A method for graph regularization of beidou positioning data with geographic information features

Abstract

This invention discloses a method for regularizing BeiDou positioning data graphics with geographic information features, belonging to the field of satellite positioning data processing technology, and includes the following steps: Step 1, calculate the plot graphic P w The working area S w1 and contour area S w2 Step 2: For the current plot graphic P w All trajectory points are rotated counterclockwise by θ° according to the coordinate rotation transformation formula; Step 3: Generate the circumscribed rectangle and obtain its length L, width w, and area S. w3 Step 4: Calculate the current plot shape P. w If the area ratio r is greater than r_th, then proceed to step 5 and other steps. The beneficial effects of the present invention are: (1) It can identify, classify and effectively remove different types of irregular plot graphics; (2) It can effectively extract and retain normal plot graphics from irregular plot graphics; (3) The method is simple to operate, has low time complexity and strong engineering feasibility; (4) It can be used alone or as a supplement to existing methods.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of satellite positioning data processing, and particularly relates to a Beidou positioning data graph regularization method with geographical information characteristics. BACKGROUND

[0002] The existing platform calculates the working area of a machine in the following manner: a Beidou positioning module installed on the machine is used to realize real-time positioning of the machine, and track points are uploaded to the platform; the platform draws lines and divides a field graph according to the data characteristics of the track points, and then calculates the working area.

[0003] At present, irregular field graphs may appear when the platform divides a field graph, mainly referring to two types of field graphs formed by the track of the machine on a road and between working areas and the field graph formed by the drift points of the positioning device when the machine is parked. Common irregular field graphs can be summarized as follows: as shown in FIG. 1, a field graph formed by the drift of the machine when parked in place; as shown in FIG. 2, a field graph on a road close to a straight line; as shown in FIG. 3, a field graph on a road with an arbitrary shape; and as shown in FIG. 4, a field graph on a road connected to a normal working field graph. Figure 1 Figure 2 Figure 3 Figure 4

[0004] For the above problems, the inventor of the present application previously applied for an invention patent for a method for calculating the working area of a farm machine based on the space-time characteristics of BDS / GPS positioning data, which discloses using an improved DBSCAN clustering algorithm to process the field graph on a road and using a heading difference statistical determination algorithm to process the drift field graph, and both have good processing effects. However, the efficiency of the clustering algorithm is sometimes difficult to guarantee when the amount of track point data is large, and when the track points of the irregular field graph are dense, the recognition effect is affected. The heading difference statistical determination algorithm is based on the statistical thought, and sometimes a large amount of data is required to reflect the rule, and the algorithm is suitable for large-range drift situations, and small-range drift affects the recognition effect. SUMMARY

[0005] The present application aims to provide a Beidou positioning data graph regularization method with geographical information characteristics, which effectively processes irregular field graphs from the geometric characteristics of the field graph and can be used alone or as a supplementary method for existing methods.

[0006] To solve the above technical problems, the technical scheme adopted by the present application is as follows: a Beidou positioning data graph regularization method with geographical information characteristics, comprising the following steps: step 1, calculating the working area S of a field graph P in a list P of field graphs i w w1 ​​​​​​and the contour area S w2 , wherein i = 1…n, n is the number of land block patterns, and w ranges from 1 to n; step 2, taking a as an interval, counting the direction angles of all the trajectory lines of the current land block pattern P w , and averaging all the direction angles of the interval with the most trajectory lines to obtain θ, which is taken as the main direction of the land block pattern; according to the coordinate rotation transformation formula, the all the trajectory points of the current land block pattern P w are rotated counterclockwise by θ°, wherein the coordinate rotation transformation formula is as follows:

[0007] x’ = x cos(θ) - y sin(θ)

[0008] y’ = y cos(θ) + x sin(θ)

