Method and system for evaluating fused track quality based on track correlation

Abstract

The invention discloses a method and system for evaluating the quality of a fused track based on track correlation, and the method comprises the steps: selecting a first track point at a current moment from a fused track, and selecting an original track point with the minimum time difference with the current moment from an original track, calculating a first distance between the first track point corresponding to the current moment and the original track point; calculating an absolute value of a difference value between the second distance and the third distance, and recording the absolute value as a fourth distance; recording the absolute value of the difference value between the eighth distance and the seventh distance as a ninth distance; all track points in the fused track are traversed, all original track data in the original track library are traversed, and track codes of all original tracks with correlation larger than a preset threshold value are written into fused track matching parameters; and calculating the association accuracy, the positioning error improvement rate, the track integrity improvement rate and the track continuity improvement rate of the fused track, and evaluating the quality of the fused track. According to the invention, the quality of the fusion track is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a method and system for evaluating fusion track quality based on track correlation. BACKGROUND

[0002] In order to improve the tracking accuracy of the target object, reduce error accumulation, eliminate the influence of blind area, enhance the reliability of track point data, avoid single point failure, and improve the anti-interference ability, it is necessary to comprehensively process data from different sensors and different time points to obtain a more accurate motion trajectory of the target object, that is, track fusion technology. At present, track fusion technology has been widely used in many fields, such as radar tracking, marine navigation, air traffic control, etc. For different application scenarios, track fusion algorithms are constantly being researched and improved to better meet the needs. However, whether the actual use effect of the track fusion algorithm in the actual scene is consistent with the experimental results needs to form a complete fusion data evaluation and analysis process to evaluate the fusion effect, so as to locate the shortcomings of the fusion algorithm and optimize the algorithm.

[0003] Currently, in actual application, the overall performance of the fused track is not good, mainly including high track breakage rate, frequent association platform jump, large position estimation deviation and other problems. The root cause lies in the lack of a systematic and quantifiable quality evaluation system, which makes it difficult to objectively and accurately measure the performance of the fusion result, thus it is difficult to effectively locate the defects of the fusion algorithm, and ultimately leads to the long-term low-level cycle of the fused track quality, making it difficult to adapt to complex and variable actual application scenarios. SUMMARY In order to solve the problem of being difficult to evaluate the performance of the fusion algorithm, the present application provides a method and system for evaluating the quality of the fused track based on track correlation, which analyzes the correlation between the fused track data and each original track data, calculates each evaluation index of the fused track and the related original track, and comprehensively evaluates the performance of the fusion algorithm.

[0004] The present application discloses a method for evaluating the quality of the fused track based on track correlation, which comprises: Step 1: selecting a group of original track data from the original track database, and setting the matching parameters of the fused track to be empty; Step 2: selecting a first track point at the current time from the fused track, and selecting an original track point with the smallest time difference from the current time from the original track, calculating the distance between the first track point and the original track point at the current time, and recording it as the first distance; the fused track is obtained by fusing all original track data in the original track database; Step 3: if the first track point is not the first track point in the fusion track and is not the last track point in the fusion track, a second track point at the next time is selected from the fusion track, a track point with the minimum time difference from the next time is selected from the original track, a distance between the second track point and the first track point is calculated and recorded as a second distance, a distance between the track points at the current time and the next time is calculated and recorded as a third distance, an absolute value of a difference between the second distance and the third distance is calculated and recorded as a fourth distance; Step 4: a third track point at the previous time is selected from the fusion track, a track point with the minimum time difference from the third track point is selected from the original track, a distance between the first track point and the third track point is calculated and recorded as a fifth distance, a distance between the track points at the current time and the previous time is calculated and recorded as a sixth distance, a difference between the fifth distance and the sixth distance is recorded as a seventh distance, a difference between the second distance and the third distance is recorded as an eighth distance, and an absolute value of a difference between the eighth distance and the seventh distance is recorded as a ninth distance; Step 5: steps 2 to 4 are repeated until all track points in the fusion track are traversed, a correlation between the fusion track and the original track is obtained according to the first distance, the fourth distance and the ninth distance, and if the correlation is greater than a preset threshold, a track code of the original track is written into the fusion track matching parameter; Step 6: steps 1 to 5 are repeated until all original track data in the original track library are traversed, and track codes of all original tracks with a correlation greater than the preset threshold are written into the fusion track matching parameter; Step 7: a correlation accuracy, a positioning error improvement rate, a track integrity improvement rate and a track continuity improvement rate of the fusion track are calculated to evaluate the quality of the fusion track.

