A method and system for re-optimizing map annotation configuration results

Abstract

The present invention discloses a method for re-optimizing the map annotation configuration result. For the initially determined annotation configuration result, by calculating and comparing the scoring values of the selected candidate positions of the elements with other candidate positions ranked higher, the annotations not at the theoretically optimal candidate positions are moved one by one, so as to achieve moving the element annotations with initially determined candidate positions one by one to make them approach the better candidate positions, ensuring that the configuration result can continuously approach the optimal annotation candidate position combination as much as possible, reducing the convergence value of the configuration result again, and achieving the purpose of improving the annotation configuration quality.

Description

Technical Field

[0001] The present invention belongs to the field of annotation configuration, and more specifically, particularly relates to a method for re-optimizing the result of map annotation configuration. At the same time, the present invention also relates to a system for re-optimizing the result of map annotation configuration. Background Art

[0002] The essence of annotation configuration is an NP-hard problem, which is to select one from all candidate positions of each feature for permutation and combination. In this process, generally, algorithmic loop iteration is used to screen out high-quality permutation and combination ways of annotation candidate positions. For example:

[0003] A map has 71 features to be annotated, and each feature generates 24 candidate positions. Then, the number of candidate position combination ways of all features will be 24 71 species (about 9.89E+97). If we want to use some current algorithms to find the optimal permutation and combination way, it is very difficult because the number of candidate position combination ways formed by feature candidate positions is an extremely large number. Therefore, only a sub-optimal solution can be found within a certain time, and the obtained configuration result is not the optimal one.

[0004] Therefore, it is necessary to explore other methods to optimize the already configured result, improve the speed of the algorithm for finding high-quality permutation and combination of annotation candidate positions, further optimize the quality of the annotation configuration result, and make the configuration result as close as possible to the optimal combination.

[0005] Currently, in the research of annotation configuration, common algorithms include genetic algorithm, greedy algorithm, backtracking algorithm, simulated annealing algorithm, ant colony algorithm, tabu search algorithm, discrete differential evolution and genetic algorithm (DDEGA). These algorithms mostly screen out high-quality permutation and combination ways of annotation candidate positions through loop iteration. Although they can solve the annotation configuration problem to a certain extent, when the algorithm score value drops to a certain level, it tends to converge, and it is very difficult to obtain a solution smaller than the convergence value. Only a sub-optimal solution can be found as the annotation configuration result. Therefore, we propose a method and system for re-optimizing the result of map annotation configuration. Summary of the Invention

[0006] The purpose of the present invention is to solve the deficiencies existing in the prior art, and to propose a method and system for re-optimizing the result of map annotation configuration, which can re-optimize the already configured result of the annotation, thereby improving the quality of map configuration.

[0007] To achieve the above purpose, the present invention provides the following technical solutions:

[0008] A method for re-optimizing the result of map annotation configuration includes the following steps:

[0009] S1. Input the preliminarily configured result list Outcome and the experimental map Map0;

[0010] The map Map0 contains different layers of points, lines, and surfaces. Each layer has multiple features, with a total of S features;

[0011] S2. Optimize the input annotation configuration result to obtain a new annotation configuration result Result, which contains the selected annotation candidate positions Name, Type, X, and Y data information for each feature. Result = [Name, Type, X, Y];

[0012] S3. Use V1, V2, and V3 to represent the scoring values of the annotation configuration result. At the same time, define a two-dimensional list Labels_2 to represent the set of all candidate positions for all features;

[0013] S4. Define the output parameter two-dimensional list Result, which contains the selected annotation candidate positions Name, Type, X, and Y data information for each feature. Result = [Name, Type, X, Y];

[0014] S5. Based on the map Map0, obtain the list Labels_2 of the candidate positions for each feature sorted by the theoretical limit value and the annotation configuration result Outcome, and replace the configuration result Outcome with Result;

[0015] S6. Calculate the annotation scoring value of the configuration result Result, represented by V1;

[0016] S7. Define V2 = V1;

[0017] S8. Find the features whose corresponding candidate positions from Label1 - Label S are not the theoretically optimal candidate positions;

[0018] S9. Move the candidate positions that are not at the theoretically optimal candidate positions for the features one by one, and define T1 = 1;

[0019] S10. End and output the re-optimized result Result of the annotation configuration.

