A data processing system based on map image erosion and dilation

Abstract

The application provides a data processing system based on map image erosion and expansion, which comprises a database, a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are implemented: a target geographical image is preprocessed to obtain a first intermediate geographical image; the first intermediate geographical image is subjected to expansion processing using a first convolution kernel to obtain a second intermediate geographical image; the second intermediate geographical image is subjected to erosion processing using a second convolution kernel to obtain a final map image; and after the erosion and expansion, the AOI boundary can be more accurately identified when the AOI boundary is identified.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a data processing system based on map image erosion and expansion. BACKGROUND

[0002] AOI is the abbreviation of Area of Interest, which can be called interest surface, and is surface data. AOI is a higher level of abstraction of POI, which is composed of polygon fence boundary and feature data. POI is the abbreviation of Point of interest, which is translated into Chinese as interest point, and is point data. Any position on the map can represent an interest point, which is composed of longitude and latitude and other feature data. POI is also quite common in the field of map application, such as: surrounding search, real-time position acquisition, etc. However, POI (Point of Interest) data is a powerful tool for identifying urban areas. In the past, point information was commonly used to view the degree of aggregation of various urban functions, but the area, volume, nature and other properties of the functional area were ignored, making it difficult to reflect the reality. In the prior art, AOI is often obtained by downloading a map and then manually drawing a map. However, manual drawing is labor-intensive and not simple. SUMMARY

[0003] In view of the above technical problems, the technical scheme adopted by the present application is: a data processing system based on map image erosion and expansion, the data processing system based on map image erosion and expansion includes a database, a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are implemented:

[0004] S201, pre-processing a target geographic image to obtain a first intermediate geographic image;

[0005] S203, performing expansion processing on the first intermediate geographic image using a first convolution kernel to obtain a second intermediate geographic image;

[0006] The first convolution kernel meets the following conditions: S 11 is positively correlated with X*Y, S 11 is the number of "1"s in the first convolution kernel, and X*Y is the area of the target map image;

[0007] S205, performing erosion processing on the second intermediate map image using a second convolution kernel to obtain a final map image;

[0008] The second convolution kernel meets the following conditions: S 21 is positively correlated with X*Y and S 11 > S 21 S21 is the number of "1"s in the second convolution kernel.

[0009] The present application has at least the following technical effects: the range of the first intermediate map image inflation is greater than the range of the second intermediate map image erosion, that is, after the inflation and erosion operations, the originally white area will increase to some extent, which can expand the white area in the middle of the adjacent to-be-processed areas in the first intermediate map image, or for some special cases, such as the interval with white and black in the middle of the to-be-processed area in the first intermediate map image, after the erosion and inflation, the white area can be connected, and the AOI boundary can be more accurately identified when the AOI boundary is identified. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0011] Figure 1 The flowchart of the data processing system for obtaining the target geographic image provided by the embodiment of the present application executes the computer program.

[0012] Figure 2 The flowchart of the data processing system based on map image erosion and inflation provided by the embodiment of the present application executes the computer program.

[0013] Figure 3 The flowchart of the data processing system for obtaining the AOI boundary provided by the embodiment of the present application executes the computer program. DETAILED DESCRIPTION

[0014] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0015] It is to be understood that the terminology "first", "second" and the like used in the specification and the claims of the application as well as the foregoing drawings is merely intended to distinguish between similar objects and not necessarily to describe a particular sequential or chronological order. It is to be understood that the use of data herein so described makes it possible, in appropriate cases, to adapt the application to varied embodiments without necessarily departing from the scope of the application. Furthermore, the terms "comprise" and "include", and any variation thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, article, or apparatus that comprises a list of steps or units can not necessarily be limited to those steps or units which are expressly listed, but can include additional steps or units not expressly listed or inherent to such process, method, article, or apparatus.

