Method for generating position reference lines, method and device for determining position of points of interest

Abstract

The application discloses a position reference line generation method, a point of interest position determination method and device. The position reference line generation method comprises the following steps: acquiring a closed area located at a preset collection side of a collection road based on map data; acquiring a closed area in which a point of interest can fall as a target closed area; and generating a position reference line of the collection road based on the target closed area. The position reference line can be provided for the delivery of the point of interest, so as to overcome the problem that the coordinates of the collected point of interest are inaccurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electronic map, in particular to a position reference line generation method, a point of interest position determination method and device. BACKGROUND

[0002] A point of interest (POI) in an electronic map can correspond to a house, a shop, a mailbox, a bus stop, a tourist attraction, etc. in the real world. Currently, a map data producer collects POIs through crowdsourcing to obtain collection data for generating POIs, which at least includes images recording POI names, addresses, etc. collected by a camera of a collection device, and position information of the POIs collected by a positioning module of the collection device. The position information of an outdoor POI is generally collected by a GPS receiver integrated in the collection device to collect GPS position information of the POI. However, the latitude and longitude coordinates recorded in the GPS position information of the POI may drift, lag, etc. due to the influence of environmental factors, which may cause the map position coordinates of the POI produced finally to deviate from the actual position coordinates. How to eliminate the deviation to make the map position coordinates of the POI consistent with the actual position coordinates is a problem to be solved by those skilled in the art. SUMMARY

[0003] In view of the above problems, the present application is proposed to provide a position reference line generation method, a point of interest position determination method and device which can overcome the above problems or at least partially solve the above problems.

[0004] In a first aspect, an embodiment of the present application provides a position reference line generation method, comprising:

[0005] Based on map data, an enclosed area located at a preset collection side of a collection road is acquired;

[0006] An enclosed area in which a point of interest can fall is acquired as a target enclosed area;

[0007] Based on the target enclosed area, a position reference line of the collection road is generated.

[0008] In some optional embodiments, based on map data, an enclosed area located at a preset collection side of a collection road is acquired, comprising:

[0009] From the map data, a road located at the preset collection side of the collection road and having a distance to the collection road not exceeding a preset distance threshold is acquired as a target road;

[0010] From the map data, a building block located on a preset collection side of the collection road and having a distance from a lower edge line of the building block to the collection road not exceeding a preset distance threshold is obtained as a target building block, the lower edge line of the building block being a closest edge line along the edge lines of the collection road;

[0011] Based on the target road and the target building block, a geographical area located on a preset collection side of the collection road is divided into one or more closed areas.

[0012] In some optional embodiments, based on the target road and the target building block, a geographical area located on a preset collection side of the collection road is divided into one or more closed areas, specifically including:

[0013] An area covered by each target building block is determined as a closed area;

[0014] If there is a target road along a direction away from the collection road on an upper edge line of the target building block, the target road and the upper edge line and the left and right edge lines of the target building block along the extension of the direction away from the collection road form a closed area;

[0015] If there is another target building block along a direction away from the collection road on an upper edge line of the target building block, the extension of the upper edge line and the left and right edge lines of the target building block and the lower edge line of the another target building block form a closed area;

[0016] If there is a target road between the lower edge line of the target building block and the collection road, the target road and the lower edge line and the left and right edge lines of the target building block along the extension of the direction close to the collection road form a closed area;

[0017] If there is no target road between the lower edge line of the target building block and the collection road, the extension of the lower edge line and the left and right edge lines of the target building block along the direction close to the collection road and the collection road form a closed area;

[0018] If there is another target building block within a preset distance along the direction of the collection road, a closed area is obtained according to the positional relationship of the edge lines of the two target building blocks.

[0019] In some optional embodiments, the closed area in which a building block can be located is obtained as a target closed area, specifically including:

[0020] When a building block is searched in the closed area, the closed area is determined as a closed area in which a building block can be located;

[0021] When no building blocks are found within the enclosed area, satellite imagery of the enclosed area is acquired. Using a pre-trained building classification model, if a building is identified within the satellite imagery, the enclosed area is determined to be an enclosed area where points of interest can be located.