[0005] Further, the step 3 further comprises: if the first track point is the first track point in the fusion track, a second track point at the next time is selected from the fusion track, a track point with the minimum time difference from the next time is selected from the original track, a distance between the second track point and the first track point is calculated and recorded as a second distance, and the ninth distance is set to 0.

[0006] Further, the correlation between the fusion track and the original track obtained according to the first distance, the fourth distance and the ninth distance comprises: an average value of all first distances is obtained as a first average value; an average value of all fourth distances is obtained as a second average value; an average value of all ninth distances is obtained as a third average value; the correlation between the fusion track and the original track is obtained according to the first average value, the second average value and the third average value.

[0007] Further, the correlation between the fused track and the original track is obtained according to the first average value, the second average value and the third average value, and the correlation between the fused track and the original track comprises: The correlation between the fused track and the original track is obtained by the following formula:

[0008] wherein, , , the first average value, the second average value and the third average value respectively, is the correlation between the fused track and the original track.

[0009] Further, the correlation correct rate of the fused track is calculated, and the correlation correct rate of the fused track comprises: The source information of all original tracks in the fused track matching parameter is traversed, if the platforms of all original tracks are the same, the average weighted sum method is used to calculate the correlation correct rate of the fused track by using the weight and the correlation; If the platforms of the original tracks are not the same, and the number of different platforms does not exceed one half of the total number of original tracks corresponding to the fused track matching parameter, the platform with the largest number of original tracks is recorded, the number of original tracks of the platform is n, the number of original tracks a in all fused track matching parameters is counted, the average weight is 1 / a, the weight of the original track from the platform different from the platform with the largest number of original tracks is set to 1 / a P, and the remaining weight w is 1-1 / a P b, the weight of the original track of the platform with the largest number of original tracks is w / n; the correlation correct rate of the fused track is calculated by using the weight and the correlation; b is the number of original tracks of different platforms; If the platforms of the original tracks are not the same, and the number of different platforms exceeds one half and does not exceed two thirds of the total number of original tracks corresponding to the fused track matching parameter, the platform with the largest number of original tracks is recorded, the number of original tracks of the platform is n, the number of original tracks a in all fused track matching parameters is counted, the average weight is 1 / a, the weight of the original track from the platform different from the platform with the largest number of original tracks is set to 1 / a Q, and the remaining weight w is 1-1 / a Q b, the weight of the original track of the platform with the largest number of original tracks is w / n; the correlation correct rate of the fused track is calculated by using the weight and the correlation; If the platforms of the original tracks are not the same, and the number of different platforms exceeds two thirds of the total number of original tracks corresponding to the fused track matching parameter, it is indicated that the quality of the fused track does not meet the requirements.

[0010] Further, the positioning error improvement rate is calculated, comprising: corresponding to the platform recorded in the fusion track matching parameter of each fusion track point, extracting platform real track data of the same number of platforms most from the platform real track data set, respectively using the fusion track and the original track data, correlating the real track data in the real track data set according to the time parameter, calculating the positioning error of the track point in the fusion track corresponding to each time point, obtaining the average positioning error E1 of all track points in the fusion track and the average positioning error E2 of the original track corresponding to the platform, and the positioning error improvement rate is [1-(E1 / E2)]x100%.

[0011] Further, the track integrity improvement rate is calculated, comprising: The overall duration of the fusion track is recorded as T1, the same number of platforms most are recorded as T2, and the track integrity improvement rate is [(T1 / T2)-1]x100%.

[0012] Further, the track continuity improvement is calculated, comprising: The longest continuous duration of the fusion track is recorded as CT1, the longest continuous duration of the same number of platforms most is recorded as CT2, and the track continuity improvement rate is [(CT1 / CT2)-1]x100%.