[0016] Embodiment one

[0017] The embodiment of the application provides a data processing system for acquiring a target geographical image, the data processing system for acquiring the target geographical image comprises a database, a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are realized, such as Figure 1

[0018] S101, acquiring a geographical image set P corresponding to a target POI, P = {P1,..., P i ,..., P m}, P i = {P i1 ,..., P ij ,..., P in}, P ij = (W ij , L ij ), W ij is the width of the jth geographical image in the ith type geographical image set corresponding to the target POI, and L ij is the length of the jth geographical image in the ith type geographical image set corresponding to the target POI.

[0019] Specifically, before S101, the following steps are further included:

[0020] S1, acquiring parameter information of a to-be-processed region.

[0021] Specifically, the parameter information comprises the size and shape of the to-be-processed region, the total number of divided images, the length X of the to-be-processed region and the width Y of the to-be-processed region.

[0022] Preferably, the to-be-processed region is a rectangle; more preferably, the to-be-processed region is a square.

[0023] ​S2, according to the parameter information of the region to be processed, dividing the region to be processed in a preset manner to obtain a map type list P corresponding to the target POI, P = {P1, …, P i , …, P m}.

[0024] In one embodiment, the image type list P = {P1, …, P i , …, P m} is obtained in the following preset manner: the region to be processed is divided into each type of image, i.e., each type of image is one image, and adjacent images share the same boundary.

[0025] wherein W ij meets the following condition:

[0026] W ij = W (i-1)j + W α , wherein W α is a width growth parameter factor;

[0027] wherein L ij meets the following condition:

[0028] L ij = L (i-1)j + L α , wherein L α is a length growth parameter factor;

[0029] Further, W α meets the following condition:

[0030] W ɑ = K1*W 1j ; K1 is a first size adjustment factor.

[0031] L α meets the following condition: L ɑ = K1*L 1j .

[0032] It can be understood that the length and width of P i are adjusted according to the first size adjustment factor of the length and width of P1, gradually expanding the length and width of the map image.

[0033] In another embodiment of the present application, W α meets the following condition:

[0034] W ɑ = K2*W (i-1)j ; K2 is a second size adjustment factor.

[0035] L α meets the following condition: Lɑ =K2*L (i-1)j .

[0036] Technical effect: It can be understood that P i The length and width of P i-1 The length and width of the map image are adjusted by the second size adjustment factor, and the length and width of the map image are gradually expanded. Compared with adjusting directly according to P1, according to P i-1 More flexibility in making adjustments.

[0037] Among them, K2 satisfies the following condition: K2≤1.

[0038] The present invention further includes the following steps after S101:

[0039] S1011, obtaining the jth geographic image P in the mth type map image set mj =(W mj , L mj );

[0040] S1013, when W mj ≥X and L mj ≥Y, execute S103, wherein X is the width of the target map image corresponding to the target POI, and Y is the length of the target map image corresponding to the target POI.

[0041] Based on this, it can be understood that when W mj ≥X and L mj When ≥Y, the mth type map image includes the entire map image corresponding to the target POI, ensuring that the largest type image includes the entire target geographic image.

[0042] S103 : Input P into a preset geographic image model to obtain a target geographic image corresponding to the target POI, wherein the target geographic image is a geographic image including the target POI and a boundary area corresponding to the target POI.

[0043] Specifically, the geographic image set P corresponding to the target POI is input into the map image model = {P1, ..., P i ,…,P m} and a scaling index, where the scaling index is a fixed value and can be determined according to actual conditions.

[0044] Optionally, when the map image model outputs the target geographic image corresponding to the target POI, different colors can be selected for different areas according to actual needs.

[0045] Based on S101-S103, through iteration of the step length, the overlapped map images can be acquired, the acquired AOI boundary can be completely presented on one map image, the length and width of different types of images are acquired according to the preset step length on multiple map images, the overlapped part is formed between each type of image, the AOI is avoided to be split in several images, the incomplete AOI boundary in the image is avoided, therefore, the step length is adopted to acquire the overlapped image, so that the complete AOI boundary can be acquired in one image, and the situation that the complete AOI boundary needs to be acquired by splicing after the partial AOI boundary is acquired on multiple map images is avoided.