[0022] In some optional embodiments, the building classification model is pre-trained in the following manner:

[0023] Label satellite image samples to indicate the presence or absence of buildings;

[0024] From the labeled satellite image samples, obtain satellite image samples labeled with buildings and satellite image samples labeled without buildings at a set ratio;

[0025] The building classification model is trained using the acquired satellite image samples.

[0026] In some optional embodiments, it also includes:

[0027] If the building classification model cannot identify whether there are buildings in the satellite image, it is determined whether the closed area is a closed area enclosed by the up and down roads. If so, the attribute parameters of the closed area are obtained, and a pre-trained decision tree is used to determine whether there are buildings in the closed area.

[0028] In some optional embodiments, the decision tree is pre-trained in the following manner:

[0029] Label the presence or absence of buildings in closed area samples divided based on map data;

[0030] From the marked closed areas, obtain a set proportion of closed area samples with buildings marked and closed area samples without buildings marked;

[0031] The decision tree is trained using the attribute parameters of the obtained closed region samples.

[0032] In some optional embodiments, generating a location reference line for the data collection road based on the target enclosed area includes:

[0033] Obtain perpendicular lines to the collection road within the target enclosed area at preset intervals, and obtain reference points for each perpendicular line in the direction close to the collection road.

[0034] The line connecting the reference points is used as a position reference line for the collected road.

[0035] In some optional embodiments, after obtaining the closed area located on the preset collection side of the collection road, the method further includes:

[0036] Delete closed regions whose area is smaller than a preset area threshold.

[0037] Secondly, embodiments of the present invention provide a method for determining the location of a point of interest, comprising:

[0038] The position reference line is generated using the method described above.

[0039] Determine the projection point of the point of interest onto the nearest reference line, and use the projection point as the location point of the point of interest.

[0040] Thirdly, embodiments of the present invention provide a device for generating a position reference line, comprising:

[0041] The first acquisition module is used to acquire a closed area located on a preset acquisition side of the acquisition road based on map data;

[0042] The second acquisition module is used to acquire closed regions where points of interest can be located as target closed regions.

[0043] The generation module is used to generate a location reference line for the collected road based on the target closed area obtained by the second acquisition module.

[0044] Fourthly, embodiments of the present invention provide a device for determining the location of a point of interest, comprising:

[0045] The generation module is used to generate position reference lines using the aforementioned method for generating position reference lines;

[0046] The projection module is used to determine the projection point of the point of interest on the nearest location reference line, and to use the projection point as the location point of the point of interest.

[0047] Fifthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the above-described method for generating a position reference line or the above-described method for determining the position of a point of interest.

[0048] In a sixth aspect, embodiments of the present invention provide a server, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described method for generating a position reference line or the above-described method for determining the position of a point of interest.

[0049] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0050] The method for generating location reference lines provided in this invention, based on map data, obtains a closed area on a preset collection side of the collection road; obtains a closed area where points of interest can be placed as a target closed area; and generates a location reference line for the collection road based on the target closed area. This method can provide location reference lines for the placement of points of interest, overcoming the problem of inaccurate coordinates of the collected points of interest.

[0051] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0052] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0053] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0054] Figure 1 This is a flowchart of the method for generating position reference lines in Embodiment 1 of the present invention;

[0055] Figure 2 This is a flowchart illustrating the training process of the building classification model in Embodiment 1 of the present invention.

[0056] Figure 3 This is an example diagram of position reference line generation in Embodiment 1 of the present invention;

[0057] Figure 4 This is a flowchart of the method for determining the location of a point of interest in Embodiment 2 of the present invention;

[0058] Figure 5 This is an example diagram of the closed region division in Embodiment 2 of the present invention;

[0059] Figure 6 This is an example diagram of another closed region division in Embodiment 2 of the present invention;

[0060] Figure 7 This is a flowchart illustrating the training process of the decision tree in Embodiment 2 of the present invention;

[0061] Figure 8 This is a schematic diagram of the structure of the position reference line generation device in an embodiment of the present invention;

[0062] Figure 9 This is a schematic diagram of the structure of the device for determining the location of a point of interest in an embodiment of the present invention. Detailed Implementation

[0063] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0064] To address the problem of inaccurate coordinates of points of interest (POIs) in existing technologies, this invention provides a method for generating location reference lines, a method for determining the location of POIs, and an apparatus that can provide reference lines for the placement of POIs, thereby overcoming the problem of inaccurate coordinates of the collected POIs.

