CN107784079A

CN107784079A - A kind of magnanimity static object management method under space coordinates

Info

Publication number: CN107784079A
Application number: CN201710895323.5A
Authority: CN
Inventors: 李彭伟; 李亚钊; 王玉珠; 吴青松; 郑伟; 丁及堃
Original assignee: CETC 28 Research Institute
Current assignee: CETC 28 Research Institute
Priority date: 2017-09-28
Filing date: 2017-09-28
Publication date: 2018-03-09

Abstract

The invention discloses the magnanimity static object management method under a kind of space coordinates, static object of this method based on spatial scene tree management magnanimity, cutting control is carried out to spatial scene tree simultaneously, plus-minus counting processing is carried out to the data in node according to the observability of spatial scene tree node, and dynamic plus unloading are carried out to data by multithreading offline access technology.This method has higher recall precision, solve the problems, such as in plane scene to object height, bounding box processing it is difficult, effectively solve the problems such as magnanimity static object EMS memory occupation is big, display management is difficult, efficiency is low under space coordinates.

Description

A kind of magnanimity static object management method under space coordinates

Technical field

The present invention relates to the magnanimity target display management method under CyberSpace coordinate system, more particularly to a kind of space to sit Magnanimity static object management method under mark system.

Background technology

Current various GIS-Geographic Information System with emulation platform, the use to static object engender capacity it is big, when Effect is high, the requirement that delay is small, EMS memory occupation is low.As the landmark data in GIS-Geographic Information System is ten hundreds of, space galaxy management In it is even more hundreds of millions of.The static object data of these magnanimity how are efficiently managed, have become the key of measurement system Technology, traditional linear regulation control, two dimensional surface scene management etc. is all increasingly unable to meet demand, such as travel time in linear regulation control It is long, serious image efficiency；Two dimensional surface scene management can not then solve the height problem of target.

The content of the invention

Goal of the invention：For problem above, the present invention proposes the magnanimity static object manager under a kind of space coordinates Method, it is effective to solve the problems such as magnanimity static object EMS memory occupation is big, display management is difficult, efficiency is low under space coordinates.

Technical scheme：To realize the purpose of the present invention, the technical solution adopted in the present invention is：Under a kind of space coordinates Magnanimity static object management method, comprise the following steps：

(1) the static object point of input is standardized, forms the static object data of standard；

(2) spatial scene tree is created, static object data is traveled through, adds data in suitable node, as node is deposited Data are then being deployed to the node, new node is otherwise being created and data is deployed to the node；

(3) cutting control is carried out according to current observation parameters on space scene tree, the data in invisible node is carried out Subtract reference count operation, the data in visible node are carried out plus reference count operates；

(4) start processed offline thread, the data in spatial scene tree count using multithreading and unloaded with adding Management is carried, data of the reference count in scene less than 0 are discharged, the data quoted in scene more than 0 are loaded.

Beneficial effect：Compared with prior art, method advantage disclosed by the invention is：1st, managed by spatial scene tree The static object data of magnanimity, there is higher recall precision；2nd, solve in plane scene to object height, bounding box processing The problem of difficult；3rd, it is whether visible to object count progress add-subtract control based on scenario node, while realized using multithreading The dynamic of target data adds unloading, and the EMS memory occupation amount of system is effectively reduced on the premise of actual use is not influenceed.

Brief description of the drawings

Fig. 1 is the workflow diagram of the present invention；

Fig. 2 is the spatial scene tree structure schematic diagram of the present invention；

Fig. 3 is that regarding for the present invention cuts body space nodes cutting schematic diagram.

Embodiment

Technical scheme is further described with reference to the accompanying drawings and examples.

As shown in figure 1, the magnanimity static object management method under the space coordinates of the present invention comprises the following steps that：

(1) longitude and latitude, bounding box processing are carried out to the static object of input, forms standard compliant static object data.

N number of static object point is sequentially input, and target point is standardized, major parameter includes longitude and latitude, height Degree, bounding box, visible level etc..Static object data structure expression after note standardization is PointStruct：

PointStruct_i={ Lng, Lat, Alt, BoxSize, MinLevel, MaxLevel, Ref }

Wherein, PointStruct_iI-th of target point is represented, Lng represents the longitude of the target, and Lat represents the target Latitude, Alt represent the height of the target, and BoxSize represents the bounding box size of the target, and MinLevel represents minimum visible layer Level, MaxLevel represent the visible level of highest, usually MinLvel >=0, and MaxLevel≤20, Ref represent that the target quotes meter Number, is initially 0.

