Micro-map personalized layout information mining method

Abstract

The application provides a micro-map personalized layout information mining method, relates to the related professional knowledge of image processing and cartography, and solves the problem of difficult mapping caused by the lack of professional map knowledge of ordinary people; the method comprises the following steps: obtaining the combination characteristics and position characteristics of map layout elements, obtaining the relationship characteristic information between the layout elements, and constructing a micro-map layout knowledge base; the map component personalized layout information mining method can effectively obtain the distribution characteristics of the map components in the map, provide help for the ordinary people who lack mapping basis, and improve the map production efficiency and map quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cartography and image processing, and particularly relates to a micro-map personalized layout information mining method. BACKGROUND

[0002] A map is one of the important tools for conveying information besides text and language. A traditional map is drawn by professional cartographers after understanding the cartographic requirements, and then professional cartographic tools are used for drawing. However, the users of the map are mostly ordinary people, which makes it difficult for the cartographers to fully understand the requirements of the users, and it is even more difficult to draw the ideal map in the minds of the users. We-Map is an extension of the traditional map in the era of self-media, and it provides the possibility for ordinary people to make their own maps. As a kind of "grass-roots" map for the general public, it has the characteristics of micro-content, low threshold, fast production and personalization, that is, the cartographer does not need to undergo professional training, and can draw and publish a map at any time and anywhere through a personal electronic device, and complete information transmission.

[0003] Layout is a key factor affecting information transmission, and high-quality map layout can effectively improve the communication efficiency of the map and enhance the visual appeal. For ordinary people who lack professional map knowledge, it is a very difficult task to select and layout map elements while considering the readability of the map, the transmission of information, and the beauty of the layout. However, since We-Map provides a low-threshold production method for cartographers, cartographers can delete map component elements and adjust the layout structure according to their personal wishes, which has a negative impact on the thematic expression of the map, the readability of the content, and the transmission of spatial information. However, providing a reasonable layout framework or guidance to the user is not considered. Therefore, it is an urgent problem to construct a micro-map element layout knowledge base to assist ordinary people in drawing maps.

[0004] The state invention patent CN202310098622.1 discloses an emergency map drawing method based on a set of drawing schemes, aiming at the problems of large variety demand of emergency map drawing and high drawing threshold, proposes to make corresponding drawing schemes for different emergency drawing scenarios respectively, and form a set of drawing schemes, the system can complete automatic drawing by selecting the drawing scheme matched with the drawing scenario, and improves the efficiency and timeliness of emergency map drawing. Although the method proposes map decoration rules, it only involves the font and style of decoration elements, and does not analyze the position characteristics of each element in the drawing and the spatial relationship between elements. At the same time, the method proposes to count the usage frequency of each material under various drawing scenarios, but only includes various drawing scenarios in emergencies and does not analyze the combination form of different materials under different scenarios. Therefore, the automatic emergency map drawing method proposed by the invention cannot effectively reduce the threshold of ordinary people lacking map professional knowledge to draw personalized maps.

[0005] The state invention patent CN202310183001.3 discloses a thematic map frame decoration method and system, aiming to automatically draw the standard map frame of certain specific thematic maps, avoid modifying the frame decoration effect multiple times, and improve the professionalization and automation level of map making. However, its production process is only for the standard style of certain specific thematic maps, such as explicitly specifying the content of left and right lower corner notes, and does not reduce the personalized map making technology, nor can it meet the needs of ordinary people for personalized map making technology. SUMMARY

[0006] The technical problem to be solved by the present application is to provide a micro map personalized layout information mining method to overcome the technical defect that ordinary people need to spend a lot of effort to adjust the layout when making maps due to lack of professional knowledge.

[0007] To solve the above problems, the present application provides a micro map personalized layout information mining method to reduce the difficulty of micro map drawing, solve the technical problem that the drawn map lacks readability and aesthetic due to lack of drawing knowledge, fill the gap of personalized map layout element rules in micro map drawing, provide scientific layout strategy for mass drawing, and thus improve the drawing efficiency and map quality.

[0008] To achieve the above goal, the present application adopts the following technical scheme:

[0009] A micro map personalized layout information mining method, characterized in that it comprises data screening, layout feature analysis and layout knowledge extraction, aiming to analyze the map cases that satisfy the public, and condense a set of micro map element layout rules suitable for ordinary people, to improve the effect of conveying information and enhancing design.

[0010] Preferably, the data screening, including map data screening and layout element screening, solves the problems of data non-compliance and data duplication, and provides basic data for layout element analysis;

[0011] The map data screening refers to selecting a set of personalized map data for feature analysis of map layout elements.