[0065] Example 1

[0066] Embodiment 1 of the present invention provides a method for generating position reference lines. The generated position reference lines are mainly used to determine the position of points of interest. The process is as follows: Figure 1 As shown, it includes the following steps:

[0067] Step S11: Based on map data, obtain the closed area located on the preset collection side of the collection road.

[0068] Specifically, based on the pre-specified data collection roads, multiple closed areas are divided on the preset data collection side of the data collection roads in the electronic map.

[0069] Among them, the collection road is the reference road used to determine the target closed area where the point of interest is placed. The collection road can be determined in advance based on information such as the length of the road, the direction of the road extension, and the node data of the road. Generally, roads with a longer length, an extension method close to a straight line, and a node data distribution density that meets the requirements are selected as the collection road.

[0070] To obtain a closed area located on a preset collection side of the collection road, the area to be divided can be determined within a preset distance threshold range on the electronic map. Within this area, closed features are searched, and multiple closed areas are divided based on these features and the collection road. The preset collection side of the collection road is either the left or right side, generally the side with more buildings. The preset distance threshold can be determined based on experience; typically, it can be set to 60 meters. Closed features can be roads, buildings, or waterways, etc.

[0071] The specific method for dividing the enclosed area will be detailed in the subsequent second embodiment.

[0072] Step S12: Obtain the closed region where points of interest can be landed as the target closed region.

[0073] In one embodiment, the method may include: when a building is found within a closed area, the closed area is determined to be a closed area where a point of interest can be placed; when no building is found within the closed area, satellite imagery of the closed area is acquired, and when a pre-trained building classification model is used to identify buildings within the satellite imagery, the closed area is determined to be a closed area where a point of interest can be placed.

[0074] The above building classification model refers to Figure 2 As shown, it can be pre-trained in the following way:

[0075] Step S21: Label the satellite image samples to indicate whether or not buildings are present.

[0076] Satellite imagery information is downloaded using the coordinates of the bounding box of a closed region and then processed into satellite imagery at a preset resolution, such as a 224x224 image. A binary classification method can be used to label the presence or absence of buildings in multiple satellite imagery samples.

[0077] Step S22: From the labeled satellite image samples, obtain satellite image samples labeled with buildings and satellite image samples labeled without buildings at a set ratio.

[0078] Because enclosed areas are usually more likely to contain buildings, it is necessary to filter the multiple labeled satellite image samples. The ratio of the filtered satellite image samples labeled with buildings to those labeled without buildings can be 1:1.

[0079] Step S23: Train a building classification model using the acquired satellite image samples.

[0080] For closed areas lacking building information, satellite imagery can be used to supplement the information to determine whether buildings exist. Closed areas where buildings are ultimately identified can be designated as target closed areas where points of interest can be landed.

[0081] Specifically, the building classification model mentioned above can be the GoogleNet model or other classification models.

[0082] Step S13: Generate a location reference line for the road based on the target closed area.

[0083] In one embodiment, it may be that, at preset intervals, perpendicular lines to the acquisition road are obtained within the target enclosed area, and reference points for the perpendicular lines are obtained in the direction close to the acquisition road on each perpendicular line; the line connecting the reference points is used as a position reference line for the acquisition road.

[0084] Specifically, the reference points mentioned above can be the points where each vertical line is equally divided, or points on each vertical line at a set distance from the road being collected. This implementation does not limit the specific rules for determining the reference points.

[0085] Reference Figure 3 As shown, for the target closed area ABCD located on the collection side of the collection road L, perpendicular lines to the collection road L are obtained in the target closed area ABCD at preset intervals, namely perpendicular lines E1, E2 and E3... On each perpendicular line, 1 / 3 reference points are obtained in the direction close to the collection road, and equal division points F1, F2 and F3... are obtained; the line connecting the equal division points is used as a position reference line F of the collection road L.

