CN110730001B - Track compression method for ship berthing - Google Patents

Abstract

The invention discloses a track compression method for a ship hanging port, which is based on a window sliding algorithm for track compression, wherein the algorithm idea is to always process only 3 points, and the data is transmitted in a streaming form by applying a gradual compression idea for online compression, so that all the data are not required to be stored and processed, and the data processing efficiency is high; and judging whether the initial track point and the end track point are the same point or not before calculating the vertical distance from the current track point to be compressed to the straight line of the initial track point and the end track point, so that the situation that track compression cannot be performed due to incapacity of calculating the distance from the current port to the connecting line of the initial port and the end port is eliminated, and the algorithm can run efficiently.

Description

Track compression method for ship berthing

Technical Field

The invention relates to the field of application of geographic information systems, in particular to a track compression method for a ship to lean on a port.

Background

Data compression is one of key technologies for preprocessing track data, and the number of original track points used for representing is reduced on the premise of guaranteeing the characteristics and the topological structure of the original track data. Track compression can be divided into two main categories according to whether all ship track point data are needed before compression: offline compression and online compression. In the offline compression algorithm, the elimination or retention of a certain point is determined according to the relation between the point and all track points, and the deformation generated by the point is smaller than that of the original track, but the compression efficiency is low, and particularly, the method is used for mass ship track data exceeding the computer storage capacity. The online algorithm determines the choice of points by analyzing the relationship between adjacent ship track points, and the method is simple, convenient and efficient and has a good compression effect on massive ship track data.

The forced installation and use of the shipborne automatic identification system are beneficial to realizing the monitoring of marine systems and shipborne companies on ships, but the storage and use of massive AIS data bring great challenges, various ship track compression methods are created for extracting the normal behavior mode of the ships from the massive data, the Target-Puck method is the most classical ship track compression method at present, the method can be described as the method for virtually connecting an initial track point and a termination track point of one track with one straight line, the vertical distance between the rest track points and the straight line is obtained, the largest one is compared with a preset threshold value, if the maximum distance is smaller than or equal to the threshold value, all track points between two ends of the straight line are deleted, if the maximum vertical distance is larger than the threshold value, the track points which are the largest from the straight line are reserved, the two parts are divided into two parts, the method is reused until the two parts cannot be compressed finally, the algorithm idea is that the track starting point and the termination point are connected with the track, the middle point to the connecting line distance is obtained, and the key point is selected according to the set distance threshold value, so that the compression purpose is realized. However, the method has huge calculation amount and long time consumption, and when the data amount exceeds the maximum storage level of a computer, the on-line track compression cannot be realized by the Targelas-Pockey method.

Disclosure of Invention

The track compression method for the ship landing port is capable of compressing ship track data on line, and is kept compressed all the time in a state that the data is continuously updated, so that the defect that online track compression cannot be realized by the track compression method when the data exceeds the maximum storage level of a computer in the prior art is overcome.

In order to achieve the above object, the present invention adopts the following technical scheme:

a track compression method for a ship berth, the method comprising the steps of:

step one: initializing a sliding window and setting a track compression distance threshold;

step two: judging whether the initial track point and the termination track point are the same track point, if so, moving the termination track point backwards by one point, executing the fifth step, otherwise, executing the third step;

step three: calculating the vertical distance from the current track point to be compressed in the sliding window to the straight line of the initial track point and the ending track point;

step four: comparing the vertical distance from the current track point to be compressed to the straight line of the initial track point and the end track point with a track compression distance threshold value, adding the current track point to be compressed into a compression track set if the track compression distance threshold value is smaller, setting a new sliding window by taking the current track point to be compressed as the initial track point, and if not, moving the current track point to be compressed and the end track point backwards by one point;

step five: judging whether the ending track point is the last track point, if so, adding the ending track point into the compressed track set to complete track compression, otherwise, returning to the step two.

In the scheme, track compression is performed based on a window sliding algorithm, the algorithm idea is to always process only 3 points, the idea of gradual compression is applied, data is transmitted in a stream form, online compression is performed, all the data are not required to be stored and then processed, and the data processing efficiency is high; and by judging whether the initial track point and the termination track point are the same point or not before calculating the vertical distance from the current track point to the straight line of the initial track point and the termination track point, excluding the condition that the initial track point and the termination track point are the same point, the vertical distance from the current track point to the straight line of the initial track point and the termination track point can be always calculated, and the algorithm can not cause the incapability of track compression because the distance from the current harbor to the connecting line of the initial harbor and the termination harbor can not be calculated, so that the algorithm can run efficiently.