[0009] step 3, according to the coordinates of the rotated points, taking the maximum values x ’ max and y ’ max in the horizontal and vertical coordinate directions, and the minimum values x ’ min and y ’ min to generate a circumscribed rectangle, and obtaining the length L, the width w and the area S w3 of the circumscribed rectangle, wherein L = max(x ’ max - x ’ min , y ’ max - y ’ min ), w = min(x ’ max - x ’ min , y ’ max - y ’ min ), and S w3 = L × w; step 4, calculating the area ratio r of the current land block pattern P w , if r > r_th, executing step 5, otherwise executing step 7, wherein r = S w1 / S w3 ; step 5, calculating the rectangularity d of the current land block pattern P w , if d > d_th, executing step 6, otherwise executing step 10, wherein d = S w2 / S w3 ; step 6, calculating the body ratio T of the current land block pattern P w , if T > T_th, retaining the current land block pattern P wStep 7: Divide the circumscribed rectangle into L / max(v,plouth) sub-rectangles along the length L and number each sub-rectangle. Count the number of plot points (num) within each sub-rectangle. k Calculate the point density q for each sub-rectangle. k And map it into (0,1) to get qz k =q k / (max(q k )-min(q k )),

[0010] Where, q k =num k / (w×max(v,plouth),k=1…L / max(v,plouth),v is the average velocity of all trajectory points on the plot map,plouth is the width of the machine; Step 8, for all qz k Calculate the mean u and variance D. If D > D_th, proceed to step 9; otherwise, proceed to step 10. Step 9: Delete qz. k <u-3×D 1 / 2 The corresponding trajectory points and rectangle numbers within the sub-rectangles are assigned. For the remaining sub-rectangles, the trajectory points and rectangle numbers within sub-rectangles without adjacent numbers are deleted. The remaining sub-rectangle numbers are merged, and the coordinates of all trajectory points are rotated clockwise by θ° before proceeding to step 11; Step 10: The current plot graphic P... w Identified as an irregular plot shape, delete the current plot shape P. w All trajectory points before coordinate rotation and the plot graphic number; Step 11: Determine the current plot graphic P w Is this the last plot of land? If so, reorder the plot numbers P' of all plots in sequence. j j=1…m, where m is the number of updated plot graphics; otherwise w=w+1 and return to step 1.

[0011] The beneficial effects of the present invention are: (1) It can identify, classify and effectively remove different types of irregular plot graphics; (2) It can effectively extract and retain normal plot graphics from irregular plot graphics; (3) The method is simple to operate, has low time complexity and strong engineering implementation; (4) It can be used alone or as a supplement to existing methods.

[0012] The present invention will now be described in detail with reference to the accompanying drawings. Attached Figure Description

[0013] Figure 1 A schematic diagram of the terrain generated by stationary parking drift;

[0014] Figure 2 Schematic diagram of field pattern on road for approaching a straight line;

[0015] Figure 3 Schematic diagram of field pattern on road for approaching an arbitrary shape;

[0016] Figure 4 Schematic diagram of field pattern on road connected to normal operation field pattern.

[0017] Figure 5 Flow chart of the method of the present application. DETAILED DESCRIPTION

[0018] Referring to the accompanying drawings Figure 5 , the present application provides a Beidou positioning data pattern regularization method with geographic information features, comprising the following steps.

[0019] Step 1, calculating a field pattern list P i i=1…n, n is the number of field patterns, and w ranges from 1 to n. w The operation area S w1 and the contour area S w2 of each field pattern P i are calculated.

[0020] The field pattern list is the result of field pattern division, and the division method can be an existing clustering method or other methods. The method of the present application is based on the division result, taking each field pattern as a processing unit to perform the next recognition and removal operation. The information of each field pattern includes the longitude and latitude coordinates, heading angle, and speed of all track points in the field pattern based on Beidou positioning data; the length and direction of the operation line, the number of operation lines; and the machine width and other data.

[0021] The operation area of the field pattern is the length of each operation line multiplied by the width and then accumulated, and the contour area is obtained by a geometric polygon processing method.

[0022] In the specific implementation, the value of w in P w starts from 1.

[0023] Step 2, taking α as an interval, counting the direction angles of all track lines in the current field pattern P w , and taking the average value of all direction angles of the interval with the most track lines as θ and taking θ as the main direction of the field pattern.

[0024] Rotating all track points of the current field pattern P w counterclockwise by θ° according to the coordinate rotation transformation formula, wherein the coordinate rotation transformation formula is as follows:

[0025] x' = x cos(θ) - y sin(θ)

[0026] y' = y cos(θ) + x sin(θ).