[0012] The layout element screening refers to selecting elements that affect map readability as the object of layout analysis and constructing layout element sample data.

[0013] Preferably, the layout feature analysis, including a layout element extraction and identification module, a basic feature analysis module, and a spatial relationship analysis module, is based on layout element sample data to detect and classify all basic components of the map, construct a basic feature analysis model, and further mine the relationship features between elements.

[0014] The layout element extraction and identification module refers to automatically extracting layout elements in the map using layout element sample data and classifying the layout elements.

[0015] The basic feature analysis module is used to obtain element combination features and position features.

[0016] The spatial relationship analysis module understands the potential position of layout elements by calculating the orientation, distance, and overlapping relationship between elements.

[0017] Preferably, the layout knowledge extraction module specifically includes feature screening and cluster analysis.

[0018] The feature screening refers to selecting layout features with higher public recognition according to user satisfaction evaluation results.

[0019] The cluster analysis refers to counting the frequency of each layout feature, with higher frequency representing higher recognition, sorting them according to frequency, and storing them in the knowledge base.

[0020] Preferably, the map data screening specifically includes a personalized map data set and a sample data set.

[0021] The personalized map data set refers to selecting personalized maps other than standard maps and map sets as the map data set for layout information mining, considering map readability.

[0022] The sample data set refers to selecting some representative maps as sample data sets to support layout element screening, in order to reduce the cost of manual annotation and improve the accuracy of automatic standards.

[0023] Preferably, the layout element extraction and identification module, specifically including layout element extraction and category identification, is used to automatically extract layout elements in the map, aiming to improve labeling efficiency and reduce labeling cost.

[0024] The layout element extraction refers to automatic segmentation of map entities according to layout element samples;

[0025] The layout element category identification refers to element classification by identifying text information in the layout element.

[0026] Preferably, the basic feature analysis module, specifically including type structure analysis and spatial structure analysis of layout elements, is used to obtain element combination features and position features;

[0027] The type structure analysis of the layout element refers to a calculation method of element type combination features contained in each map;

[0028] The spatial structure analysis of the layout element refers to a calculation method of position feature values of each layout element in the map, and a description of parameters involved in the calculation method.

[0029] Preferably, the spatial relationship analysis module, specifically including three sub-modules of orientation relationship, distance relationship and overlap relationship, is used to solve the problem that cartographers without professional knowledge cannot understand the spatial layout of the map and the depth sorting of map elements;

[0030] The layout element distance relationship refers to a method of calculating the Euclidean distance between elements;

[0031] The layout element orientation relationship refers to a method of calculating the eight-orientation relationship of elements;

[0032] The layout element overlap relationship refers to selecting overlap area, overlap rate and IoU as factors for analyzing the overlap relationship of elements, and a calculation method of two factors. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of the provided drawings.

[0034] Figure 1 The structure schematic diagram of the micro-map personalized layout information mining method provided by the present application;

[0035] Figure 2 The layout element position feature value calculation method and its description diagram provided by the present application;

[0036] Figure 3 Layout element position characteristic value schematic diagram provided by the present application;

[0037] Figure 4 Eight-direction cone model schematic diagram provided by the present application;

[0038] Figure 5 IoU example diagram provided by the present application. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0040] The embodiment of the present application discloses a micro-map personalized layout information mining method, characterized in that it comprises three parts of data screening, layout feature analysis and layout knowledge extraction (as shown in Figure 1 The embodiment of the present application discloses a micro-map personalized layout information mining method, characterized in that it comprises three parts of data screening, layout feature analysis and layout knowledge extraction (as shown in

[0041] In order to further optimize the above technical solutions, the data screening comprises map data screening and layout element screening, so as to solve the problems of data not meeting the requirements and data duplication, and provide basic data for layout element analysis;

[0042] The map data screening refers to selecting a set of personalized map data for feature analysis of map layout elements;

[0043] The layout element screening refers to selecting elements that have an impact on the readability of the map as the object of layout analysis, and constructing layout element sample data.

[0044] In order to further optimize the above technical solutions, the layout feature analysis comprises a layout element extraction and identification module, a basic feature analysis module and a spatial relationship analysis module, so as to detect and classify all basic components of the map based on the layout element sample data, construct a basic feature analysis model, and further mine the relationship features between elements;

[0045] The layout element extraction and identification module refers to automatically extracting layout elements in the map by using layout element sample data, and classifying the layout elements;

[0046] The basic feature analysis module is used to obtain element combination features and position characteristics;

[0047] The spatial relationship analysis module understands the potential position of layout elements by calculating the orientation, distance, and overlapping relationship between elements.