Preferably, the method specifically comprises the following steps:

step 1: assuming that a track set of a ship to be compressed is P= { Pi }, wherein Pi is an I-th track point, I epsilon [1, N ], and N is the total number of track points to be compressed; let the sliding window be { Ps, pt, pe }, wherein Ps and Pe are the initial track point and the end track point of the sliding window respectively, e and s represent the positions of the initial track point and the end track point of the sliding window respectively, pt is the track point to be compressed currently in the sliding window, and t represents the position of the track point to be compressed currently in the sliding window; let the compressed track distance threshold be L;

step 2: initializing a sliding window, so that s=1, t=2 and e=3; meanwhile, let the compressed trace set q= { P1};

step 3: judging whether Ps and Pe are the same point, if Ps is equal to Pe, adding Pt into the track set Q, resetting the sliding window to enable s=t, t=t+1 and e=t+2, and entering step 6; otherwise, enter step 4;

step 4: calculating the distance dt from the point Pt to the straight line Ps-Pe;

step 4: calculating the distance dt from the point Pt to the straight line Ps-Pe;

step 5: comparing the magnitudes of dt and L, adding Pt to the trace set Q if dt is greater than L, resetting the sliding window to let s=t, t=t+1, e=t+2; otherwise, pt and Pe are moved backward by one point at the same time, let t=t+1, e=t+2;

step 6: comparing the sizes of e and N, if e is larger than N, adding Pt into the track set Q, and completing track compression of the ship landing port; otherwise, go back to step 3.

Preferably, the compressed track distance threshold l=6.

Preferably, the track distance threshold l=7 is compressed.

The online compression method has the advantages that online compression of the ship track is realized based on the sliding pane algorithm without acquiring the termination point of the ship track; the initial track point and the end track point are judged, when the initial track point and the end track point are the same point, the current track point is reserved, and the whole sliding window is moved backwards by one point, so that the algorithm cannot compress the track because the initial track point and the end track point are the same point.

Drawings

Fig. 1 is a flow chart of the present invention.

Detailed Description

The invention is further described below with reference to the drawings and specific embodiments.

Specific examples: a track compression method for a ship berth, the method comprising the steps of:

step one: initializing a sliding window and setting a track compression distance threshold;

step two: judging whether the initial track point and the termination track point are the same track point, if so, moving the termination track point backwards by one point, executing the fifth step, otherwise, executing the third step;

step three: calculating the vertical distance from the current track point to be compressed in the sliding window to the straight line of the initial track point and the ending track point;

step four: comparing the vertical distance from the current track point to be compressed to the straight line of the initial track point and the end track point with a track compression distance threshold value, adding the current track point to be compressed into a compression track set if the track compression distance threshold value is smaller, setting a new sliding window by taking the current track point to be compressed as the initial track point, and if not, moving the current track point to be compressed and the end track point backwards by one point;

step five: judging whether the ending track point is the last track point, if so, adding the ending track point into the compressed track set to complete track compression, otherwise, returning to the step two.

The algorithm idea is to always process only 3 points, apply the idea of gradual compression, transmit data in the form of stream, and perform online compression. For example, the initial window is { P1, P2, P3}, P1 is an initial point, is set as a key feature point to be reserved, judges whether P1 and P3 are the same point, if P3 is discarded, P4 is added, judgment is carried out again, if P1 and P3 are not the same point, the distance between P2 point and the connecting lines P1-P3 of 2 endpoints is calculated, and the calculation formula is as follows;

if the distance is smaller than the preset threshold distance L, discarding the middle point P2, and adding P4; the current window is updated to { P1, P3, P4}, whether P1 and P4 are the same point is judged, if not, the distance between the P3 point and the connection lines P1-P4 of 2 endpoints is calculated; if the current window is larger than the preset threshold distance L, P3 is marked as key feature point reservation, the window slides and P5 is added, and the current window is updated to { P3, P4, P5}; p3 is used as a new sliding starting point, and is gradually compressed, and so on.

The method specifically comprises the following steps:

step 1: assuming that a track set of a ship to be compressed is P= { Pi }, wherein Pi is an I-th track point, I epsilon [1, N ], and N is the total number of track points to be compressed; let the sliding window be { Ps, pt, pe }, wherein Ps and Pe are the initial track point and the end track point of the sliding window respectively, e and s represent the positions of the initial track point and the end track point of the sliding window respectively, pt is the track point to be compressed currently in the sliding window, and t represents the position of the track point to be compressed currently in the sliding window; let the compressed track distance threshold be L;

step 2: initializing a sliding window, so that s=1, t=2 and e=3; meanwhile, let the compressed trace set q= { P1};