[0027] x and y are respectively the horizontal and vertical coordinate values of the trajectory point before rotation transformation, and x' and y' are respectively the horizontal and vertical coordinate values of the trajectory point after rotation transformation.

[0028] Step 3, according to the rotated point coordinates, the maximum value x ’ max and y ’ max in the horizontal and vertical coordinate directions ’ min and y ’ min are generated to generate the circumscribed rectangle, and the length L, width w and area S w3 of the circumscribed rectangle are obtained.

[0029] wherein, L = max(x ’ max - x ’ min , y ’ max - y ’ min ), w = min(x ’ max - x ’ min , y ’ max - y ’ min ), S w3 = L x w.

[0030] The purpose of steps 2 and 3 is to obtain the circumscribed rectangle of the landform, and the general method for obtaining the circumscribed rectangle is to find the maximum and minimum latitude and longitude of the point set. However, the direction of the landform is not necessarily north-south or east-west, and if a diagonal circumscribed rectangle consistent with the direction of the landform is obtained, the direction needs to be determined first. The interval can be divided according to α, which can be any value in (0, 180°), and α is taken as 30° in this case. The direction with the most working lines is used as a reference to obtain the diagonal circumscribed rectangle by statistical method. Then, all point coordinates are converted to new coordinates by rotation method, and the length, width and area of the diagonal circumscribed rectangle are obtained as input parameters for the following determination logic.

[0031] Step 4, calculate the area ratio r of the current landform P w , if r > r_th, execute step 5, otherwise execute step 7, wherein r = S w1 / S w3 . r_th is the area ratio threshold.

[0032] The ratio of the working area and the area of the circumscribed rectangle can be used as the first basis for determining the field pattern on the road. Generally, the field pattern of normal working is close to a rectangle, that is, the ratio is close to 1; while Figure 3 In the case shown, the ratio is very small, even approaching 0; while Figure 4 The case shown is between 0 and 1, depending on the area proportion of the normal field pattern. r_th is generally taken as a value between (0, 1), which can be taken as 0.85 here. Greater than this value, it can be a normal field pattern or Figure 1 the case shown or Figure 2 the case shown, the next step of step 5 needs to be judged; otherwise, it is determined that the field pattern is an irregular field pattern, belonging to Figure 3 or one of the cases shown in 4, because Figure 4 The case shown contains a normal field pattern, so it cannot be directly deleted here and needs to be further judged by step 7 to distinguish.

[0033] Step 5, calculate the rectangularity d of the current field pattern P w , if d > d_th, execute step 6, otherwise execute step 10, where d = S w2 / S w3 . d_th is the rectangularity threshold.

[0034] Rectangularity represents the degree to which a figure is close to a rectangle. The ratio of the contour area of a normal field pattern and the area of the circumscribed rectangle is theoretically 1. Considering the actual working situation, it is impossible to be completely rectangular, but at least it is close to 1. Figure 1 In the case shown, the figure generated by the drift field pattern is an irregular polygon, even close to a circle. The rectangularity value of a circle is π / 4, about 0.78. Therefore, d_th can be taken as a value between (π / 4, 1), which is taken as 0.9 in this example. Less than this value can be identified as Figure 1 the drift field pattern shown in the case, the deletion operation of step 10 is executed; greater than this value is considered to be a normal field pattern or Figure 2 one of the cases shown in 4, which needs step 6 for further discrimination.

[0035] Step 6, calculate the body ratio T of the current field pattern P w , if T > T_th, keep all the track points of the current field pattern P w before coordinate rotation and execute step 11, otherwise execute step 10, where T = w / L. T_th is the body ratio threshold.

[0036] Body ratio represents the "fatness" of a rectangular figure, Figure 2The illustrated straight plot of land on the road has a variable length but a very narrow width, typically one width. Considering that the actual travel distance on the road is much greater than the width of a single piece of equipment, T_th generally takes a value between (0,1). In this example, it can be taken as 0.05, meaning that a value less than this is considered... Figure 1 For the plot of land shown, perform the deletion operation in step 10; otherwise, it can be determined as a normal plot of land by combining the previous steps 4 and 5. At this point, Figure 1 The situation shown Figure 2 The situation shown and the normal plot graphics have been processed.