[0020] Preferably, in step S1, the Outcome contains the selected annotation candidate positions Name, Type, X, and Y data information for each feature, denoted as: Outcome = [Name, Type, X, Y].

[0021] Preferably, in step S5, a list Labels_2 after sorting the candidate positions of each element according to the theoretical limit value and the annotation configuration result Outcome are obtained, and the DDEGA optimization algorithm is adopted. The DDEGA optimization algorithm is as follows:

[0022] Optimize the initial population, and improve the quality of the initial population by adding a local optimal permutation and combination method and controlling the range of randomly generated initial population.

[0023] Optimize the genetic process, change the proportion of generating new population individuals by the differential algorithm and the genetic algorithm, and add a part of individuals randomly generated from high-quality candidate positions.

[0024] Preferably, the list Labels_2 is specifically:

[0025] Labels_2 = [Label 1-[1] , Label 1-[2] , Label 1-[3] ,......, Label 1-[N*M] , Label 2-[1] , Label 2-[2] , Label 2-[3] , Label 2-[N*M] ,......, Label S-[1] , Label S-[2] , Label S-[3] ,......, Label S-[N*M] ;

[0026] Where Label S-[N*M] represents the candidate position information set ranked N*M in the theoretical limit value F in the S-th element; and replace the configuration result Outcome with Result.

[0027] Preferably, in step S8, the elements whose candidate positions corresponding to Label1 - Label S are not the theoretically optimal candidate positions are represented by the list Y[Label P1-x , Label P2-x , Label P3 -x,......, Label Pn-x .

[0028] Preferably, the specific process of defining T1 = 1 in step S9 is as follows:

[0029] 1) Define T = T1, and calculate the annotation score V3 of the T-th candidate position Label P1 of the element Label in Y and the candidate positions of other elements. P1-T

[0030] If the score value is less than V2, replace the Label in Result P1-x with Label P1-T , and let V3 overwrite the original value of V2, V2 = V3, then go to step 3), and move the candidate position of the next element;

[0031] If V3 is greater than V2, go to step 2);

[0032] 2) Define T1 = T1 + 1, and determine whether T1 is greater than the x value of Label P1-x ;

[0033] If T1 is less than x, go to step 1), and determine whether the next candidate position is better than the existing candidate position;

[0034] If T1 is greater than or equal to x, go to step 3);

[0035] 3) Move the selected candidate position of the next element in Y according to the method in step 1) until all the candidate positions of the elements are moved, and then determine whether Result changes;

[0036] If Result changes, calculate the annotation score value of the configuration result Result again and start the next move;

[0037] If Result does not change, output the re-optimized result Result of the annotation configuration.

[0038] Preferably, in step S7, the specific method for defining V2 = V1 is:

[0039] According to the candidate position coordinates X and Y of each element in Result, find the permutation serial number corresponding to the candidate position configuration result of the element in the permutation serial number Labels_2 generated by sorting the candidate positions of each element obtained based on the map Map0 according to the theoretical limit value, and use Label 1-X -Label S-X to represent.

[0040] A system for re-optimizing the result of map annotation configuration includes:

[0041] A central processing unit and a memory;

[0042] The memory is a transient storage memory or a persistent storage memory;

[0043] The central processing unit is configured to communicate with the memory and execute the instruction operations in the memory on the system for re-optimizing the result of map annotation configuration to execute the method for re-optimizing the result of map annotation configuration.

[0044] A computer-readable storage medium includes instructions that, when run on a computer, cause the computer to execute a method for re-optimizing based on map annotation configuration results.