step 3: judging whether Ps and Pe are the same point, if Ps is equal to Pe, adding Pt into the track set Q, resetting the sliding window to enable s=t, t=t+1 and e=t+2, and entering step 6; otherwise, enter step 4;

step 4: calculating the distance dt from the point Pt to the straight line Ps-Pe;

step 4: calculating the distance dt from the point Pt to the straight line Ps-Pe;

step 5: comparing the magnitudes of dt and L, adding Pt to the trace set Q if dt is greater than L, resetting the sliding window to let s=t, t=t+1, e=t+2; otherwise, pt and Pe are moved backward by one point at the same time, let t=t+1, e=t+2;

step 6: comparing the sizes of e and N, if e is larger than N, adding Pt into the track set Q, and completing track compression of the ship landing port; otherwise, go back to step 3.

In this embodiment, the compressed track distance threshold l=6, n=25, pi (Xi, yi) is coordinates of each port of the ship in the running process, the coordinates of the track point Pi (Xi, yi) are composed of longitude and latitude of each port of the ship, xi is longitude of the ith port of the ship, yi is latitude of the ith port of the ship, dt is a distance from the point Pt to the straight line Ps-Pe, dt is denoted as d (Ps, pt, pe) in this embodiment, and the data in the set P are substituted into the above steps, and the following is further performed in combination with specific data:

the first step: the track sets of the ship to be compressed are p= { P1 (55.050503,25.002617), P2 (122.416667,30.666667), P3 (122.066667,30.633333), P4 (128.75,35.066667), P5 (128.75,35.066667), P6 (121.8833,29.9333), P7 (-123.116667,49.283333), P8 (-123.388233,48.415872), P9 (-122.3333,47.6), P10 (128.7,35.116667), P11 (113.945392,22.360472), P12 (114.25,22.5833), P13 (128.7,35.116667), P14 (-123.116667,49.283333), P15 (-123.388233,48.415872), P16 (-123.388233,48.415872), P17 (-122.3333,47.6), P18 (128.7,35.116667), P19 (113.945392,22.360472), P20 (114.25,22.5833), P21 (121.883333,29.933333), P22 (122.416667,30.683333), P23 (122.066667,30.633333), P24 (128.7,35.116667), P25 (-123.116667,49.283333) }; let the sliding window be { Ps, pt, pe }, wherein Ps and Pe are the initial track point and the end track point of the sliding window respectively, e and s represent the positions of the initial track point and the end track point of the sliding window respectively, pt is the track point to be compressed currently in the sliding window, and t represents the position of the track point to be compressed currently in the sliding window; let the compressed track distance threshold be l=6;

and a second step of: initializing a sliding window, so that s=1, t=2, e=3, ps=p1, pt=p1, pe=p3; meanwhile, let the compressed trace set q= { P1};

and a third step of: determining if point P1 and point P3 are the same point, P1 (55.050503,25.002617) is not equal to P3 (122.066667,30.633333), so step 4 of the algorithm is performed;

fourth step: calculating the distance dt, d (P1, P2, P3) = 0.00391311865678253 from the point P2 to the straight line P1-P3;

fifth step: 0.0039 is less than 6, i.e. dt is less than L, pt and Pe are simultaneously moved back by one point, let t=t+1, e=t+2, to obtain pt=p3, pe=p4;

sixth step: 4 is less than 25, i.e. e is less than N, so step 3 of the algorithm is performed;

seventh step: determining if point P1 and point P4 are the same point, P1 (55.050503,25.002617) is not equal to P4 (128.75,35.066667), so step 4 of the algorithm is performed;

eighth step: calculating the distance dt, d (P1, P3, P4) = 3.48831642459002 from the point P3 to the straight line P1-P4;

ninth step: 3.4883 is smaller than 6, i.e. dt is smaller than L, pt and Pe are simultaneously moved backward by one point, let t=t+1, e=t+2, and pt=p4, pe=p5 are obtained;

tenth step: 5 is less than 25, i.e., e is less than N, so step 3 of the algorithm is performed;

eleventh step: determining if point P1 and point P5 are the same point, P1 (55.050503,25.002617) is not equal to P5 (128.75,35.066667), so step 4 of the algorithm is performed;

twelfth step: calculating the distance dt, d (P1, P4, P5) = 8.8001138829084E-16 from the point P4 to the straight line P1-P5;

thirteenth step: 8.8001138829084E-16 is smaller than 6, i.e. dt is smaller than L, pt and Pe are simultaneously moved back by one point, let t=t+1, e=t+2, to obtain pt=p5, pe=p6;

fourteenth step: 6 is less than 25, i.e., e is less than N, so step 3 of the algorithm is performed;