[0037] Step 7: Divide the circumscribed rectangle into L / max(v,plouth) sub-rectangles along the length L, number each sub-rectangle, and count the number of plot points (num) within each sub-rectangle. k Calculate the point density q for each sub-rectangle. k And map it into (0,1) to get qz k =q k / (max(q k )-min(q k )),

[0038] Where, q k =num k / (w×max(v,plouth),k=1…L / max(v,plouth),v is the average velocity of all trajectory points of the plot,plouth is the width of the machine.

[0039] Step 8: For all qz k Calculate the mean u and variance D. If D > D_th, proceed to step 9; otherwise, proceed to step 10. D_th is the variance threshold.

[0040] Step 9: Delete qz k <u-3× D 1 / 2 For the corresponding sub-rectangles, assign trajectory points and rectangle numbers, and for the remaining sub-rectangles, delete the trajectory points and sub-rectangle numbers within sub-rectangles that do not have adjacent numbers.

[0041] After merging the remaining sub-rectangle numbers and rotating the coordinates of all trajectory points clockwise by θ°, proceed to step 11.

[0042] Steps 7 to 9 are for differentiation Figure 3 The situation shown and Figure 4In the case shown, step 7 uses the unit plot pattern point density to divide the circumscribed rectangle into several sub-rectangles of the same size, and then calculates the point density of each rectangle. Since the size of the circumscribed rectangle of each plot pattern is different, the point density cannot be regulated by a unified standard, and therefore needs to be normalized, i.e. mapped to (0, 1); Figure 3 In the case shown, step 7 uses the unit plot pattern point density to divide the circumscribed rectangle into several sub-rectangles of the same size, and then calculates the point density of each rectangle. Since the size of the circumscribed rectangle of each plot pattern is different, the point density cannot be regulated by a unified standard, and therefore needs to be normalized, i.e. mapped to (0, 1); Figure 4 The essential difference between the case shown and Figure 3 In the case shown, the point density of each sub-rectangle is similar and very small, while Figure 4 In the case shown, the point density of each sub-rectangle is similar and very small, while Figure 3 In the case shown, the point density of each sub-rectangle is similar and very small, while Figure 4 In the case shown, according to the difference in point density, the sub-rectangles with a point density lower than u / 10 are determined as road plot patterns and deleted in step 9. Among the remaining sub-rectangles, the ones with a large plot pattern point density but no adjacent sub-rectangles are single road plot patterns with the working line direction consistent with the direction of the width of the circumscribed rectangle, and should also be removed. Finally, the remaining sub-rectangles are numbered and merged, and the coordinates of all the track points are rotated clockwise to restore the normal plot patterns after filtering. Thus, Figure 3 In the case shown, step 7 uses the unit plot pattern point density to divide the circumscribed rectangle into several sub-rectangles of the same size, and then calculates the point density of each rectangle. Since the size of the circumscribed rectangle of each plot pattern is different, the point density cannot be regulated by a unified standard, and therefore needs to be normalized, i.e. mapped to (0, 1); Figure 4 The case shown is processed.

[0043] Step 10, delete the current plot pattern P w identified as an irregular plot pattern, delete the current plot pattern P w and the coordinates of all the track points before rotation.

[0044] Step 11, determine whether the current plot pattern P w is the last one. If so, rearrange the numbers of all the plot patterns P' j in order, j = 1…m, m is the number of updated plot patterns; otherwise, w = w + 1 and return to step 1.

[0045] After irregular plot pattern recognition and removal, the number of new plot patterns may be less than the original number due to the removal of Figures 1-3 the plot patterns in the case shown, or may be more due to the processing of Figure 4The situation shown results in an increase in the number, and the change in the plot pattern of the land block caused by the operation, so that the remaining normal operation land block pattern needs to be renumbered. The land block pattern list after renumbering is the final result obtained after processing the irregular land block pattern, thereby realizing the regularization of the land block pattern.

[0046] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the specific embodiments of the present application can be modified or some technical features can be replaced by equivalent ones; without departing from the spirit of the technical solutions of the present application, they should be covered in the technical solution range of the present application claimed.