[0086] The above-described solution provided in this embodiment of the invention, based on map data, obtains a closed area located on a preset collection side of the collection road; obtains a closed area where points of interest can be placed as a target closed area; and generates a location reference line for the collection road based on the target closed area. This provides a location reference line for the placement of points of interest, overcoming the problems of inaccurate coordinates or even points of interest crossing roads.

[0087] Example 2

[0088] Embodiment 2 of the present invention provides a specific implementation flow of a method for generating position reference lines, referring to... Figure 4 As shown, it includes the following steps:

[0089] Step S41: Based on map data, obtain the closed area located on the preset collection side of the collection road.

[0090] In one embodiment, a road located on a preset collection side of the collection road and whose distance to the collection road does not exceed a preset distance threshold can be obtained from map data as a target road; a building located on the preset collection side of the collection road and whose lower edge of the building is located on the preset collection side of the collection road and whose distance to the collection road does not exceed a preset distance threshold can be obtained from map data as a target building; and based on the target road and the target building, the geographical area located on the preset collection side of the collection road can be divided into one or more closed areas.

[0091] Specifically, the bottom edge of the target building is the edge line closest to the data collection road among all the edges along the data collection road.

[0092] The aforementioned target roads and target buildings are the closing elements of the closed area. Optionally, water systems located on a preset collection side of the collection road, within a preset distance threshold from the collection road, can also be used as target water systems and treated as closing elements of the closed area. Finally, based on each closing element and the collection road, the geographical area on the preset collection side of the collection road is divided into more than one closed area.

[0093] Based on the target road and target building, the geographical area located on the preset collection side of the collection road is divided into more than one closed area. Specifically, this may include the following situations:

[0094] ① Define the area covered by each target building block as a closed area.

[0095] Reference Figure 5 The target blocks CDEF in (1), ABCD and EFGH in (2), ABCD in (3), and ABCD and GHIJ in (4) can all be divided into a closed area based on the boundary lines of the target blocks.

[0096] ② If there is a target road along the top edge of the target building block in a direction away from the data collection road:

[0097] The target road and the top and left and right sides of the target building block, extended along the direction away from the data collection road, form a closed area.

[0098] Figure 5 (1) If the top edge of the target building block CDEF is along the direction away from the collection road, then the target road and the extension lines of the top edge and left and right edges of the target building block CDEF along the direction away from the collection road form a closed area ABCD.

[0099] ③ If there is another target building along the direction away from the data collection road on the upper boundary of the target building:

[0100] The extensions of the top and left and right sides of the target building block and the bottom line of another target building block form a closed area.

[0101] Figure 5 (2) If there is another target building block ABCD along the direction away from the collection road, the upper edge of the target building block EFGH and the extension of its left and right edges and the lower edge of the other target building block ABCD form a closed area.

[0102] When two target building blocks are adjacent to each other along a direction perpendicular to the data collection road, a closed area is formed by the adjacent boundary lines and their extensions, as well as the extensions of the left and right boundary lines. Figure 5 The closed region EFGC in (5).

[0103] Optionally, when two target blocks are adjacent along a direction perpendicular to the data collection road, and the distance between adjacent edges is less than a preset distance, the area between adjacent edges can be ignored, or the area between adjacent edges can be evenly divided into the adjacent closed areas on both sides.

[0104] ④ If there is a target road between the bottom edge of the target building and the data collection road:

[0105] The target road and the bottom and left and right sides of the target building block, along the extension of the lines closest to the road where the data was collected, form a closed area.

[0106] Figure 5 (3) If there is a target road between the bottom edge of the target building block ABCD and the data collection road, then the target road and the extension of the bottom edge and left and right edge of the target building block ABCD along the direction close to the data collection road form a closed area CDEF.

[0107] ⑤ There is no target road between the bottom edge of the target building and the data collection road:

[0108] The bottom edge and left and right edge of the target building block, along with the extension of the road leading to the data collection point and the data collection road, form a closed area.

[0109] Figure 5 (1) If there is no target road between the lower edge of the target building block CDEF and the data collection road, then the lower edge of the target building block CDEF, the extensions of its left and right edges along the direction closest to the data collection road, and the data collection road together form a closed area EFGH. Also... Figure 5 The closed region GHIJ in (2), the closed region CDEF in (4), and the closed region IJKL all belong to case ⑤.