(2) spatial scene tree is created, travels through magnanimity static object data list, according to position and bounding box by data in sky Between search in scene tree and specify node, add data in suitable node.Data are deployed to the section if node is present Point, otherwise create new node and data are deployed to the node.

First, it is divided into thing two spaces four directions body node from global range by the earth, obtains following two node regions：

The Eastern Hemisphere node：

SpacialNode_East=Level=0, and 0 °≤Lng≤180 °, -90 °≤Lat≤90 °, NodeSize= WorldRadius}

The Western Hemisphere node：

SpacialNode_East=Level=0, and -180 °≤Lng≤0 °, -90 °≤Lat≤90 °, NodeSize= WorldRadius}

Above-mentioned formula meaning representation is that the Eastern Hemisphere longitude range arrives 180 degree for 0 at the 0th layer, and latitude scope is -90 to 90 Degree, the Western Hemisphere longitude range are -180 to 0 degree, and latitude is spent to 90 degree for -90, and height of node is earth hemisphere WorldRadius, Typically it is taken as 6378137 meters.

With Level increase, space tetragonal body size of node is the 1/4 of last layer node, and node is still WorldRadius。

Then, i-th of target PointStruct in static data list is traveled through_i, according to its longitude and latitude scope and bounding box Determine which root node it should be deployed to.

Finally, search level number is equal to PointStruct downwards successively_i.MaxLevel node, will if node is present PointStruct_iThe node is deployed to, otherwise creates the node.

Establishment process is expressed as：

(a) assume that the 0th layer of the Eastern Hemisphere space tetragonal body range of nodes is 0 °≤Lng≤180 °, -90 °≤Lat≤90 °, NodeSize=WorldRadius.

(b) then the Eastern Hemisphere in the 1st split layer into four spaces four directions body node Node₁、Node₂、Node₃、Node₄, it is empty Between region be respectively：

Node₁=Level=1, and 0 °≤Lng≤90 °, 0 °≤Lat≤90 °, NodeSize=WorldRadius }

Node₂=Level=1, and 90 °≤Lng≤180 °, 0 °≤Lat≤90 °, NodeSize=WorldRadius }

Node₃=Level=1, and 0 °≤Lng≤90 °, -90 °≤Lat≤0 °, NodeSize=WorldRadius }

Node₄=Level=1, and 90 °≤Lng≤180 °, -90 °≤Lat≤0 °, NodeSize=WorldRadius }

(c) by that analogy, each node continues to be split into 4 nodes in next layer, as shown in Fig. 2 and node size all For earth radius, by the fission process, an incomplete spatial scene tree is finally constructed.

The equal representation space coordinate of definition space data structure GVector3={ X, Y, Z }, X, Y, Z, unit is rice.

Definition space node Node_iBounding box be VexBox_i={ P₁、P₂、P₃、P₄、P₅、P₆、P₇、P₈, P₁~P₈It is The data of GVector3 types.

The data structure in definition space face is Plane={ a, b, c, d }, then the four of range of observation section body ViewFrustum It is made up of six faces that type is Plane：

ViewFrustum={ P_Front, P_Back, P_Left, P_Right, P_Top, P_Bottom}；

Each point in VexBox is done whether regarding an internal judgement is cut, then and if only if meet following condition when, P_i Internal regarding cutting：

P_front.a*P_i.x+P_front.b*P_i.y+P_front.c*P_i.z+P_front.d≥0

P_back.a*P_i.x+P_back.b*P_i.y+P_back.c*P_i.z+P_back.d≥0

P_Left.a*P_i.x+P_Left.b*P_i.y+P_Left.c*P_i.z+P_Left.d≥0

P_Right.a*P_i.x+P_Right.b*P_i.y+P_Right.c*P_i.z+P_Right.d≥0

P_Top.a*P_i.x+P_Top.b*P_i.y+P_Top.c*P_i.z+P_Top.d≥0

P_Bottom.a*P_i.x+P_Bottom.b*P_i.y+P_Bottom.c*P_i.z+P_Bottom.d≥0

When have in VexBox a point regarding cut it is internal when, then it represents that the node is visible, owns to the intra-node administration Target data carries out plus an operation, otherwise carries out subtracting one operation to all target datas of intra-node administration, as shown in Figure 3.

Claims

1. a massive static object management method under a space coordinate system, is characterized in that: comprise the steps:

(1) Standardize the input static target points to form standard static target data;

(2) Create a spatial scene tree, traverse the static target data, add the data to the appropriate node, if the node exists, deploy the data to the node, otherwise create a new node and deploy the data to the node;

(3) According to the current observation parameters, the spatial scene tree is clipped and controlled, and the reference counting operation is performed on the data in the invisible nodes, and the reference counting operation is performed on the data in the visible nodes;

(4) Start the offline processing thread, use multi-threading technology to count and load and unload the data in the spatial scene tree, release the data with a reference count less than 0 in the scene, and load the data with a reference count greater than 0 in the scene.