[0013] Further, the quality of the fusion track is evaluated, comprising: The four indexes of correlation accuracy, positioning error improvement rate, track integrity improvement rate and track continuity improvement rate are comprehensively obtained by using equal weight weighting method, and the total score is obtained. The total score is compared with the preset quality threshold to evaluate whether the quality of the fusion track meets the standard.

[0014] The application also discloses a system for evaluating the quality of the fusion track based on track correlation, which realizes the method described above, and comprises: The selection module is used for selecting a group of original track data from the original track database, and setting the fusion track matching parameter to be empty. The first calculation module is used for selecting a first track point at the current time from the fusion track, selecting an original track point with the smallest time difference from the current time from the original track, and calculating the distance between the first track point and the original track point at the current time, which is recorded as the first distance. The fusion track is obtained by fusing all original track data in the original track database. The second calculating module is configured to: if the first track point is not the first track point in the fused track and is not the last track point in the fused track, select a second track point at a next time from the fused track, select a track point from the original track that has the smallest time difference with the second track point, calculate a distance between the first track point and the second track point, and record the distance as a second distance; calculate a distance between the track point at the current time and the track point at the next time, and record the distance as a third distance; calculate an absolute value of a difference between the second distance and the third distance, and record the absolute value as a fourth distance; The third calculating module is configured to: select a third track point at a previous time from the fused track, select a track point from the original track that has the smallest time difference with the third track point, calculate a distance between the first track point and the third track point, and record the distance as a fifth distance; calculate a distance between the track point at the current time and the track point at the previous time, and record the distance as a sixth distance; record a difference between the fifth distance and the sixth distance as a seventh distance; record a difference between the second distance and the third distance as an eighth distance; and record an absolute value of a difference between the eighth distance and the seventh distance as a ninth distance. The first judging module is configured to: until all track points in the fused track are traversed, obtain a correlation between the fused track and the original track according to the first distance, the fourth distance, and the ninth distance; and if the correlation is greater than a preset threshold, write a track code of the original track into the fused track matching parameter. The second judging module is configured to: until all original track data in the original track library are traversed, write track codes of all original tracks whose correlations are greater than the preset threshold into the fused track matching parameter. The quality evaluating module is configured to: calculate a correlation correct rate, a positioning error improvement rate, a track integrity improvement rate, and a track continuity improvement rate of the fused track, and evaluate a quality of the fused track.

[0015] Thanks to the above technical solutions, the present application has the following advantages: 1. The present application can evaluate the performance of the fusion algorithm without knowing the specific source of the fused data. The present application analyzes the correlation between the fused track and each original track, establishes a relevant track data set, compares the fused track data and the relevant track data, calculates the correlation correct rate, the positioning error improvement rate, the track integrity improvement rate, and the track continuity improvement rate, uses the weighted sum method to comprehensively evaluate the performance of the fusion algorithm, locates the defects of the fusion algorithm, guides the optimization of the fusion algorithm, and thus improves the quality of the fused track.

[0016] 2. The present application can still be applied in the case where the specific source of the fused data is unclear, reduces the system cost, and is suitable for promotion. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application. Other drawings can be obtained by those skilled in the art based on these drawings.

[0018] Figure 1 A flowchart of a method for evaluating fusion track quality based on track correlation according to an embodiment of the present application. DETAILED DESCRIPTION

[0019] The present application is further illustrated by the drawings and embodiments. The described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments obtained by those skilled in the art should be within the scope of protection of the embodiments of the present application.