[0110] ⑥ There is another adjacent target building within a preset distance along the direction of the data collection road:

[0111] Based on the positional relationship between the edges of these two target building blocks, the enclosed area is obtained.

[0112] It is possible that when the distance between the adjacent edges of two adjacent target blocks is greater than a set threshold, the two adjacent edges and the two edges perpendicular to one of the adjacent edges extend along the line towards the other target block to form a closed area.

[0113] Figure 5 (4) If there is another target building GHIJ adjacent to the target building ABCD along the direction of the collection road within a preset distance, then the side lines BC, GJ and the side lines AB, CD perpendicular to the side line BC along the extension line towards the other target building GHIJ form a closed area BCJG.

[0114] When two target building blocks are adjacent to each other along the data collection road, a closed area is formed by the adjacent boundary lines and their extensions, as well as the extensions of the upper and lower boundary lines. Figure 6 The closed region MGNC in (6).

[0115] Optionally, when two target blocks are adjacent along the direction of the data collection road, and the distance between the adjacent edges is less than the preset distance, the area between the adjacent edges can be ignored, or the area between the adjacent edges can be evenly divided into the adjacent closed areas on both sides.

[0116] Step S42: Determine whether a building can be found within the closed area.

[0117] If yes, proceed to step S44; otherwise, proceed to step S43.

[0118] Step S43: Obtain satellite imagery of the enclosed area and use a pre-trained building classification model to identify whether there are buildings in the satellite imagery.

[0119] Specifically, a pre-trained building classification model is used to determine whether there are buildings in satellite imagery. First, the result of whether there are buildings in the satellite imagery and the accuracy of the result are obtained. For example, if the result is "yes, 80%", it means that there are buildings in the satellite imagery with an accuracy of 80%. Then, it is determined whether the obtained accuracy is not less than a preset accuracy threshold. If yes, the result of whether there are buildings is determined to be reliable. If no, the result of whether there are buildings is determined to be unreliable, that is, the building classification model cannot determine whether there are buildings in the satellite imagery.

[0120] If the building classification model identifies buildings in the satellite image, proceed to step S46; if the building classification model identifies no buildings in the satellite image, proceed to step S47; if the building classification model cannot identify whether there are buildings in the satellite image, proceed to step S48.

[0121] Step S44: Determine the closed region as the target closed region where points of interest can be placed.

[0122] Step S45: Generate a location reference line for the road based on the target closed area.

[0123] Step S46: Determine whether the area of ​​the closed region is greater than the preset area threshold.

[0124] If yes, proceed to step S44; otherwise, proceed to step S47.

[0125] Step S47: Determine that the closed area is a non-target closed area.

[0126] This means determining which points of interest cannot be placed within a closed area.

[0127] Step S48: Identify whether the closed area is a closed area enclosed by the up and down roads.

[0128] A closed area enclosed by roads in both directions refers to a closed area where two parallel roads within the closed element are in an upward-downward relationship, meaning the two roads have the same or similar names, are close to each other, and extend in nearly parallel directions. The specific rules for determining the upward-downward roads can also be other rules, which are not limited in this embodiment.

[0129] Reference Figure 7 As shown, when the road being collected is Road 1, the closed areas defined on the collection side include Closed Area 1 and Closed Area 2, etc.; however, when the road being collected is Road 2, the closed areas defined on the collection side also include the area between Road 1 and Road 2.

[0130] If step S48 is determined to be yes, proceed to step S49; if step S48 is determined to be no, proceed to step S47.

[0131] Step S49: Obtain the attribute parameters of the closed area, and use the pre-trained decision tree to determine whether there are buildings within the closed area.

[0132] The decision tree mentioned above can be an iterative decision tree algorithm (Gradient Boosting Decision Tree) or other decision trees.

[0133] Reference Figure 8 As shown, decision trees can be pre-trained in the following way:

[0134] Step S71: Label the closed area samples divided based on map data to indicate whether there are buildings.

