Arrow positioning method and system applicable to command situation diagram

Abstract

The present invention discloses an arrow positioning method and system suitable for a command situation map, and the steps are as follows: 1) obtaining the position of the arrow in the GIS command situation map; 2) calculating the starting point and end point coordinates of the arrow using a vanishing point detection algorithm. The present invention can use relevant technologies such as knowledge graphs to perform semantic conversion on the arrow symbols that have completed the positioning, and convert the situation map information into text semantic information, which greatly reduces the amount of data transmitted in the command situation map and improves the transmission efficiency.

Description

Technical Field

[0001] The invention belongs to the technical field of target recognition, and in particular relates to an arrow positioning method and system suitable for a command situation diagram. Background Art

[0002] At present, the command and dispatch system based on GIS is widely used in multiple business fields such as emergency response, fire fighting, and smart cities. The command and dispatch system based on GIS tracks and locates the command and operation personnel at the scene of various emergencies, realizes visual remote command and dispatch, and plays an important role in urban governance and emergency rescue. However, the current command and dispatch system needs to transmit image files when transmitting the command situation map, and when the receiving party's network conditions are poor, such as in mountainous areas and forest areas with poor signal coverage, the transmission often takes a lot of time.

[0003] The mainstream GIS command and dispatch system on the market can only transmit traditional command situation picture files, but cannot extract the structured semantic text in the command picture, which requires a large amount of data to be transmitted. Therefore, it is necessary to propose an arrow positioning method to identify the coordinates of the starting point and the end point of the arrow, so as to extract the command structured semantics in the situation picture and output it as text to compress the data volume. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide an arrow positioning method and system suitable for a command situation map, so as to solve the problem that the GIS command and dispatch system in the prior art can only transmit traditional command situation picture files, cannot extract structured semantic text in the command picture, and requires a large amount of data to be transmitted. The present invention uses a target recognition algorithm based on a deep neural network to detect the coverage of the arrow in the command situation map, and then calculates the starting point and pointing point coordinates of the arrow through a vanishing point algorithm, thereby extracting the semantic information in the command situation map, which plays a role in compressing the amount of transmitted data.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] The arrow positioning method of the present invention is applicable to a command situation diagram, and the steps are as follows:

[0007] 1) Obtain the position of the arrow in the GIS command situation picture;

[0008] 2) Use the vanishing point detection algorithm to calculate the coordinates of the arrow’s start and end points.

[0009] Furthermore, the step 1) specifically includes:

[0010] 11) Obtain GIS command situation images, pre-process the acquired images through Gaussian filtering to eliminate existing image noise;

[0011] 12) The preprocessed image is input into the trained deep learning detection model to obtain the pixel area and size of the arrow in the image.

[0012] Furthermore, the step 2) specifically includes:

[0013] 21) Slice the pixel area where the arrow is located, and use the Hough transform edge detection algorithm to obtain an edge image;

[0014] 22) Using a vanishing point detection algorithm to calculate the edge image, obtain the straight line equations Y1 and Y2 of the two edge line segments at the tip of the arrow, and the end point of the arrow; wherein Y1: a1x+b1y+c1=0, Y2: a2x+b2y+c2=0, a and b are variable parameters, and c is a constant;

[0015] 23) According to the equations of the two edge line segments Y1 and Y2 and the angle bisector equation formula (a1x+b1y+c1)±λ(a2x+a2y+c2)=0, Find the angle bisectors Y3 and Y4;

[0016] 24) Calculate the intersection points of Y3, Y4 and the pixel area where the arrow is located obtained in step 12) respectively, calculate the length of the overlapping part of the two angular bisectors and the pixel area where the arrow is located based on the intersection points, and select the angular bisector with the longer overlapping part as the input for subsequent calculations; then calculate the distances between the two intersection points of the overlapping part of the angular bisector with the longer overlapping part and the pixel area where the arrow is located and the end point of the arrow, and select the one with the longest distance as the starting point of the arrow.

[0017] Furthermore, the vanishing point detection algorithm is specifically as follows:

[0018] 221) Use Hough transform to detect straight line segments and merge close line segments into one line segment;

[0019] 222) Sort all the merged line segments according to polar coordinates, and find the intersection points of the lines with adjacent line segments with close angles as candidate points;

[0020] 223) All line segments vote for all candidate points. The voting method is to calculate the angle θ between the line from the end point of the line segment to the specified point and the original line segment. The voting value is |l| is the length of the line segment, u is the robustness parameter, and e is the base of the natural logarithm function;

[0021] 224) Perform hierarchical clustering on the candidate points after voting, and the clustering end condition is that the minimum distance is greater than a certain pixel; for each cluster, calculate the weighted centroid of the votes as the new candidate point; the votes are the sum of the votes of all candidate points in the cluster;

[0022] 225) Select the cluster with the highest number of votes as the output point (i.e., the end point of the arrow), which is the intersection of two straight lines in the straight line segment detected in step 221), and the two straight lines are the edge segments of the arrow tip.