[0048] To further optimize the above technical solution, the layout knowledge extraction module specifically selects layout features with high public acceptance as layout knowledge and constructs a layout knowledge base based on this.

[0049] To further optimize the above technical solution, the map data screening specifically includes a personalized map data set and a sample data set.

[0050] The personalized map data set refers to selecting personalized maps other than standard maps and map sets as the map data set for mining layout information based on the consideration of map readability.

[0051] The sample data set refers to selecting some representative maps as sample data sets to support layout element screening in order to reduce the cost of manual annotation and improve the accuracy of automatic standards.

[0052] More specifically, the selection method of the sample data set refers to selecting the top 150 maps with the highest scores from the satisfaction S, functionality F, explanatory E, aesthetics A, authenticity R, and comprehensive I score results according to the user satisfaction evaluation results. Remove the map sets, incomplete map content, and duplicate maps that do not meet the requirements, and finally retain 400 maps as sample data.

[0053] Further, the user satisfaction evaluation only includes five evaluation indexes of satisfaction S, functionality F, explanatory E, aesthetics A, and authenticity R. According to the map readability requirement, a comprehensive index is added, and the comprehensive I is described as ;

[0054] More specifically, the layout elements in the layout element screening include map subjects, map names, legends, compasses, scales, explanatory information, and other elements.

[0055] To further optimize the above technical solution, the layout element extraction and recognition module specifically includes layout element extraction and category recognition to automatically extract layout elements in maps, aiming to improve annotation efficiency and reduce annotation cost.

[0056] The layout element extraction refers to automatically segmenting map entities based on layout element samples and automatically extracting layout elements in all maps based on manually annotated element sample data.

[0057] More specifically, the map subject is an instance segmentation result, and the map name, legend, compass, scale, explanatory information, and other elements are target detection results (i.e., the minimum bounding box of the element).

[0058] The layout element category recognition refers to element classification by recognizing text information in the layout element, recognizing the text information in the layout element by using an OCR technology, and performing element classification;

[0059] To further optimize the above technical solution, the basic feature analysis module specifically includes type structure analysis and spatial structure analysis of the layout element, to obtain element combination features and position features;

[0060] The type structure analysis of the layout element refers to a calculation method of element type combination features contained in each map, and the number of element types and element types contained in each map are obtained by statistics;

[0061] The spatial structure analysis of the layout element refers to a calculation method of position feature values of each layout element in the map and a description of parameters in the calculation method, as shown in FIGS. Figure 2 、 Figure 3 The feature values involved include a minimum bounding box, a center point, and a distance from an edge of a map.

[0062] To further optimize the above technical solution, the spatial relationship analysis module specifically includes three sub-modules of distance relationship, orientation relationship, and overlapping relationship. Considering that the layout elements of a map are not single individuals but a group composed of many individuals, the center point of the layout element is selected to calculate the distance and orientation relationship. Meanwhile, considering that a map has a feature that a main element is located in the middle of a map sheet and other layout elements are located at edges of the map sheet, the relationship between the main element and the other layout elements is calculated when calculating the distance, orientation, and overlapping relationship.

[0063] The distance relationship of the layout element refers to a method of calculating Euclidean distance between elements, and the Euclidean distance D between two elements is described as

[0064] (1)

[0065] Wherein, the center coordinates of the main element of the map are , and the center coordinates of the other element 1 are ;

[0066] The orientation relationship of the layout element refers to a method of calculating eight orientation relationships of elements, and the orientation relationship between the main element of the map and the other layout elements is obtained by using an eight-direction conical model.

[0067] More specifically, the angle between the main element of the map and the other layout elements is calculated, and the orientation relationship of the elements is obtained according to the eight-direction conical model, and the angle between the main element of the map and the other layout elements is described as

[0068] (2)

[0069] Wherein, the center coordinate of the map subject is The center coordinate of the remaining element 2 is ;

[0070] As shown in Figure 4 , in combination with the habit of the public describing the figure relationship, the 8 directions of the eight-direction conical model are {right, right-up, up, left-up, left, left-down, down, right-down};

[0071] The layout element overlapping relationship is designed to feature values including overlapping area, overlapping rate and IoU;

[0072] The overlapping area refers to being described as

[0073] (3)

[0074] Wherein, two sets A and B, which are respectively the map subject figure area and a certain remaining element figure area;

[0075] The overlapping rate is described as

[0076] (4)

[0077] Wherein, two sets A and B, which are respectively the map subject figure area and a certain remaining element figure area;

[0078] As shown in Figure 5 , the IoU is described as

[0079] (3)

[0080] Wherein, two sets A and B, which are respectively the map subject figure area and a certain remaining element figure area.