[0135] Step S72: From the marked closed areas, obtain a set proportion of closed area samples with buildings marked and closed area samples without buildings marked.

[0136] The ratio of the selected samples labeled with buildings to the samples labeled without buildings can be 1:1.

[0137] Step S73: Obtain the attribute parameters of each selected sample.

[0138] Specifically, the attribute parameters may include:

[0139] 1. The area of ​​the enclosed zone and whether there are any buildings within the enclosed zone.

[0140] 2. Attributes of the up and down roads in the closed area: such as road status, road rendering level, road extension direction, road length and width, construction status, cycling level and road level. The road status can include: normal traffic 0 (default), under construction 1, no traffic 2, and pre-opening 3; the rendering level can include: not surveyed 0 (default): main road 1 and secondary road 2; the extension direction (traffic direction) can include: two-way traffic 1, forward traffic 2, reverse traffic 3, and two-way no traffic 4; the construction status can include: not under construction 1, (two-way) under construction 2, forward construction 3, and negative construction 4; the cycling level can include: not surveyed 0 (default), dedicated 1, recommended 2, ordinary 3, not recommended 4, recommended prohibition 5, and mandatory prohibition 6; the road level can include: expressway 41000, national highway 42000, urban expressway 43000, urban main road 44000, urban secondary road 45000, urban ordinary road 47000, provincial highway 51000, county road 52000, township road 53000, county and village internal road 54000, and small road 49000.

[0141] 3. Are the same attributes of the up and down roads in the closed area consistent? For example, are the rendering levels of the up and down roads consistent, are the construction status of the up and down roads consistent, are the riding levels of the up and down roads consistent, and are the road levels of the up and down roads consistent?

[0142] Step S74: Train a decision tree using the attribute parameters of the obtained closed region samples.

[0143] If step S49 is determined to be yes, proceed to step S46; if step S49 is determined to be no, proceed to step S47.

[0144] Based on the inventive concept of this invention, embodiments of this invention also provide a method for determining the location of a point of interest, comprising:

[0145] A position reference line is generated using the above-described method for generating position reference lines; the projection point of the point of interest on the nearest position reference line is determined, and the projection point is used as the position point of the point of interest.

[0146] Based on the inventive concept of this invention, embodiments of this invention also provide a device for generating a position reference line, the structure of which is as follows: Figure 9 As shown, it includes:

[0147] The first acquisition module 81 is used to acquire a closed area located on a preset acquisition side of the acquisition road based on map data;

[0148] The second acquisition module 82 is used to acquire a closed region where points of interest can be located as the target closed region;

[0149] The generation module 83 is used to generate the location reference line of the collected road based on the target closed area obtained by the second acquisition module 82.

[0150] In some embodiments, the first acquisition module 81 is specifically used for:

[0151] From the map data, a road located on a preset collection side of the collection road and whose distance to the collection road does not exceed a preset distance threshold is selected as the target road; from the map data, a building located on the preset collection side of the collection road and whose lower edge of the building is located on the preset collection side and whose distance to the collection road does not exceed a preset distance threshold is selected as the target building, wherein the lower edge of the building is the edge line closest to the collection road among the edges of the collection road; based on the target road and the target building, the geographical area located on the preset collection side of the collection road is divided into one or more closed areas.

[0152] In some embodiments, the first acquisition module 81 is specifically used for:

[0153] The area covered by each target building is defined as a closed area. If there is a target road along the top edge of the target building away from the collection road, then the target road, the extension of the top edge and left and right edges of the target building along the direction away from the collection road, constitute a closed area. If there is another target building along the top edge of the target building away from the collection road, then the extension of the top edge and left and right edges of the target building and the bottom edge of the other target building constitute a closed area. If there is a target road between the bottom edge of the target building and the collection road, then the target road, the extension of the bottom edge and left and right edges of the target building along the direction close to the collection road, constitute a closed area. If there is no target road between the bottom edge of the target building and the collection road, then the extension of the bottom edge and left and right edges of the target building along the direction close to the collection road, and the collection road, constitute a closed area. If there is another adjacent target building within a preset distance along the direction of the collection road, then the closed area is obtained based on the positional relationship of the edges of these two target buildings.