[0023] The arrow positioning system of the present invention is applicable to a command situation diagram, comprising:

[0024] Position acquisition module, used to obtain the position of the arrow in the GIS command situation picture;

[0025] The calculation module is used to calculate the coordinates of the starting point and the end point of the arrow using a vanishing point detection algorithm.

[0026] Beneficial effects of the present invention:

[0027] The present invention can utilize knowledge graphs and other related technologies to perform semantic conversion on the arrow symbols that complete positioning, and convert the situation map information into text semantic information, which greatly reduces the amount of data transmitted in the command situation map and improves transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0029] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with embodiments and drawings. The contents mentioned in the implementation modes are not intended to limit the present invention.

[0030] Reference Figure 1 As shown, the arrow positioning method of the present invention is applicable to the command situation diagram, and the steps are as follows:

[0031] 1) Obtain the position of the arrow in the GIS command situation picture;

[0032] Wherein, the step 1) specifically includes:

[0033] 11) Obtain 6IS command situation image, pre-process the acquired image through Gaussian filtering to eliminate the existing image noise;

[0034] 12) Input the preprocessed image into the trained deep learning detection model to obtain the pixel area (x coordinate and y coordinate) and size (H height and W width) of the arrow in the image.

[0035] 2) Use the vanishing point detection algorithm to calculate the coordinates of the starting and ending points of the arrow;

[0036] Wherein, the step 2) specifically includes:

[0037] 21) Slice the pixel area where the arrow is located, and use the Hough transform edge detection algorithm to obtain an edge image;

[0038] 22) Using a vanishing point detection algorithm to calculate the edge image, obtain the straight line equations Y1 and Y2 of the two edge line segments at the tip of the arrow, and the end point of the arrow; wherein Y1: a1x+b1y+c1=0, Y2: a2x+b2y+c2=0, a and b are variable parameters, and c is a constant;

[0039] 23) According to the equations of the two edge line segments Y1 and Y2 and the angle bisector equation formula (a1x+b1y+c1)±λ(a2x+a2y+c2)=0, Find the angle bisectors Y3 and Y4;

[0040] 24) Calculate the intersection points of Y3, Y4 and the pixel area where the arrow is located obtained in step 12) respectively, calculate the length of the overlapping part of the two angular bisectors and the pixel area where the arrow is located based on the intersection points, and select the angular bisector with the longer overlapping part as the input for subsequent calculations; then calculate the distances between the two intersection points of the overlapping part of the angular bisector with the longer overlapping part and the pixel area where the arrow is located and the end point of the arrow, and select the one with the longest distance as the starting point of the arrow.

[0041] Specifically, the vanishing point detection algorithm is as follows:

[0042] 221) Use Hough transform to detect straight line segments and merge close line segments into one line segment;

[0043] 222) Sort all the merged line segments according to polar coordinates, and find the intersection points of the lines with adjacent line segments with close angles as candidate points;

[0044] 223) All line segments vote for all candidate points. The voting method is to calculate the angle θ between the line from the end point of the line segment to the specified point and the original line segment. The voting value is |l| is the length of the line segment, u is the robustness parameter, and e is the base of the natural logarithm function;

[0045] 224) Perform hierarchical clustering on the candidate points after voting, and the clustering end condition is that the minimum distance is greater than a certain pixel; for each cluster, calculate the weighted centroid of the votes as the new candidate point; the votes are the sum of the votes of all candidate points in the cluster;

[0046] 225) Select the cluster with the highest number of votes as the output point (i.e., the end point of the arrow), which is the intersection of two straight lines in the straight line segment detected in step 221), and the two straight lines are the edge segments of the arrow tip.

[0047] The arrow positioning system of the present invention is applicable to a command situation diagram, comprising:

[0048] Position acquisition module, used to obtain the position of the arrow in the GIS command situation picture;

[0049] The calculation module is used to calculate the coordinates of the starting point and the end point of the arrow using a vanishing point detection algorithm.

[0050] The present invention has many specific application paths. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements can be made without departing from the principle of the present invention. These improvements should also be regarded as the protection scope of the present invention.