[0046] Embodiment two

[0047] The application also comprises a data processing system based on map image erosion and expansion, when the computer program is executed by the processor, the following steps are realized, as shown in Figure 2 .

[0048] S201, pre-processing the target geographical image to acquire a first intermediate geographical image.

[0049] Specifically, the pre-processing of the target geographical image comprises the following steps:

[0050] S2011, performing image binarization on the target geographical image to acquire a preliminary geographical image.

[0051] Specifically, performing image binarization on the target geographical image means setting the pixel value of the pixel point of the target geographical image as 0 or 255.

[0052] In an embodiment of the application, the image binarization is realized through the following steps:

[0053] S2011, acquiring a pixel value list of all pixel points of the target geographical image ij ={S ij1 ,……,S ijx1 ,……,S ijx},S ijx1 is the pixel value of the x1th pixel point of the target geographical image, the value range of x1 is 1 to x, x is the total number of pixel points in the target geographical image; when S ijx1 ≥T, S′ ijx1 =255; when S ijx1 <T, S′ ijx1 =0; wherein, T is a preset gray threshold value, and T can be input according to the actual situation, S′ ijx1 is the pixel value of the x1th pixel point of the first intermediate geographical image.

[0054] Based on this, the binaryzation of the target geographical image can reduce the data amount and the calculation amount, so that the data processing system occupies less memory and runs more smoothly.

[0055] In another embodiment of the application, the image binaryzation is realized by the following steps:

[0056] Obtain a list of pixel values S of all pixel points of the target geographical image ij = {S ij1 , …, S ijx1 , …, S ijx}, S ijx1 is the pixel value of the x1th pixel point of the target geographical image, the value range of x1 is 1 to x, and x is the total number of pixel points in P ij ; for S ijx1 , take a circular region with a radius of r around S ijx1 , calculate the area mean of the circular region and mark it as T, which satisfies the following conditions:

[0057] T = [∑ r s＝-r ∑ r t＝-r f(x+s, y+t)] / (2r+1) 2 ;

[0058] When S ijx1 ≥ T, S' ijx1 = 255; when S ijx1 < T, S' ijx1 = 0; S' ijx1 is the pixel value of the x1th pixel point of the first intermediate geographical image.

[0059] Based on this, using adaptive threshold values can use different threshold values according to the local features of the map image, avoiding the influence of shadows or high brightness in part of the image on the preset gray threshold value of the entire image, thereby avoiding the wrong judgment of image binaryzation and improving the accuracy of image binaryzation.

[0060] S2013, delete the boundary block in the prepared geographical image to obtain the first intermediate geographical image, the boundary block being a region in the prepared geographical image that does not contain the boundary region corresponding to the target POI.

[0061] Specifically, the boundary block is obtained by the following steps:

[0062] S10, obtain the RGB value B1 of the boundary point B.

[0063] S30, obtain the RGB values {B2, B3, B4} of the 3 neighborhoods of the boundary point B.

[0064] S50, when B1 = Bt When B t is added to the boundary block, t = 2 or t = 3 or t = 4.

[0065] S70, traversing B t 3 neighborhoods of the corresponding boundary point.

[0066] S90, when the RGB values are all different, ending the traversal.

[0067] Based on S10-S90, the boundary block is obtained according to the boundary point and the neighborhood of the boundary point, and the boundary block is deleted, which can be understood as obtaining the AOI boundary on the map image containing the complete AOI boundary. The boundary block of the map image does not contain the complete boundary of the AOI, so the boundary block is deleted to reduce the calculation amount.

[0068] S203, performing inflation processing on the first intermediate geographic image using a first convolution kernel to obtain a second intermediate image;

[0069] Specifically, those skilled in the art know that the first convolution kernel is composed of numbers "0" and "1", and the first convolution kernel has a plurality of structures with a center point, wherein the first convolution kernel of the present application can be input according to actual conditions.