[0081] In order to further optimize the above technical scheme, the layout knowledge extraction module specifically includes feature screening and cluster analysis;

[0082] The feature screening refers to selecting the layout features corresponding to 1000 maps with higher comprehensiveness according to the user satisfaction evaluation results;

[0083] The cluster analysis refers to performing cluster analysis on the screened layout features, and counting the frequency of occurrence of each type of knowledge, sorting it according to the high and low of the frequency of occurrence, and storing it in the knowledge base.

[0084] The micro-map intelligent production system and device provided by the application have the following characteristics:

[0085] (1) The technical defect that the public needs to spend a lot of energy to adjust the layout when making a map due to lack of professional knowledge background is overcome, and the professional requirements for cartographers in micro-map making are reduced;

[0086] (2) The problem of time-consuming and laborious manual labeling of map layout elements is solved, providing a basis for layout analysis;

[0087] (3) The influence of subjectivity in knowledge extraction is reduced, and the layout knowledge is sorted through satisfaction evaluation results and frequency of use, effectively improving the universality of the knowledge.

[0088] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts between the various embodiments can be mutually referred to. For the device disclosed by the embodiments, since it corresponds to the method disclosed by the embodiments, the description is relatively simple, and the related parts can be referred to the method part.

[0089] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A micro-map personalized layout information mining method, characterized in that, Comprise: Data screening, layout feature analysis and layout knowledge extraction; The data screening comprises map data screening and layout element screening; The map data screening refers to selecting a set of personalized map data for feature analysis of map layout elements; The layout element screening refers to selecting elements that affect map readability as the object of layout analysis and constructing layout element sample data; The layout elements in the layout element screening include map body, map name, legend, compass, scale, explanatory information and other elements; The layout feature analysis comprises a layout element extraction and recognition module, a basic feature analysis module and a spatial relationship analysis module, which is based on layout element sample data to detect and classify the basic components of all maps, construct a basic feature analysis model, and further mine the relationship characteristics between elements; The layout element extraction and recognition module refers to automatically extracting layout elements in the map using layout element sample data and classifying the layout elements; The layout element extraction refers to automatically segmenting map entities based on layout element samples and automatically extracting layout elements in all maps based on manually labeled element sample data; The map body is the instance segmentation result, and the map name, legend, compass, scale, explanatory information and other elements are the target detection results; The layout element category recognition refers to classifying elements by recognizing the text information in the layout elements, and classifying elements according to the recognition results; The basic feature analysis module specifically includes type structure analysis and spatial structure analysis of layout elements to obtain element combination features and position features; The spatial relationship analysis module understands the potential position of layout elements by calculating the orientation, distance and overlap relationship between elements; The layout knowledge extraction module specifically selects layout features with high public acceptance as layout knowledge and constructs a layout knowledge base based on it.

2. The micro-map personalized layout information mining method of claim 1, wherein, The layout knowledge extraction module comprises: Feature screening refers to selecting layout features with high public acceptance according to user satisfaction evaluation results; Cluster analysis refers to counting the frequency of each layout feature, where higher frequency represents higher acceptance, and sorting them according to frequency and storing them in the knowledge base.

3. The micro map personalized layout information mining method according to claim 1, wherein the map data screening specifically comprises: The personalized map data set refers to selecting personalized maps other than standard maps and map sets as the map data set for mining layout information based on considering map readability; The sample data set refers to selecting some representative maps as sample data sets to support layout element screening in order to reduce the cost of manual labeling and improve the accuracy of automatic standards.

4. The micro map personalized layout information mining method according to claim 1, wherein the basic feature analysis module specifically comprises: The type structure analysis of layout elements refers to the calculation method of the element type combination features contained in each map; The spatial structure analysis of the layout element refers to a calculation method of a position characteristic value of each layout element in a map and a description of a parameter involved in the calculation method.

5. The micro-map personalized layout information mining method according to claim 1, wherein the spatial relationship analysis module specifically comprises: The distance relationship of the layout element refers to a method of calculating the Euclidean distance between elements; The orientation relationship of the layout element refers to a method of calculating the eight-orientation relationship of the elements; The overlapping relationship of the layout element refers to selecting an overlapping area, an overlapping rate and an IoU as factors for analyzing the overlapping relationship of the elements and a calculation method of the two factors.

Images