[0154] In some embodiments, the second acquisition module 82 is specifically used for:

[0155] When a building is found within the closed area, the closed area is determined to be a closed area where a point of interest can be placed; when no building is found within the closed area, satellite imagery of the closed area is acquired, and a pre-trained building classification model is used to identify buildings within the satellite imagery, thus determining the closed area to be a closed area where a point of interest can be placed.

[0156] In some embodiments, the second acquisition module 82 is further configured to:

[0157] If the building classification model cannot identify whether there are buildings in the satellite image, it is determined whether the closed area is a closed area enclosed by the up and down roads. If so, the attribute parameters of the closed area are obtained, and a pre-trained decision tree is used to determine whether there are buildings in the closed area.

[0158] In some embodiments, the generation module 83 is specifically used for:

[0159] At preset intervals, obtain perpendicular lines to the collection road within the target enclosed area, and obtain reference points for each perpendicular line in the direction close to the collection road; use the line connecting the reference points as a position reference line for the collection road.

[0160] In some embodiments, the above-described apparatus further includes a deletion module 84, which is used to delete closed areas with an area smaller than a preset area threshold after the first acquisition module 81 acquires a closed area located on a preset acquisition side of the acquisition road.

[0161] Based on the inventive concept of this invention, embodiments of this invention also provide a device for determining the location of a point of interest, the structure of which is as follows: ​ As shown, it includes:

[0162] Generation module 91 is used to generate a position reference line using the above-described method for generating position reference lines;

[0163] The projection module 92 is used to determine the projection point of the point of interest on the nearest location reference line, and to use the projection point as the location point of the point of interest.

[0164] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0165] Based on the same inventive concept, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the above-described method for generating a position reference line or the above-described method for determining the position of a point of interest.

[0166] Based on the same inventive concept, embodiments of the present invention also provide a server, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the above-mentioned method for generating position reference lines or the above-mentioned method for determining the position of points of interest.

[0167] Unless otherwise specifically stated, terms such as processing, calculation, operation, determination, display, etc., may refer to the actions and / or processes of one or more processing or computing systems or similar devices that represent the manipulation and conversion of data representing physical (e.g., electronic) quantities within the registers or memory of the processing system into other data similarly representing physical quantities within the memory, registers, or other such information storage, transmission, or display devices of the processing system. Information and signals can be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the above description can be represented by voltage, current, electromagnetic waves, magnetic fields or particles, light fields or particles, or any combination thereof.

[0168] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to the specific order or hierarchy described.

[0169] In the detailed description above, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features in a single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, with each claim representing a separate preferred embodiment of the invention.

[0170] Those skilled in the art will also understand that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments herein can be implemented as electronic hardware, computer software, or a combination thereof. To clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the described functionality in alternative ways for each specific application; however, such implementation decisions should not be construed as departing from the scope of this disclosure.

[0171] The steps of the methods or algorithms described in conjunction with the embodiments herein can be directly embodied in hardware, software modules executed by a processor, or a combination thereof. The software modules can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in a user terminal. Alternatively, the processor and storage medium can exist as discrete components in the user terminal.

[0172] For software implementation, the techniques described in this application can be implemented using modules (e.g., procedures, functions, etc.) that perform the functions described in this application. This software code can be stored in memory units and executed by a processor. The memory units can be implemented within the processor or outside the processor; in the latter case, they are communicatively coupled to the processor via various means, as is well known in the art.

[0173] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

Claims

1. A method of generating a position reference line, characterized by, The method comprises the following steps: Based on the map data, the closed area located on the preset collection side of the collection road is obtained; The closed area where the interest point can fall is obtained as the target closed area; Based on the target closed area, the position reference line of the collection road is generated, wherein the position reference line is used to determine the position of the interest point.

2. The method of claim 1, wherein, Based on the map data, the closed area located on the preset collection side of the collection road is obtained, comprising: From the map data, the road located on the preset collection side of the collection road and having a distance from the collection road not exceeding a preset distance threshold is obtained as a target road; From the map data, the building located on the preset collection side of the collection road and having a distance from the lower edge line of the building to the collection road not exceeding a preset distance threshold is obtained as a target building, wherein the lower edge line of the building is the edge line closest to the collection road among the edge lines of the building; Based on the target road and the target building, the geographical area located on the preset collection side of the collection road is divided into one or more closed areas.