Claims

1. An arrow positioning method suitable for a command situation diagram, characterized in that: Here are the steps: 1) Obtain the position of the arrow in the GIS command situation picture; 2) Use the vanishing point detection algorithm to calculate the coordinates of the starting and ending points of the arrow; The step 1) specifically includes: 11) Obtain GIS command situation images, pre-process the acquired images through Gaussian filtering to eliminate existing image noise; 12) Input the preprocessed image into the trained deep learning detection model to obtain the pixel area and size of the arrow in the image; The step 2) specifically includes: 21) Slice the pixel area where the arrow is located, and use the Hough transform edge detection algorithm to obtain an edge image; 22) Using a vanishing point detection algorithm to calculate the edge image, obtain the straight line equations Y1 and Y2 of the two edge line segments at the tip of the arrow, and the end point of the arrow; wherein Y1: a1x+b1y+c1=0, Y2: a2x+b2y+c2=0, a and b are variable parameters, and c is a constant; 23) According to the equations of the two edge line segments Y1 and Y2 and the angle bisector equation formula (a1x+b1y+c1)±λ(a2x+b2y+c2)=0, Find the angle bisectors Y3 and Y4; 24) respectively calculating the intersection points of Y3, Y4 and the pixel area where the arrow is located obtained in step 12), calculating the length of the overlapping part of the two angle bisectors and the pixel area where the arrow is located according to the intersection points, and selecting the angle bisector with the longer overlapping part as the input for subsequent calculations; then calculating the distances between the two intersection points of the overlapping part of the angle bisector with the longer overlapping part and the pixel area where the arrow is located and the end point of the arrow, and selecting the one with the longest distance as the starting point of the arrow; The vanishing point detection algorithm is specifically: 221) Use Hough transform to detect straight line segments and merge close line segments into one line segment; 222) Sort all the merged line segments according to polar coordinates, and find the intersection points of the lines with adjacent line segments with close angles as candidate points; 223) All line segments vote for all candidate points. The voting method is to calculate the angle θ between the line from the end point of the line segment to the specified point and the original line segment. The voting value is |l| is the length of the line segment, u is the robustness parameter, and e is the base of the natural logarithm function; 224) Perform hierarchical clustering on the candidate points after voting, and the clustering end condition is that the minimum distance is greater than a certain pixel; for each cluster, calculate the weighted centroid of the votes as the new candidate point; the votes are the sum of the votes of all candidate points in the cluster; 225) Select the cluster with the highest number of votes as the output point, which is the intersection of two straight lines in the straight line segment detected in step 221), and the two straight lines are the edge segments at the tips of the arrows.

2. An arrow positioning system suitable for a command situation diagram, characterized in that: include: Position acquisition module, used to obtain the position of the arrow in the GIS command situation picture; A calculation module, used to calculate the coordinates of the starting point and the end point of the arrow using a vanishing point detection algorithm; Specifically include: Slice the pixel area where the arrow is located, and use the Hough transform edge detection algorithm to obtain the edge image; The edge image is calculated using a vanishing point detection algorithm to obtain the straight line equations Y1 and Y2 of the two edge line segments at the tip of the arrow, and the end point of the arrow; wherein Y1: a1x+b1y+c1=0, Y2: a2x+b2y+c2=0, a and b are variable parameters, and c is a constant; According to the equations of the two edge line segments Y1 and Y2 and the angle bisector equation formula (a1x+b1y+c1)±λ(a2x+b2y+c2)=0, Find the angle bisectors Y3 and Y4; Calculate the intersection points of Y3 and Y4 with the pixel area where the arrow is located respectively, calculate the length of the overlapping part of the two angle bisectors and the pixel area where the arrow is located according to the intersection points, and select the angle bisector with the longer overlapping part as the input for subsequent calculations; then calculate the distance between the two intersection points of the overlapping part of the angle bisector with the longer overlapping part and the pixel area where the arrow is located and the end point of the arrow, and select the one with the longest distance as the starting point of the arrow; The vanishing point detection algorithm is specifically: Use Hough transform to detect straight line segments and merge close line segments into one line segment; Sort all the merged line segments by polar coordinates, and find the intersection points of the line segments with close angles to the adjacent lines as candidate points; All line segments vote for all candidate points. The voting method is to calculate the angle θ between the line from the end point of the line segment to the specified point and the original line segment. The voting value is |l| is the length of the line segment, u is the robustness parameter, and e is the base of the natural logarithm function; Perform hierarchical clustering on the candidate points after voting. The clustering end condition is that the minimum distance is greater than a certain number of pixels. For each cluster, calculate the weighted centroid of the votes as the new candidate point. The votes are the sum of the votes of all candidate points in the cluster. The cluster with the highest number of votes is selected as the output point, which is the intersection of two straight lines in the detected straight line segment, and the two straight lines are the edge segments at the tip of the arrow.

Images