[0045] Technical effects and advantages of the present invention: A method for re-optimizing based on map annotation configuration results provided by the present invention, compared with traditional methods, for the initially determined annotation configuration results, by calculating and comparing the score values of the selected candidate positions of elements with other candidate positions ranked higher, to move one by one the annotations that are not in the theoretically optimal candidate positions, and then by moving one by one the element annotations with initially determined candidate positions towards more optimal candidate positions, so as to ensure that the configuration results can approach the optimal annotation candidate position combination as much as possible, and reduce the convergence value of the configuration results again, achieving the purpose of improving the quality of annotation configuration. Description of the Drawings

[0046] Figure 1 It is a flowchart of the method for re-optimizing based on map annotation configuration results of the present invention. Detailed Embodiments

[0047] To make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the present invention to be protected, but only represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0048] The present invention provides a method for re-optimizing based on map annotation configuration results as Figure 1 described below, including input description, output description, and annotation configuration optimization process. The detailed process of the technical solution is described as follows:

[0049] (1) Input description

[0050] The input is a list of the initially configured results Outcome and the experimental map Map0. Outcome contains the selected annotation candidate position Name, Type, X, and Y data information of each element. Outcome = [Name, Type, X, Y], as shown in the following table:

[0051] Table 1 Output candidate position coordinate information Outcome

[0052] Field Name Field Meaning Field Type Field Description Name Element Name TEXT Name information of the element, Type Element Type TEXT Divided into three types: point, polyline, and polygon X Abscissa DECIMAL Abscissa information of the center point of the determined candidate position Y Ordinate DECIMAL Ordinate information of the center point of the determined candidate position

[0053] The map Map0 contains different layers of points, lines, and surfaces. Each layer has multiple elements, with a total of S elements.

[0054] (2) Output description

[0055] The output is based on the technical method of this invention patent, optimizing the input annotation configuration result to obtain a new annotation configuration result Result, which contains the selected annotation candidate position Name, Type, X, and Y data information for each element. Result = [Name, Type, X, Y], as shown in the following table:

[0056] Table 2 Optimized output candidate position coordinate information Result

[0057] Field Name Field Meaning Field Type Field Description Name Element Name TEXT Name information of the element, Type Element Type TEXT Divided into three types: point, polyline, and polygon X Abscissa DECIMAL Abscissa information of the center point of the determined candidate position Y Ordinate DECIMAL Ordinate information of the center point of the determined candidate position

[0058] (3) Annotation configuration optimization process

[0059] (3-1) Definition of parameters

[0060] (3-1-1) Define V1, V2, V3, all used to represent the scoring values of the annotation configuration results; go to (3-1-2);

[0061] (3-1-2) Define the two-dimensional list Labels_2, used to represent the set of all candidate positions of all elements; go to (3-1-3);

[0062] (3-1-3) Define the output parameter two-dimensional list Result, which contains the selected annotation candidate position Name, Type, X, and Y data information for each element. Result = [Name, Type, X, Y]; go to (3-2);

[0063] (3-2) Based on the map Map0, obtain the list Labels_2 of the candidate positions of each element sorted by the theoretical limit value and the annotation configuration result Outcome, and replace the configuration result Outcome with Result; among them, to obtain the list Labels_2 of the candidate positions of each element sorted by the theoretical limit value and the annotation configuration result Outcome, the DDEGA optimization algorithm is used. The DDEGA optimization algorithm is as follows:

[0064] Optimize the initial population by adding a local optimal permutation and combination method and controlling the range of randomly generated initial populations to improve the quality of the initial population;

[0065] Optimize the genetic process, change the ratio of generating individuals in the new population by the differential algorithm and the genetic algorithm, and add a part of individuals randomly generated from high-quality candidate positions;

[0066] Further, the list Labels_2 is specifically:

[0067] Labels_2 = [Label 1-[1] , Label 1-[2] , Label 1-[3] ,......, Label 1-[N*M]， Label 2-[1] , Label 2-[2] , Label 2-[3] , Label 2-[N*M] ,......, Label S-[1] , Label S-[2] , Label S-[3] ,......, Label S-[N*M] ;

[0068] Among them, Label S-[N*M] represents the set of candidate position information ranked N*M in the theoretical limit value F of the S-th element; and replace the configuration result Outcome with Result;

