TWI245233B - Method for locating a position in a computer terrain model - Google Patents

Method for locating a position in a computer terrain model Download PDF

Info

Publication number
TWI245233B
TWI245233B TW90131245A TW90131245A TWI245233B TW I245233 B TWI245233 B TW I245233B TW 90131245 A TW90131245 A TW 90131245A TW 90131245 A TW90131245 A TW 90131245A TW I245233 B TWI245233 B TW I245233B
Authority
TW
Taiwan
Prior art keywords
point
model
interval
points
trench
Prior art date
Application number
TW90131245A
Other languages
Chinese (zh)
Inventor
Ming-Fang Weng
Shiue-Wu Wang
Jen-Duo Liou
Original Assignee
Inst Information Industry
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Inst Information Industry filed Critical Inst Information Industry
Priority to TW90131245A priority Critical patent/TWI245233B/en
Application granted granted Critical
Publication of TWI245233B publication Critical patent/TWI245233B/en

Links

Landscapes

  • Processing Or Creating Images (AREA)

Abstract

This invention relates to a method for searching a position in a computer terrain model. It is employed in computer animation for searching a certain point (such as the contact point between a vehicle wheel and the ground) and a position in the computer terrain model. According to trench-type block division, the input terrain data is divided into plural sections to form a computer terrain model. By means of this way, the computer terrain model data can be divided into ""blocks,"" where ""trench regions"" are formed between blocks, and ""trench intersections"" are formed at intersections of the trench regions. While applying a testing point to the computer terrain model, different searching models are applied upon determining which specific region among the ""blocks"", ""trench regions"", or ""trench intersections"".

Description

經濟部智慧財產局員工消費合作社印製Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs

------、丨丨訂---------綠I (請先閲讀背面之注意事項再填寫本頁) 1245233------ 、 丨 丨 Order --------- Green I (Please read the precautions on the back before filling this page) 1245233