fifteenth step: determining if point P1 and point P6 are the same point, P1 (55.050503,25.002617) is not equal to P6 (121.8833,29.9333), so step 4 of the algorithm is performed;

sixteenth step: calculating the distance dt, d (P1, P5, P6) = 4.61422613535309 from the point P5 to the straight lines P1-P6;

seventeenth step: 4.61422613535309 is smaller than 6, i.e. dt is smaller than L, pt and Pe are simultaneously moved backward by one point, let t=t+1, e=t+2, and pt=p6, pe=p7 are obtained;

eighteenth step: 7 is less than 25, i.e. e is less than N, so step 3 of the algorithm is performed;

nineteenth step: determining whether the point P1 and the point P7 are the same point, wherein P1 (55.050503,25.002617) is not equal to P7 (-123.116667,49.283333), so that step 4 of the algorithm is executed;

twenty-step: calculating the distance dt, d (P1, P6, P7) = 13.9101141362529 from the point P6 to the straight line P1-P7;

twenty-first step: 13.9101141362529 is greater than 6, i.e. dt is greater than L, pt is added to the trace set Q, the sliding window is reset, s=t, t=t+1, e=t+2, resulting in q= { P1, P6}, ps=6, pt=p7, pe=p8;

twenty-second step: 8 is less than 25, i.e. e is less than N, so step 3 of the algorithm is performed;

by analogy, the method in the embodiment of the port coordinate of each wall in the set P is operated, the final compression track set is Pt= { P1, P6, P9, P10, P12, P13, P17, P18, P23 and P25}, the ship track compression method in the embodiment has the advantages that the track compression method is improved based on a sliding pane algorithm, the online ship track is compressed, the occupied resources of the system are fewer, the processing efficiency is higher, and the online compression of the ship track can be realized without acquiring the termination points of the ship track; the starting track point and the ending track point are judged, and when the starting track point and the ending track point are the same point, one point is moved backwards, so that the algorithm cannot compress the track because the starting track point and the ending track point are the same point.

The foregoing description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and any simple modification, variation and equivalent transformation of the above embodiment according to the technical substance of the present invention still fall within the scope of the technical solution of the present invention.

Claims (4)

1. A track compression method for a ship berth, the method comprising the steps of:

step one: initializing a sliding window, wherein the length of the sliding window is three track points, and setting a track compression distance threshold;

step two: judging whether the initial track point and the termination track point are the same track point, if so, moving the termination track point backwards by one point, executing the fifth step, otherwise, executing the third step;

step three: calculating the vertical distance from the current track point to be compressed in the sliding window to the straight line of the initial track point and the ending track point;

step four: comparing the vertical distance from the current track point to be compressed to the straight line of the initial track point and the end track point with a track compression distance threshold value, adding the current track point to be compressed into a compression track set if the track compression distance threshold value is smaller, setting a new sliding window by taking the current track point to be compressed as the initial track point, and if not, moving the current track point to be compressed and the end track point backwards by one point;

step five: judging whether the ending track point is the last track point, if so, adding the ending track point into the compressed track set to complete track compression, otherwise, returning to the step two.

2. The track compression method for a ship berth according to claim 1, comprising the steps of:

step 1: assuming that a track set of a ship to be compressed is P= { Pi }, wherein Pi is an I-th track point, I epsilon [1, N ], and N is the total number of track points to be compressed; let the sliding window be { Ps, pt, pe }, wherein Ps and Pe are the initial track point and the end track point of the sliding window respectively, e and s represent the positions of the initial track point and the end track point of the sliding window respectively, pt is the track point to be compressed currently in the sliding window, and t represents the position of the track point to be compressed currently in the sliding window; let the compressed track distance threshold be L2;

step 2: initializing a sliding window, so that s=1, t=2 and e=3; meanwhile, let the compressed trace set q= { P1};

step 3: calculating a linear distance L1 between the point Ps and the point Pe, if L1=0, making s=t, t=t+1, e=t+2, and entering step 6; otherwise, enter step 4;

step 4: calculating the distance dt from the point Pt to the straight line Ps-Pe;

step 5: comparing the magnitudes of dt and L2, adding Pt to the trace set Q if dt is greater than L2, resetting the sliding window to let s=t, t=t+1, e=t+2; otherwise, pt and Pe are moved backward by one point at the same time, let t=t+1, e=t+2;

step 6: comparing the sizes of e and N, if e is larger than N, adding Pt into the track set Q, and completing track compression of the ship landing port; otherwise, go back to step 3.

3. The track compression method for a ship harbor according to claim 2, wherein the compressed track distance threshold l2=6.

4. The track compression method for a ship harbor according to claim 2, wherein the compressed track distance threshold l2=7.