[0020] Referring to Figure 1 The present application provides an embodiment of a method for evaluating fusion track quality based on track correlation, which comprises: Step 1: Select a group of original track data from the original track database, and set the fusion track matching parameters to be empty; Step 2: Select the first track point point00 at the current time time0 from the fusion track, and select the original track point point10 with the smallest time difference from the current time time0 from the original track. Calculate the distance between the first track point point00 at the current time time0 and the original track point, denoted as the first distance delta0. The fusion track is obtained by fusing all original track data in the original track database; Step 3: If the first track point point00 is not the first track point in the fusion track, and is not the last track point in the fusion track, then select the second track point point01 at the next time time1 from the fusion track, and select the original track point point11 with the smallest time difference from the next time time1 from the original track. Calculate the distance between the second track point point01 and the first track point point00, denoted as the second distance delta10. Calculate the distance between the original track points (point10 and point11) at the current time time0 and the next time time1, denoted as the third distance delta11. Calculate the absolute value of the difference between the second distance delta10 and the third distance delta11, denoted as the fourth distance delta1. Step 4: Select the third track point point02 from the merged track at the previous time 2. Select the original track point with the smallest time difference from the third track point point02 from the original track. Calculate the distance between the first track point point00 and the third track point point02, and record it as the fifth distance delta20. Calculate the distance between the original track point point10 at the current time 0 and the original track point point12 at the previous time 2, and record it as the sixth distance delta21. Record the difference between the fifth distance delta20 and the sixth distance delta21 as the seventh distance. Record the difference between the second distance delta10 and the third distance delta11 as the eighth distance. Record the absolute value of the difference between the eighth distance and the seventh distance as the ninth distance. ; Step 5: Repeat steps 2 to 4 until all track points in the fused track have been traversed. Based on the first distance delta0, the fourth distance delta1, and the ninth distance delta2, obtain the correlation between the fused track and the original track. If the correlation is greater than the preset threshold, write the track code of the original track into the fused track matching parameters. Step 6: Repeat steps 1 to 5 until all original track data in the original track database has been traversed, and write the track codes of all original tracks with a correlation greater than the preset threshold into the fusion track matching parameters. Step 7: Calculate the correlation accuracy, positioning error improvement rate, track integrity improvement rate, and track continuity improvement rate of the fused track to evaluate the quality of the fused track.

[0021] Optionally, step 3 further includes: If the first track point point00 is the first track point in the fused track, then select the second track point point01 at the next time 1 from the fused track, select the original track point with the smallest time difference with the next time 1 from the original track, calculate the distance between the second track point point01 and the first track point point00, and record it as the second distance delta10; and set the ninth distance delta2 to 0.

[0022] Optionally, obtaining the correlation between the fused track and the original track based on the first distance delta0, the fourth distance delta1, and the ninth distance delta2 includes: Calculate the average of all first distances delta0 to obtain the first average. Calculate the average of all fourth distances delta1 to obtain the second average; Calculate the average of all ninth distances delta2 to obtain the third average; The correlation between the fused track and the original track is obtained based on the first average, the second average, and the third average.

[0023] Optionally, obtaining the correlation between the fused track and the original track based on the first average, the second average, and the third average includes: The correlation between the fused track and the original track can be obtained using the following formula:

[0024] in, , , These are the first draw value, the second average value, and the third average value. To integrate the correlation between the original track and the flight path.

[0025] Optionally, calculating the association accuracy of the fused tracks includes: Iterate through the source information of all original tracks in the fusion track matching parameters. If all original tracks originate from the same platform, use the average weighted sum method and calculate the association accuracy of the fusion track by utilizing the weights and correlation. If the platforms from which the original tracks originate are different, and the number of different platforms does not exceed half of the total number of original track groups corresponding to the fused track matching parameters, record the platform with the most original tracks, with its original track count as n. Count the original track count 'a' among all fused track matching parameters, and then the average weight is 1 / a. Set the weight of original tracks originating from platforms different from the platform with the most original tracks to 1 / a. P, then the remaining weight w is 1 - 1 / a P b, the weight of the original track of the platform with the most original tracks is w / n; the association accuracy of the fused track is calculated using the weight and correlation; b is the number of original tracks of different platforms; P is a positive number, for example, the value of P can be 1.1; If the platforms from which the original tracks originate are different, and the number of different platforms exceeds half but does not exceed two-thirds of the total number of original track groups corresponding to the fused track matching parameters, then the platform with the most original tracks is recorded, and its number of original tracks is n. The number a of original tracks in all fused track matching parameters is counted, and the average weight is 1 / a. The weights of original tracks originating from platforms different from the platform with the most tracks are all set to 1 / a. Q, then the remaining weight w is 1 - 1 / a Q b. The weight of the original track of the platform with the most original tracks is w / n; the association accuracy of the fused track is calculated using the weight and correlation; Q is greater than P, for example, the value of Q can be 1.5; If the platforms from which the original tracks originate are different, and the number of different platforms exceeds two-thirds of the total number of original track groups corresponding to the fused track matching parameters, then the quality of the fused track does not meet the requirements.