[0069] (3-3) Calculate the annotation score value of the configuration result Result, denoted as V1, and enter (3-4);

[0070] (3-4) V2 = V1; According to the candidate position coordinates X and Y of each element in Result, find out the corresponding permutation numbers of the candidate position configuration results of the elements in the permutation numbers Labels_2 generated in (3-2), denoted as Label 1-X -Label S-X ; Enter (3-5);

[0071] (3-5) Find out the elements whose corresponding candidate positions of Label1-Label S are not the theoretical optimal candidate positions (the candidate position ranks first), and represent them with the list Y[Label P1-x , Label P2-x , Label P3-x ,......, Label Pn-x , where P1, P2, P3,......, Pn represent the elements not at the theoretical optimal candidate positions, and x represents the actual candidate position number of the corresponding element; Enter (3-6);

[0072] (3 - 6) Move the candidate positions that are not at the theoretically optimal element candidate positions one by one; define T1 = 1;

[0073] (3 - 6 - 1) T = T1, calculate the note score V3 of the element Label in Y P1 The Label of the T - th candidate position P1-T and the note scores of other element candidate positions; if the score value is less than V2, replace the Label in Result P1-x with Label P1-T , and let V3 overwrite the original V2 value, V2 = V3, then enter (3 - 6 - 3) to move the candidate position of the next element; if V3 is greater than V2, enter (3 - 6 - 2);

[0074] (3 - 6 - 2) T1 = T1 + 1, determine whether T1 is greater than the x value of Label P1-x ; if T1 is less than x, enter (3 - 6 - 1) to determine whether the next candidate position is better than the existing candidate position; if T1 is greater than or equal to x, enter (3 - 6 - 3);

[0075] (3 - 6 - 3) Move the selected candidate position of the next element in Y according to the method in (3 - 6 - 1) until all element candidate positions are moved. Then, determine whether Result has changed. If there is a change, enter the calculation of the note score value of the configuration result Result again to start the next move; if Result has not changed, enter the output of the re - optimized result Result of the note configuration;

[0076] (3 - 7) End and output the re - optimized result Result of the note configuration.

[0077] The above method moves the element notes at the initially determined candidate positions one by one, making it approach a better candidate position, so as to ensure that the configuration result can continuously approach the optimal note candidate position combination as much as possible, further reduce the convergence value of the configuration result, and achieve the purpose of improving the quality of the note configuration.

[0078] This embodiment also provides a system for re - optimizing the result of map note configuration, including:

[0079] A central processing unit and a memory;

[0080] The memory is a transient storage memory or a persistent storage memory;

[0081] The central processing unit is configured to communicate with the memory and execute the instruction operations in the memory on the system for re - optimizing the result of map note configuration to execute the method for re - optimizing the result of map note configuration.

[0082] A computer-readable storage medium includes instructions that, when run on a computer, cause the computer to execute a method for re-optimizing based on a map annotation configuration result.

[0083] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for re - optimizing the configuration result of map annotation, characterized in that, It includes the following steps: S1. Input the initially configured Outcome list of results and the experimental map Map0; The map Map0 contains different layers of points, lines, and surfaces, and each layer has multiple elements, with a total of S elements; S2. Optimize the input annotation configuration result to obtain a new annotation configuration result Result, which contains the selected annotation candidate positions Name, Type, X, and Y data information for each element, and Result = [Name, Type, X, Y]; S3. Use V1, V2, and V3 to represent the scoring values of the annotation configuration result. At the same time, define a two-dimensional list Labels_2 to represent the set of all candidate positions for all elements; S4. Define the output parameter two-dimensional list Result, which contains the selected annotation candidate positions Name, Type, X, and Y data information for each element, and Result = [Name, Type, X, Y]; S5. Based on the map Map0, obtain the list Labels_2 of candidate positions for each element sorted by the theoretical limit value and the annotation configuration result Outcome, and replace the configuration result Outcome with Result; S7. Calculate the annotation scoring value of the configuration result Result, represented by V1; S8. Define V2 = V1; S8. Identify the elements where the corresponding candidate positions of Label1 - Label S are not the theoretically optimal candidate positions; S9. Move the candidate positions that are not at the theoretically optimal element candidate positions one by one, and define T1 = 1; S10. End and output the re-optimized result Result of the annotation configuration.