[0070] Preferably, the first convolution kernel satisfies the following conditions: S 11 positively correlated with X*Y, S 11 X*Y is the area of the target map image;

[0071] In an embodiment of the present application, the first convolution kernel satisfies the following conditions:

[0072] S 11 R1*X*Y, R1 is a first area correlation coefficient.

[0073] Specifically, the center point of the first convolution kernel traverses each pixel point of the first intermediate map image. Among the traversed pixel points of the first intermediate map image, the position corresponding to "1" of the first convolution kernel corresponds to "255", so that the position corresponding to the center point of the first convolution kernel becomes 255, thereby generating a second intermediate map image.

[0074] S205, performing erosion processing on the second intermediate map image using a second convolution kernel to obtain a final map image;

[0075] Wherein, the second convolution kernel satisfies the following conditions: S 21 positively correlated with X*Y and S 11 >S 21 S21 is the number of "1" in the second convolution kernel.

[0076] Specifically, the person skilled in the art knows that the bit number of the first convolution kernel and the second convolution kernel is the same.

[0077] In an embodiment of the present application, the second convolution kernel satisfies the following condition:

[0078] S 21 R1*X*Y, R1 is the first area correlation coefficient

[0079] Optionally, R1≥R2; preferably, R1>R2.

[0080] Specifically, the center point of the second convolution kernel traverses each pixel point of the second intermediate map image, and among the traversed pixel points of the second intermediate map image, if the position of the second convolution kernel with "1" corresponds to "0", the position corresponding to the center point of the second convolution kernel becomes "0", thereby generating the final map image.

[0081] Specifically, after S205, it further includes:

[0082] S2051, obtaining the number A1 of dilated pixel points in the second intermediate map image;

[0083] S2053, obtaining the number A2 of eroded pixel points in the final map image;

[0084] S2055, A1 and A2 satisfy the following condition:

[0085] A1-A2=A, A is a preset dilated width value.

[0086] Based on this, the first intermediate map image is dilated and eroded using the first convolution kernel and the second convolution kernel, and the number of "1"s in the first convolution kernel is greater than the number of "1"s in the second convolution kernel, which can be understood as the dilated range of the first intermediate map image being greater than the eroded range of the second intermediate map image, i.e. after the dilated and eroded operation, the originally white region will increase to some extent, which can make the white region in the middle of the adjacent to-be-processed regions in the first intermediate map image expand, or for some special cases, for example, the first intermediate map image to-be-processed region has white and black intervals, after the erosion and dilatation, the white region can be connected, and the AOI boundary can be more accurately identified when the AOI boundary is identified.

[0087] Embodiment three

[0088] On the basis of embodiment two, when the computer program is executed by the processor, the following steps are implemented on the final map image list A", as shown in Figure 3 .

[0089] S301, traversing the pixel points of the final map image A' ′ij to obtain the first boundary point;

[0090] wherein the final map image A" ij Each pixel point has an access identifier, which is used to determine whether the corresponding pixel point is recognized, and further, the access identifier is "1" or "0"; it can be understood that the skilled in the art knows that when A' ij The corresponding pixel point is recognized, the access identifier is "1" or "0"; otherwise, when A" ′ij The corresponding pixel point is not recognized, the access identifier is "0" or "1". ij The corresponding pixel point is not recognized, the access identifier is "0" or "1".

[0091] Preferably, when A" ij The corresponding pixel point is recognized, the access identifier is "1"; otherwise, when A' ij The corresponding pixel point is not recognized, the access identifier is "0".

[0092] Specifically, obtaining the first boundary point includes the following steps:

[0093] S3011, obtaining all pixel points U ij of the final map image A" ij ={U ij1 , …, U ijd , …, U ije}, U ijd is the dth pixel point of A" ij , the value range of d is 1 to e, and e is the total number of pixel points of A" ij ;

[0094] S3013, obtaining the pixel value of the pixel point U ijd ;

[0095] S3015, when the pixel value G(U ijd )=255, the access identifier of the point is marked as "1", and the next point is traversed;

[0096] S3017, when the pixel value G(U ijd )=0, the access identifier of the point is marked as "1" and the pixel values of the eight neighborhoods of U ijd are obtained;

[0097] S3019, when the pixel value of any one of the eight neighborhoods of U ijd is equal to 255, U ijd is determined as the first boundary point;

[0098] Based on S3011-S3019, the pixel points of the final map image are acquired, whether the pixel value is the first boundary point is judged until the first boundary point is found, and the access identifier of the pixel point which has been judged whether it is the first boundary point is marked as "1", so that when the access is performed again, the access identifier can be directly skipped, the access is not performed, time is saved, and efficiency is improved.