[0026] Optionally, calculating the positioning error improvement rate includes: The platform corresponding to the original track code recorded in the fusion track matching parameters of each fusion track point is counted. The platform real trajectory data with the most identical data is extracted from the platform's real motion trajectory dataset. The fusion track and the original track data are used to associate the real motion trajectory data in the real motion trajectory dataset according to the time parameter. The positioning error of the trajectory point in the fusion track corresponding to each time point is calculated. The average positioning error E1 of all trajectory points in the fusion track and the average positioning error E2 of the corresponding platform original track are obtained. The positioning error improvement rate is [1-(E1 / E2)]×100%.

[0027] Optionally, calculating the track integrity improvement rate includes: The overall duration of the fused track is recorded as T1. The platform corresponding to the original track code recorded in the fused track matching parameters of each fused track point is counted. The longest duration of the platform with the most identical numbers is recorded as T2. The track integrity improvement rate is [(T1 / T2)-1]×100%.

[0028] Optionally, calculating the track continuity improvement includes: The longest continuous duration of the statistically fused track is denoted as CT1, and the longest continuous duration of the platform with the most statistically consistent data is denoted as CT2. The track continuity improvement rate is [(CT1 / CT2)-1]×100%.

[0029] Optionally, the evaluation of the quality of the fused tracks includes: The equal-weighted method is used to comprehensively evaluate four indicators: association accuracy, positioning error improvement rate, track integrity improvement rate, and track continuity improvement rate, to obtain a total score. The total score is then compared with a preset quality threshold to assess whether the quality of the fused track meets the standard.

[0030] This application also provides an embodiment of a system for evaluating the quality of fused tracks based on track correlation, which implements the method described in the above embodiment, and includes: The selection module is used to select a set of original track data from the original track database and set the fused track matching parameters to empty. The first calculation module is used to select the first track point at the current time from the fused track, select the original track point with the smallest time difference from the current time from the original track, and calculate the distance between the first track point at the current time and the original track point, which is denoted as the first distance; the fused track is obtained by fusing all the original track data in the original track database; The second calculation module is used to select the second track point for the next moment from the fused track if the first track point is neither the first nor the last track point in the fused track; select the original track point with the smallest time difference with the next moment from the original track; calculate the distance between the second track point and the first track point, denoted as the second distance; calculate the distance between the original track points at the current moment and the next moment, denoted as the third distance; and calculate the absolute value of the difference between the second distance and the third distance, denoted as the fourth distance. The third calculation module is used to select the third track point from the fused track at the previous moment, select the original track point with the smallest time difference from the third track point from the original track, calculate the distance between the first track point and the third track point, and record it as the fifth distance; calculate the distance between the original track point at the current moment and the previous moment, and record it as the sixth distance; record the difference between the fifth distance and the sixth distance as the seventh distance; record the difference between the second distance and the third distance as the eighth distance; and record the absolute value of the difference between the eighth distance and the seventh distance as the ninth distance. The first judgment module is used to obtain the correlation between the fused track and the original track based on the first distance, the fourth distance and the ninth distance until all track points in the fused track have been traversed. If the correlation is greater than a preset threshold, the track code of the original track is written into the fused track matching parameters. The second judgment module is used to write the track codes of all original track data with a correlation greater than a preset threshold into the fusion track matching parameters until all original track data in the original track database has been traversed. The quality assessment module is used to calculate the correlation accuracy, positioning error improvement rate, track integrity improvement rate, and track continuity improvement rate of the fused track, and to evaluate the quality of the fused track.