2. The method for further optimizing based on the map annotation configuration result according to claim 1, characterized in that: In step S1, the Outcome contains the selected annotation candidate positions Name, Type, X, and Y data information for each element, denoted as: Outcome = [Name, Type, X, Y].

3. The method for further optimizing a map annotation configuration result according to claim 1, wherein: In step S5, to obtain the list Labels_2 of candidate positions for each element sorted by the theoretical limit value and the annotation configuration result Outcome, the DDEGA optimization algorithm is used. The DDEGA optimization algorithm is as follows: Optimize the initial population by adding a local optimal permutation and combination method and controlling the range of randomly generated initial populations to improve the quality of the initial population; Optimize the genetic process by changing the proportion of new population individuals generated by the differential algorithm and the genetic algorithm, and adding a part of individuals randomly generated from high-quality candidate positions.

4. A method for further optimizing the map annotation configuration result according to claim 3, characterized in that: The list Labels_2 is specifically as follows: Labels_2 = [Label 1-[1] , Label 1-[2] , Label 1-[3] ,......, Label 1-[N*M] , Label 2-[1] , Label 2-[2] , Label 2-[3] , Label 2-[N*M] ,......, Label S-[1] , Label S-[2] , Label S-[3] ,......, Label S-[N*M] ; Among them, Label S-[N*M] represents the set of candidate position information where the theoretical limit value F ranks N*M in the S-th element; and replace the configuration result Outcome with Result.

5. A method for further optimizing the map annotation configuration result according to claim 1, characterized in that: In step S8, the elements where the candidate positions corresponding to Label1 - Label S are not the theoretically optimal candidate positions are represented by the list Y[Label P1-x , Label P2-x , Label P3-x ,......, Label Pn-x , where P1, P2, P3,......, Pn represent the elements not at the theoretically optimal candidate positions, and x represents the actual candidate position serial number of the corresponding element.

6. A method for further optimizing the map annotation configuration result according to any one of claims 1 or 5, characterized in that: The specific process of defining T1 = 1 in step S9 is as follows: 1) Define T = T1 and calculate the element Label in Y P1 The Label at the T-th candidate position P1-T and the annotation score V3 at the candidate positions of other elements; If the scoring value is less than V2, replace the Label in Result P1-x with Label P1-T , and let V3 overwrite the original value of V2, V2 = V3, then go to step 3), move the candidate position of the next element; If V3 is greater than V2, go to step 2); 2) Define T1 = T1 + 1, and determine whether T1 is greater than the x value of Label P1-x ; If T1 is less than x, go to step 1) to determine whether the next candidate position is better than the existing candidate position; If T1 is greater than or equal to x, go to step 3); 3) Move the selected candidate position of the next element in Y according to the method in step 1) until all element candidate positions are moved, and then determine whether Result changes; If Result changes, calculate the annotation scoring value of the configuration result Result again and start the next move; If Result does not change, go to output the re-optimized result Result of the annotation configuration.

7. A method for further optimizing the map annotation configuration result according to claim 1, characterized in that: In step S7, the specific method of defining V2 = V1 is as follows: According to the candidate position coordinates X and Y of each element in Result, find the permutation number corresponding to the candidate position configuration result of the element in the permutation number Labels_2 generated by the list Labels_2 obtained by sorting the candidate positions of each element based on the map Map0 according to the theoretical limit value, and use Label 1-X -Label S-X to represent.

8. A system for re-optimizing map annotation configuration results, characterized in that, It includes: Central processing unit, memory; The memory is a transient storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instruction operations in the memory on the system for re-optimizing based on map annotation configuration results to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: Includes instructions that, when run on a computer, cause the computer to execute the method according to any one of claims 1 to 7.

Images