[0099] S303, the four neighborhoods of the first boundary point are traversed, and whether the four neighborhoods of the first boundary point are boundary points is judged.

[0100] Specifically, whether it is a boundary point is judged according to S805-S807.

[0101] S305, the four neighborhoods of the e-th boundary point are continuously traversed until the four neighborhoods of the boundary point are boundary points.

[0102] S307, second layer traversal is performed until the access identifier of all pixel points is "1".

[0103] Specifically, the second layer traversal refers to the traversal after the first boundary point according to the first layer traversal, and the first layer traversal refers to the traversal of the first boundary point to obtain a complete boundary.

[0104] Based on S301-S307, the final map image is traversed to obtain the first boundary point, the next boundary point is found through the four neighborhoods of the first boundary point, until a complete closed boundary is found, the method of finding a boundary through a boundary is faster and more convenient than directly traversing all pixel points; meanwhile, the access identifier of the accessed pixel point is marked as "1", so that the traversal speed is faster in the second layer traversal, and the second traversal can find that the AOI boundary contains another AOI boundary, so that the case that the AOI boundary contains another AOI boundary is not found, thereby more completely obtaining the boundaries of all AOIs in the final map image.

[0105] In another embodiment of the application, the boundary point can also be obtained through the following steps:

[0106] S3031, the pixel c*h of the final map image A" ij is acquired; wherein c refers to that the width of the final map image is c pixels, and h refers to that the length of the final map image is h pixels.

[0107] S3033, R j i , Q j i , R i j , Q j i , Q

[0108] R j i = F j i+1 - F j i ; Q j i = F j i - F j i+1 ; R i j = K i j+1 - K j i ; Q j i = K j i - K j i+1 ; wherein F j i+1 represents a pixel value of an (i+1)th row jth pixel point in a final map image, F j i represents a jth pixel value of an ith row in the final map image, K j i+1 represents a pixel value of an (i+1)th column jth pixel point in the final map image, K j i represents a pixel value of an ith column jth pixel point in the final map image.

[0109] S3035, when R j i ≠ 0, R j i is stored in a target set; when Q j i ≠ 0, Q j i is stored in the target set; when R i j ≠ 0, R i j is stored in the target set; when Q j i ≠ 0, Q j i is stored in the target set; the target set is used to store boundary points.

[0110] S309, connecting all boundary points according to corresponding longitude and latitude to obtain a target AOI boundary;

[0111] Specifically, the number of samples is determined based on the target AOI boundary size, the target boundary points are obtained by sampling the boundary points, and the target AOI boundary is obtained by connecting the corresponding latitude and longitude.

[0112] Among them, those skilled in the art know that any method of connecting the latitude and longitude of a known point to obtain a closed boundary belongs to the protection scope of the present application, and will not be repeated here.

[0113] S311, the target POI is associated with the target AOI boundary.

[0114] Specifically, those skilled in the art know that any method of associating the target POI and the target AOI boundary in the prior art belongs to the protection scope of the present application, for example, the method of using an association table to associate the target POI and the target AOI boundary.

[0115] Based on the present application, first, the map image and POI of the to-be-processed area are obtained through the third-party server, in the obtaining process, the map image is obtained in a step iteration manner, so that the map image has overlapping parts, so that the complete target AOI boundary can be obtained in one map image; the first intermediate map image obtained is subjected to an expansion and corrosion operation, so that the boundary which is not obvious or disconnected can be more obvious, which is convenient for obtaining the target AOI boundary in the subsequent process; the first boundary point is found, then the complete closed target AOI boundary is obtained according to the method of finding the boundary point from the boundary point, and the target POI and the target AOI boundary are associated, so that the target POI can be associated with the boundary of the target AOI, and the area of the associated AOI is more accurately obtained.

