering process can be implemented using clustering analysis algorithms, such as the K-means algorithm.

[0099] The data cleaning process includes filling in missing values, smoothing noisy data, and smoothing or deleting outliers. Smoothing noisy data involves first identifying the noisy data, and then smoothing it based on the data preceding and following it. The noisy data refers to values ​​whose values ​​are less than the sensor's detection sensitivity for the observed data, or greater than the sensor's measurement upper limit for the observed data. Outlier identification can be performed using a Kalman filter. The filling values ​​for missing values ​​can be determined by averaging the measurements within a certain sampling interval before and after the missing value.

[0100] The unified processing of the collected information can be achieved using a time registration method; the collected information is the time information.

[0101] The time registration process unifies different types of data onto the same time base; the time registration process can employ methods such as extrapolation / extrapolation and Lagrange three-point interpolation.

[0102] The correct detection probability is calculated based on the correct detection probability formula of radar coherent detection.

[0103] According to a second aspect of the present invention, a target fusion positioning device for maritime emergency rescue is disclosed, the device comprising:

[0104] Memory containing executable program code;

[0105] A processor coupled to the memory;

[0106] The processor calls the executable program code stored in the memory to execute the target fusion positioning method for maritime emergency rescue.

[0107] In a third aspect, the present invention discloses a computer-storable medium storing computer instructions, which, when invoked by a computer, are used to execute the target fusion positioning method for maritime emergency rescue.

[0108] In a fourth aspect, the present invention discloses an information data processing terminal, which is used to implement the target fusion positioning method for maritime emergency rescue.

[0109] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A target fusion positioning method for maritime emergency rescue, characterized in that, include: S1, Obtain a set of multi-source target location information; S2, preprocess the multi-source target location information set to obtain a registration location information set; S3, perform information fusion processing on the registered location information set to obtain target fusion positioning result information of the maritime emergency rescue area.

2. The target fusion positioning method for maritime emergency rescue as described in claim 1, characterized in that, The acquisition of the multi-source target location information set includes: S11, acquire target optical images of the maritime emergency rescue area; S12 performs detection processing on the target optical image to obtain a target location set; the target location set includes first location information and confidence information at several time points; S13, perform radar detection processing on the maritime emergency rescue area to obtain a target positioning information set; the target positioning information set includes second position information at several times and the corresponding detection signal-to-noise ratio and correct detection probability; S14, Receive the target AIS signal, perform detection and demodulation processing on the target AIS signal to obtain a target location transmission information set; the target location transmission information set includes third location information at several times and the corresponding received signal-to-noise ratio; S15, using the target location set, target positioning information set, and target location transmission information set, a multi-source target location information set is constructed.

3. The target fusion positioning method for maritime emergency rescue as described in claim 2, characterized in that, The expression for calculating the confidence level information is: z = T2(δ1) + L2(WNR), Where T2() represents the second-order polynomial of the first-kind Chebyshev polynomial, L2() represents the second-order Laguerre polynomial, z is the confidence information, δ1 is the deviation between the pixel coordinate of the maximum gray value of the target region in the target optical image acquired at a certain time and the first position information at that time, and WNR is the ratio of the average gray value of the target region in the target optical image acquired at a certain time to the average gray value of the non-target region.

4. The target fusion positioning method for maritime emergency rescue as described in claim 1, characterized in that, The preprocessing of the multi-source target location information set to obtain the registration location information set includes: S21, perform data cleaning processing on the multi-source target location information set to obtain the first dataset; S22, The first dataset is processed to obtain the second dataset; S23, perform data reduction processing on the second dataset to obtain a set of registration location information.

5. The target fusion positioning method for maritime emergency rescue as described in claim 4, characterized in that, The data reduction process performed on the second dataset to obtain a set of registration location information includes: S231, for each type of data attribute in the second dataset, using the data collection information of the data as the known independent variable and the data value of the data as the known dependent variable, the curve to be approximated is constructed using the known independent variable and the known dependent variable; S232, use the function approximation method to perform curve fitting on the curve to be approximated, and obtain the best uniform approximation polynomial of the class data attribute; S233, using the best uniform approximation polynomial, the known independent variable is calculated and processed to obtain the approximate dependent variable; S234, determine whether the absolute value of the difference between the approximate dependent variable and the corresponding known dependent variable is greater than a set first regression discrimination threshold; if it is greater than the first regression discrimination threshold, delete the data from the second dataset; if it is less than or equal to the first regression discrimination threshold, do not process the data. S235, perform fusion processing on all data in the second dataset after executing S231 to S234 to obtain a set of registration location information.

6. The target fusion positioning method for maritime emergency rescue as described in claim 1, characterized in that, The information fusion processing of the registered location information set to obtain target fusion positioning result information for the maritime emergency rescue area includes: S31, perform clustering processing on all location information in the target fusion positioning result information to obtain several location category information and the included location information; S32, perform credibility judgment processing on all location category information to obtain a set of feasible location category information; S33, perform fusion positioning processing on each location category information in the feasible location category information set to obtain the corresponding reliable location information; S34. Using all reliable location information, the target fusion positioning result information of the maritime emergency rescue area is constructed.

7. The target fusion positioning method for maritime emergency rescue as described in claim 6, characterized in that, The process of performing credibility determination on all location category information to obtain a set of feasible location category information includes: For each location category, the corresponding category confidence information is calculated; the expression for calculating the category confidence information is: Where L represents the category confidence information, m1, m2, and m3 are the number of first, second, and third location information items contained in a location category, respectively, and z i b i c i and d i Let be the confidence level of the i-th first location information contained in the location category information, the signal-to-noise ratio of the detected signal of the i-th second location information contained in the location category information, the correct detection probability of the i-th second location information contained in the location category information, and the signal-to-noise ratio of the received signal of the i-th third location information contained in the location category information, respectively; q is a preset bias factor; z0 is the mean of the confidence level of all first location information contained in the location category information; and d max The maximum value of the received signal-to-noise ratio of all third location information contained in the location category information.

8. A target fusion positioning device for maritime emergency rescue, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the target fusion positioning method for maritime emergency rescue as described in any one of claims 1 to 7.

9. A computer-storable medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked by the computer, are used to execute the target fusion positioning method for maritime emergency rescue as described in any one of claims 1 to 7.

10. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the target fusion positioning method for maritime emergency rescue as described in any one of claims 1 to 7.

Images