Claims

1. A method for graph regularization of Beidou positioning data with geographic information features, characterized in that, The method comprises the following steps: Step 1, calculating the list of plot patterns P i The central plot pattern P w The working area S w1 And the contour area S w2 , Wherein, i = 1…n, n is the number of land patterns, and the value range of w is 1-n; Step 2, count the direction angle of all track lines in the current plot pattern P with a as an interval, and average all direction angles of the interval with the most track lines to obtain θ and take θ as the main direction of the plot pattern. w Step 2, count the direction angle of all track lines in the current plot pattern P with a as an interval, and average all direction angles of the interval with the most track lines to obtain θ and take θ as the main direction of the plot pattern. All the track points of the current plot pattern P w are rotated counterclockwise by θ° according to the coordinate rotation transformation formula, Wherein, the coordinate rotation transformation formula is as follows: x'=xcos(theta)-ysin(theta) y'=ycos(theta)+xsin(theta); Step 3, according to the rotated point coordinates, the maximum value x ’ max and y ’ max in the horizontal and vertical coordinate directions ’ min and y ’ min Generate the circumscribed rectangle to obtain the length L, width w and area S of the circumscribed rectangle w3 , where L = max(x ’ max - x ’ min , y ’ max - y ’ min ), w = min(x ’ max - x ’ min , y ’ max - y ’ min ), S w3 = L x w; Step 4, calculate the area ratio r of the current plot pattern P, if r>r_th, execute Step 5, otherwise execute Step 7, where r=S w / S w1 ; S w3 ; Step 5, calculate the rectangularity d of the current plot pattern P, if d > d_th, then execute Step 6, otherwise execute Step 10, where d = S w / S w2 / S w3 ; Step 6, calculate the body posture ratio T of the current plot pattern P, if T > T_th, keep the current plot pattern P w Step 10, if T < T_th, rotate the coordinate system of the current plot pattern P by an angle of 90°, and execute Step 11, otherwise execute Step 10, wherein T = w / L w Step 11, rotate all the track points before coordinate rotation of the current plot pattern P by an angle of 90°, and execute Step 12, otherwise execute Step 10, wherein T = w / L Step 7, divide the outer rectangle into L / max(v,plouth) sub-rectangles along the direction of length L and number each sub-rectangle, count the number of land pattern points num in each sub-rectangle k , calculate the point density q for each sub-rectangle k , and map it into (0,1) to obtain qz k = q k / (max(q k ) - min(q k )) where q k = num k / (w x max(v, plouth), k = 1...L / max(v, plouth), v is the average speed of all track points of the plot pattern, and plouth is the width of the implement. Step 8, for all qz k The mean u and variance D are calculated. If D > D_th, step 9 is executed, otherwise step 10 is executed. Step 9, delete qz k < u - 3 x D 1 / 2 corresponding to the trajectory points and the rectangular number within the sub-rectangle and to the remaining sub-rectangles, delete the trajectory points and the sub-rectangular number within the sub-rectangles without adjacent numbers, The remaining sub-rectangular number is merged, and all trajectory point coordinates are rotated by theta degrees clockwise to execute step 11. Step 10, delete the current parcel figure P w Step 11, identify the irregular parcel figure, delete the current parcel figure P w All track points before coordinate rotation and the parcel figure number Step 11, judge whether the current plot pattern P is the last plot pattern, if yes, re-arrange the number of all plot patterns P in order w j , j = 1…m, m is the updated plot pattern number; otherwise w = w + 1 and return to step 1.​ 2. The method for Beidou positioning data graph regularization with geographic information features according to claim 1, characterized in that, The value range of r_th is 0 < r_th < 1.

3. The method for Beidou positioning data graph regularization with geographic information features according to claim 1, characterized in that, The value range of d_th is pi / 4 < d_th < 1.

4. The method for Beidou positioning data graph regularization with geographic information features according to claim 1, characterized in that, The value range of T_th is 0 < T_th < 1.

5. The method for Beidou positioning data graph regularization with geographic information features according to claim 1, characterized in that, The value range of D_th is 0 < D_th < 0.

1.

6. The method for Beidou positioning data graph regularization with geographic information features according to claim 1, characterized in that, The value range of alpha is 0° < alpha < 180°.

Images