[0031] The present application has been described in detail above with reference to the accompanying drawings. However, it should be noted that the examples described above are merely preferred embodiments of the present application and are not intended to limit the present application. Various modifications and variations can be made to the present application by those skilled in the art. In setting the weights, in addition to the average weights in the examples of the present application, a custom weighting method can also be used. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for evaluating the quality of fused tracks based on track correlation, characterized in that, include: Step 1: Select a set of original track data from the original track database and set the fused track matching parameters to empty; Step 2: Select the first track point at the current time from the fused track, and select the original track point with the smallest time difference from the current time from the original track. Calculate the distance between the first track point at the current time and the original track point, and record it as the first distance. The fused track is obtained by fusing all the original track data in the original track database. Step 3: If the first track point is neither the first nor the last track point in the merged track, then select the second track point for the next time step from the merged track, select the original track point with the smallest time difference with the next time step from the original track, calculate the distance between the second track point and the first track point, and record it as the second distance. Calculate the distance between the original track points at the current time and the next time step, and record it as the third distance. Calculate the absolute value of the difference between the second distance and the third distance, and record it as the fourth distance. Step 4: Select the third track point from the fused track at the previous moment, and select the original track point with the smallest time difference from the third track point from the original track. Calculate the distance between the first track point and the third track point, and record it as the fifth distance. Calculate the distance between the original track points at the current moment and the previous moment, and record it as the sixth distance. Record the difference between the fifth distance and the sixth distance as the seventh distance; record the difference between the second distance and the third distance as the eighth distance; and record the absolute value of the difference between the eighth distance and the seventh distance as the ninth distance. Step 5: Repeat steps 2 to 4 until all track points in the fused track have been traversed. Based on the first distance, the fourth distance, and the ninth distance, obtain the correlation between the fused track and the original track. If the correlation is greater than a preset threshold, write the track code of the original track into the fused track matching parameters. Step 6: Repeat steps 1 to 5 until all original track data in the original track database has been traversed, and write the track codes of all original tracks with a correlation greater than the preset threshold into the fusion track matching parameters. Step 7: Calculate the correlation accuracy, positioning error improvement rate, track integrity improvement rate, and track continuity improvement rate of the fused track to evaluate the quality of the fused track.

2. The method according to claim 1, characterized in that, Step 3 also includes: If the first waypoint is the first waypoint in the fused waypoint, then select the second waypoint for the next time step from the fused waypoint, select the original waypoint with the smallest time difference with the next time step from the original waypoint, calculate the distance between the second waypoint and the first waypoint, and record it as the second distance; and set the ninth distance to 0.

3. The method according to claim 1, characterized in that, The process of obtaining the correlation between the fused track and the original track based on the first distance, the fourth distance, and the ninth distance includes: Calculate the average of all the first distances to obtain the first average. Calculate the average of all fourth distances to obtain the second average; Calculate the average of all ninth distances to obtain the third average; The correlation between the fused track and the original track is obtained based on the first average, the second average, and the third average.

4. The method according to claim 3, characterized in that, The process of obtaining the correlation between the fused track and the original track based on the first average, second average, and third average includes: The correlation between the fused track and the original track can be obtained using the following formula: in, , , These are the first draw value, the second average value, and the third average value. To integrate the correlation between the original track and the flight path.

5. The method according to claim 1, characterized in that, Calculating the association accuracy of the fused tracks includes: Iterate through the source information of all original tracks in the fusion track matching parameters. If all original tracks originate from the same platform, use the average weighted sum method and calculate the association accuracy of the fusion track by utilizing the weights and correlation. If the platforms from which the original tracks originate are different, and the number of different platforms does not exceed half of the total number of original track groups corresponding to the fused track matching parameters, record the platform with the most original tracks, with its original track count as n. Count the original track count 'a' among all fused track matching parameters, and then the average weight is 1 / a. Set the weight of original tracks originating from platforms different from the platform with the most original tracks to 1 / a. P, then the remaining weight w is 1 - 1 / a P b, the weight of the original track of the platform with the most original tracks is w / n; the association accuracy of the fused track is calculated using weight and correlation; b is the number of original tracks of different platforms; P is a positive number; If the platforms from which the original tracks originate are different, and the number of different platforms exceeds half but does not exceed two-thirds of the total number of original track groups corresponding to the fused track matching parameters, then the platform with the most original tracks is recorded, and its number of original tracks is n. The number a of original tracks in all fused track matching parameters is counted, and the average weight is 1 / a. The weights of original tracks originating from platforms different from the platform with the most tracks are all set to 1 / a. Q, then the remaining weight w is 1 - 1 / a Q b. The weight of the original track of the platform with the most original tracks is w / n; the association accuracy of the fused track is calculated using the weight and correlation; Q is greater than P; If the platforms from which the original tracks originate are different, and the number of different platforms exceeds two-thirds of the total number of original track groups corresponding to the fused track matching parameters, then the quality of the fused track does not meet the requirements.