3. The method of claim 2, wherein, Based on the target road and the target building, the geographical area located on the preset collection side of the collection road is divided into one or more closed areas, comprising: The area covered by each target building is determined as a closed area; If the upper edge line of the target building has a target road in the direction away from the collection road, the target road and the extension of the upper edge line and the left and right edge lines of the target building in the direction away from the collection road form a closed area; If the upper edge line of the target building has another target building in the direction away from the collection road, the extension of the upper edge line and the left and right edge lines of the target building and the lower edge line of the other target building form a closed area; If there is a target road between the lower edge line of the target building and the collection road, the target road and the extension of the lower edge line and the left and right edge lines of the target building in the direction close to the collection road form a closed area; If there is no target road between the lower edge line of the target building and the collection road, the extension of the lower edge line and the left and right edge lines of the target building in the direction close to the collection road and the collection road form a closed area; If the target building has another target building adjacent to it within a preset distance in the direction of the collection road, the closed area is obtained according to the positional relationship of the edge lines of the two target buildings.

4. The method of claim 1, wherein, The target closed area where the interest point can fall is obtained, comprising: When a building is searched in the closed area, the closed area is determined as the closed area where the interest point can fall; When no building is searched in the closed area, a satellite image of the closed area is obtained, and when it is identified by a pre-trained building classification model that there is a building in the satellite image, the closed area is determined as the closed area where the interest point can fall.

5. The method of claim 4, wherein, The building classification model is pre-trained in the following manner: The satellite image samples are labeled with or without buildings; From the labeled satellite image samples, a satellite image sample labeled with a building and a satellite image sample labeled without a building are obtained in a set proportion; The building classification model is trained by using the obtained satellite image samples.

6. The method of claim 4, wherein, Further comprising: If the building classification model cannot identify whether there is a building in the satellite image, it is identified whether the closed area is a closed area surrounded by an uplink and downlink road. If so, the attribute parameters of the closed area are obtained, and a pre-trained decision tree is used to determine whether there is a building in the closed area.

7. The method of claim 6, wherein, The decision tree is pre-trained by: annotating the closed area samples based on the map data division with or without buildings; from the annotated closed area, obtaining a set proportion of closed area samples labeled with buildings and closed area samples labeled without buildings; the attribute parameters of the obtained closed area samples are used to train the decision tree.

8. The method according to any one of claims 1 to 7, characterized in that Based on the target closed area, the position reference line of the collection road is generated, including: According to the preset interval, the perpendicular line perpendicular to the collection road is obtained in the target closed area, and the reference point of the perpendicular line is obtained in the direction close to the collection road on each perpendicular line; the line connecting the reference points is taken as a position reference line of the collection road.

9. The method according to any one of claims 1 to 7, characterized in that After the closed area located on the preset collection side of the collection road is obtained, the method further comprises: deleting the closed area with an area less than the preset area threshold.

10. A method of determining a location of a point of interest, characterized by, Comprising: generating a position reference line by using the position reference line generation method of any one of claims 1-9; determining the projection point of the interest point on the nearest position reference line, and taking the projection point as the position point of the interest point.

11. An apparatus for generating a position reference line, characterized by Comprising: a first obtaining module for obtaining a closed area located on a preset collection side of a collection road based on map data; a second obtaining module for obtaining a closed area that can fall an interest point as a target closed area; a generating module for generating a position reference line of the collection road based on the target closed area obtained by the second obtaining module, wherein the position reference line is used to determine the position of the interest point.

12. A point of interest location determining apparatus characterized by comprising: Comprising: a generating module for generating a position reference line by using the position reference line generation method of any one of claims 1-9; a dropping module for determining the projection point of the interest point on the nearest position reference line, and taking the projection point as the position point of the interest point.

13. A computer readable storage medium having stored thereon computer instructions, wherein, When the instruction is executed by the processor, the position reference line generation method of any one of claims 1-9 or the interest point position determination method of claim 10 is implemented.

Images