2. the massive static target management method under the spatial coordinate system according to claim 1, is characterized in that: in described step (1), the static target data structure expression of standard is:

PointStruct _i = {Lng, Lat, Alt, BoxSize, MinLevel, MaxLevel, Ref}

Among them, PointStruct _i represents the i-th target point, Lng represents the longitude of the target, Lat represents the latitude of the target, Alt represents the height of the target, BoxSize represents the bounding box size of the target, MinLevel represents the lowest visible level, and MaxLevel represents the highest Visible level, Ref represents the target reference count.

3. the mass static object management method under the space coordinate system according to claim 2, is characterized in that: in described step (2), creating space scene tree comprises the step:

(1) Divide the earth into two spatial quadrilateral nodes in the east and west, and obtain the following two node areas:

Eastern Hemisphere Node:

SpacialNode _East ＝{Level＝0, 0°≤Lng≤180°, -90°≤Lat≤90°, NodeSize＝WorldRadius}

Western Hemisphere Node:

SpacialNode _East = {Level=0, -180°≤Lng≤0°, -90°≤Lat≤90°, NodeSize=WorldRadius};

(2) The first layer of the eastern hemisphere is split into four spatial quadrilateral nodes Node ₁ , Node ₂ , Node ₃ , and Node ₄ , and their spatial regions are respectively:

Node ₁ ＝{Level＝1, 0°≤Lng≤90°, 0°≤Lat≤90°, NodeSize＝WorldRadius}

Node ₂ ＝{Level＝1, 90°≤Lng≤180°, 0°≤Lat≤90°, NodeSize＝WorldRadius}

Node ₃ ＝{Level＝1, 0°≤Lng≤90°, -90°≤Lat≤0°, NodeSize＝WorldRadius}

Node ₄ ={Level=1, 90°≤Lng≤180°, -90°≤Lat≤0°, NodeSize=WorldRadius};

Western Hemisphere and so on;

(3) By analogy, each node continues to be split into 4 nodes in the next layer, and the size of the spatial quadrilateral node is 1/4 of the node in the previous layer. After the splitting process, the spatial scene tree is constructed.

4. The massive static object management method under the spatial coordinate system according to claim 3, characterized in that: in the step (2), the static object data is traversed, and the node to be deployed is determined according to the range of latitude and longitude and the size of the bounding box ;Search for the node whose layer number is equal to PointStruct _i .MaxLevel, if the node exists, deploy PointStruct _i to the node, otherwise create the node.

5. the massive static object management method under the spatial coordinate system according to claim 4, is characterized in that: in the described step (3), the spatial scene tree clipping method is as follows:

Define spatial data structure GVector3={X, Y, Z}, X, Y, Z all represent spatial coordinates;

Define the bounding box of the spatial node Node _i as VexBox _i = {P ₁ , P ₂ , P ₃ , P ₄ , P ₅ , P ₆ , P ₇ , P ₈ }, and P ₁ ~ P ₈ are data of type GVector3;

Define the data structure of the space plane as Plane={a,b,c,d}, then the tetrahedron ViewFrustum of the observation range is composed of six planes of type Plane:

ViewFrustum={P _Front , P _Back , P _Left , P _Right , P _Top , P _Bottom };

Make a judgment on whether each point in the VexBox is in the view frustum, then P _i is in the view frustum if and only if the following conditions are met:

P _front .a*P _i .x+P _front .b*P _i .y+P _front .c*P _i .z+P _front .d≥0

P _back .a*P _i .x+P _back .b*P _i .y+P _back .c*P _i .z+P _back .d≥0

P _Left .a*P _i .x+P _Left .b*P _i .y+P _Left .c*P _i .z+P _Left .d≥0

P _Right .a*P _i .x+P _Right .b*P _i .y+P _Right .c*P _i .z+P _Right .d≥0

P _Top .a*P _i .x+P _Top .b*P _i .y+P _Top .c*P _i .z+P _Top .d≥0

P _Bottom .a*P _i .x+P _Bottom .b*P _i .y+P _Bottom .c*P _i .z+P _Bottom .d≥0

When there is a point in the VexBox within the viewport, it means that the node is visible, and an operation is performed on all target data deployed in the node, otherwise, a subtraction operation is performed on all target data deployed in the node.