6. The method according to claim 1, characterized in that, Calculating the positioning error improvement rate includes: The platform corresponding to the original track code recorded in the fusion track matching parameters of each fusion track point is counted. The platform real trajectory data with the most identical data is extracted from the platform's real motion trajectory dataset. The fusion track and the original track data are used to associate the real motion trajectory data in the real motion trajectory dataset according to the time parameter. The positioning error of the trajectory point in the fusion track corresponding to each time point is calculated. The average positioning error E1 of all trajectory points in the fusion track and the average positioning error E2 of the corresponding platform original track are obtained. The positioning error improvement rate is [1-(E1 / E2)]×100%.

7. The method according to claim 1, characterized in that, Calculating the track integrity improvement rate includes: The overall duration of the fused track is recorded as T1. The platform corresponding to the original track code recorded in the fused track matching parameters of each fused track point is counted. The longest duration of the platform with the most identical numbers is recorded as T2. The track integrity improvement rate is [(T1 / T2)-1]×100%.

8. The method according to claim 1, characterized in that, Calculating the track continuity improvement includes: The longest continuous duration of the statistically fused track is denoted as CT1, and the longest continuous duration of the platform with the most statistically consistent data is denoted as CT2. The track continuity improvement rate is [(CT1 / CT2)-1]×100%.

9. The method according to claim 1, characterized in that, The evaluation of the quality of the merged tracks includes: The equal-weighted method is used to comprehensively evaluate four indicators: association accuracy, positioning error improvement rate, track integrity improvement rate, and track continuity improvement rate, to obtain a total score. The total score is then compared with a preset quality threshold to assess whether the quality of the fused track meets the standard.

10. A system for evaluating the quality of fused tracks based on track correlation, implementing the method described in any one of claims 1-9, characterized in that, include: The selection module is used to select a set of original track data from the original track database and set the fused track matching parameters to empty. The first calculation module is used to select the first track point at the current time from the fused track, select the original track point with the smallest time difference from the current time from the original track, and calculate the distance between the first track point at the current time and the original track point, which is denoted as the first distance. The fused track is obtained by fusing all the original track data in the original track database; The second calculation module is used to select the second track point for the next moment from the fused track if the first track point is neither the first nor the last track point in the fused track; select the original track point with the smallest time difference with the next moment from the original track; calculate the distance between the second track point and the first track point, denoted as the second distance; calculate the distance between the original track points at the current moment and the next moment, denoted as the third distance; and calculate the absolute value of the difference between the second distance and the third distance, denoted as the fourth distance. The third calculation module is used to select the third track point from the fused track at the previous moment, select the original track point with the smallest time difference from the third track point from the original track, calculate the distance between the first track point and the third track point, and record it as the fifth distance; calculate the distance between the original track point at the current moment and the previous moment, and record it as the sixth distance; record the difference between the fifth distance and the sixth distance as the seventh distance; record the difference between the second distance and the third distance as the eighth distance; and record the absolute value of the difference between the eighth distance and the seventh distance as the ninth distance. The first judgment module is used to obtain the correlation between the fused track and the original track based on the first distance, the fourth distance and the ninth distance until all track points in the fused track have been traversed. If the correlation is greater than a preset threshold, the track code of the original track is written into the fused track matching parameters. The second judgment module is used to write the track codes of all original track data with a correlation greater than a preset threshold into the fusion track matching parameters until all original track data in the original track database has been traversed. The quality assessment module is used to calculate the correlation accuracy, positioning error improvement rate, track integrity improvement rate, and track continuity improvement rate of the fused track, and to evaluate the quality of the fused track.

Images