[0116] Although some specific embodiments of the present application have been described in detail through examples, those skilled in the art should understand that the above examples are only for illustration, and are not intended to limit the scope of the present application. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the present application. The scope of the present application is defined by the appended claims.

Claims

1. A data processing system based on map image erosion dilation, characterized by, The data processing based on the map image erosion and expansion includes a database, a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are implemented: S201, pre-processing the target geographic image to obtain a first intermediate geographic image; S203, using a first convolution kernel to perform expansion processing on the first intermediate geographic image to obtain a second intermediate geographic image; Wherein, the first convolution kernel meets the following conditions: S 11 is positively correlated with X Y, S 11 is the number of "1" in the first convolution kernel, X Y is the area of the target map image; S205, using a second convolution kernel to perform erosion processing on the second intermediate geographic image to obtain a final geographic image; Wherein, the second convolution kernel meets the following conditions: S 21 X Y is positively correlated with S 11 >S 21 S21 is the number of "1"s in the second convolution kernel. S 11 = R1 X Y, R1 is a first area correlation coefficient; S 21 = R2 X Y, R2 is a second area correlation coefficient; R1 > R2; In S201, the target geographic image is obtained by the following method: S101, a target POI corresponding geographical image set P = {P1,..., P i} m} P i = (P i1 ,..., P ij ,..., P in} P ij = (W ij , L ij ), W ij is the target POI corresponding to the first type of geographical image set in the jth geographical image, L ij is the target POI corresponding to the first type of geographical image set in the jth geographical image. wherein W ij meets the following conditions: W ij =W （i-1）j+Wα where W α is a width growth parameter factor; wherein L ij meets the following conditions: L ij =L （i-1）j+Lα where L α is a length growth parameter factor; S103, inputting P into a preset geographic image model to obtain a target geographic image corresponding to the target POI, wherein the target geographic image is a geographic image including the target POI and a boundary region corresponding to the target POI.

2. The data processing system based on map image erosion and dilation according to claim 1, characterized in that, The first convolution kernel and the second convolution kernel have the same number of bits.

3. The data processing system based on map image erosion and dilation according to claim 1, characterized by, After S205, the following steps are further included: S2051, obtaining a number A1 of expanded pixel points in the second intermediate geographic image; S2053, obtaining a number A2 of eroded pixel points in the final geographic image; S2055, A1 and A2 satisfy the following condition: A1-A2=A, A is a preset expansion width value.

4. The data processing system based on map image erosion and dilation according to claim 1, characterized by, In S201, the pre-processing of the target geographic image includes the following steps: S2011, performing image binarization on the target geographic image to obtain a preliminary geographic image; S2013, deleting a boundary block in the preliminary geographic image to obtain the first intermediate geographic image, wherein the boundary block refers to a region in the preliminary geographic image that does not include the boundary region corresponding to the target POI.

5. The data processing system based on map image erosion dilation according to claim 4, characterized in that, In S2013, the boundary block is obtained by the following steps: S10, obtaining an RGB value B1 of a boundary point B; S30, obtaining RGB values {B2, B3, B4} of 3 neighborhoods of the boundary point B; S50, when B1 = B t t is added to the border block, t = 2 or t = 3 or t = 4;​ S70, traverse B t 3 neighborhoods of the corresponding boundary point; S90, when the RGB values are all different, ending the iteration.

6. The data processing system based on map image erosion and dilation according to claim 1, characterized by, After S101, the following steps are further included: In S1011, the jth geographical image P in the mth type of geographical image set is acquired. mj = (W mj , L mj ); S1013, when W mj ≥ X and L mj ≥ Y, execute S103, X is the width of the target map image corresponding to the target POI, and Y is the length of the target map image corresponding to the target POI.

Images