:隨著3D顯像技術的發展,地形模型的應用範圍愈來 愈廣,無論是軍事訓練、飛行模擬、地理資訊查詢系統或 二外場景的電腦遊戲,即時且眞實的地形顯像已經成為不 可或缺的一部分。然而,在諸多應用中,地形顯像一直以 來有以下的幾個問題: _ 1 · 一般技術無法即時產生大規模資料的地形模型, 同時亦不能有效的解決不同精細度之地形模型相接時所產 生的幾何不連續問題(模型破裂的現象)。 2 ·現有技術不提供與視點相依之多層次精細度地形 模型的碰撞偵測功能。例如,行進於地表的車子就會出現 輪胎可能凹陷於地面或浮在半空的不合理現象。 因此解決上述之缺點是有需要的。 【本發明之概述】 本發明之主要目的係在快速找某一測試點在電腦地形 模型(包括多精細度之電腦地形模型,muhi_ resolutions)當中位於哪一三角形,因為只要知道測 1 _________ 4 本纸張尺度適用中國國家標準(CNS)ilS (21〇 x 297公釐) 1245233 A7 B7 五、發明說明(> ) 二占一(,::其\厘標及γ座標)在平面投影後會位於哪 通過測試點之垂直線與三角形平面相交之點 在地形模型上的對應點)即可算出,因點 座標即可求出。 發月之另主要目的係在快速找某一測試點之步騾 找出可能涵蓋測試點之兩個三角_,在直接計算 測試點p是料哪—三角形,如此可加快尋找之速度。 人為達成上述之目的’本發明依據溝渠式的區間分割, I々地开/貝料切割成若干個區間以成為電腦地形模 型’每一區間的大小成為(2n+1)x(2n+1)維度,而且區 間的邊界不相互重疊。使得電腦地形模型資料可分為『區 訂 間』’區間與區間之間形成『溝渠區域』,且溝渠區域與 溝渠區域交叉處形成『溝渠交叉區域』。 在進行某一測試點(譬如車子的輪胎與地面接觸的 點)在電腦地形模型當中日争,先尋找測試點位於下列哪一 『區間』,『溝渠區域』或是『溝渠交又區域』特定區域 中,而有不同尋找之方式。 若測試點是位於『溝渠區域』或是『溝渠交叉區域』 當中時,尋找方式較為簡單,若是位於『區間』時,本發 明需利用描述『區間』之方位導向四元樹(〇〇Q Orlentation-Oriented Quad tree )來進行多層次地形 模型碰撞偵測之方法,以加速尋找可能涵蓋測試點之兩個 三角形0 由於本發明構造新穎,能提供產業上利用,且確有增 進功效,故依法申請發明專利。 I_____5 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐 1245233 A7 五、發明說明(3) 【圖式簡單説明】 第1圖係一般常用的數値地形模型所架構成之四方形網 狀0 第2圖係依據溝渠式的區間分割後之電腦地形模型(在X γ平面投影)。 第3圖係顯示記載電腦地形模型資料之階層式相依圖 (HDG )資料結構80。 第3 a圖係顯示一 5 X 5維度區間中之模型點。 弟4圖係顯示一溝渠區域。 弟5圖係顯示i己載一區間模型點之方位導向四元樹 (OOQ ) 〇 第6圖係顯示一測試點在一區域之例子。 【圖號説明】 (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 電腦地形模型3 0 特徵點3 5 區間4 1 溝渠交叉區域4 3 模型點3 1 四方形網狀4 0 溝渠區域4 2 三角形5 0 階層式相依圖(HDG)資料結構80 方位導向四元樹(OOQ ) 85 【較佳具體實施例之詳細説明】 ^紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公爱一: With the development of 3D imaging technology, the application range of terrain models is becoming wider and wider. Whether it is military training, flight simulation, geographic information query system or computer games in the second scene, real-time and solid terrain imaging has become impossible. The missing part. However, in many applications, terrain imaging has always had the following problems: _ 1 · General technology cannot generate terrain models of large-scale data in real time, and it cannot effectively solve the problem when terrain models of different fineness are connected. The resulting geometric discontinuities (phenomena of the model breaking). 2 · The prior art does not provide a collision detection function of a multi-level fine-grained terrain model dependent on the viewpoint. For example, a car traveling on the ground will have the unreasonable phenomenon that its tires may sink in the ground or float in the air. Therefore, it is necessary to solve the above-mentioned disadvantages. [Overview of the present invention] The main purpose of the present invention is to quickly find out which triangle is located in a computer terrain model (including multi-fine computer terrain models, muhi_resolutions) at a certain test point, as long as you know how to measure 1 _________ 4 Paper size applies Chinese National Standard (CNS) ilS (21 × x 297 mm) 1245233 A7 B7 V. Description of the invention (>) Two occupy one (, :: its \ centimeter and γ coordinate) will be after plane projection It can be calculated from the corresponding point on the terrain model by the point where the vertical line of the test point intersects the triangular plane, and the point coordinates can be obtained. The other main purpose of Fayue is to quickly find a certain test point. Find two triangles that may cover the test point. Calculate directly where the test point p is expected to be a triangle. This can speed up the search. In order to achieve the above-mentioned purpose, the present invention 'the present invention is based on trench-type interval partitioning, and I cut it into several intervals to become a computer terrain model.' The size of each interval is (2n + 1) x (2n + 1) Dimensions, and the boundaries of the intervals do not overlap each other. The computer terrain model data can be divided into "zones" and the "ditch area" is formed between the intervals, and the "ditch crossing area" is formed at the intersection of the ditch area and the ditch area. When conducting a certain test point (such as the point where the car's tires contact the ground) in the computer terrain model, first find out which of the following "intervals", "ditch areas" or "ditch crossing areas" of the test points are specific Area, and there are different ways to find it. If the test point is located in the "ditch area" or "ditch crossing area", the search method is relatively simple. If it is located in the "interval", the present invention needs to use the orientation describing the "interval" to guide the quaternary tree (〇〇Q Orlentation) -Oriented Quad tree) method for multi-level terrain model collision detection to accelerate the search for two triangles that may cover the test points. 0 Because the invention has a novel structure, it can provide industrial use, and it does improve efficiency, so apply according to law Patent. I_____5 This paper size is in accordance with China National Standard (CNS) A4 (210 X 297 mm 1245233 A7) V. Description of the invention (3) [Simplified illustration of the drawing] Figure 1 is the fourth one of the commonly used mathematical terrain model. Square mesh 0 The second picture is a computer terrain model (projected on the X γ plane) after being divided according to the ditch-like interval. The third picture is a hierarchical dependency graph (HDG) data structure 80 that records the computer terrain model data. Figure 3a shows a model point in a 5 X 5 dimension interval. Figure 4 shows a ditch area. Figure 5 shows an azimuth-oriented quaternary tree (OOQ) containing model points in an interval. Figure 6 This is an example showing a test point in an area. [Illustration of the drawing number] (Please read the notes on the back before filling out this page) Computerized terrain model printed by the Employee Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs 3 0 Feature points 3 5 Interval 4 1 Ditch crossing area 4 3 Model points 3 1 Square mesh 4 0 Ditch area 4 2 Triangle 5 0 Hierarchical dependency graph (HDG) data structure 80 Orientation-oriented quaternary tree (OOQ) 85 [Details of the preferred embodiment Description] ^ Paper Of the applicable Chinese National Standard (CNS) A4 size (210 X 297 Kimiyoshi a

--------訂---------線I 1245233 A7 B7 五、發明說明(斗) 特 產生地形模型的原始資料是一般常用的數値地形模刑 2叫al Terrain Model,DTM)0 如第 i 圖所示,在 X土 千:來看,複數之模型點31可架構成四方形網狀4 母一挺型點31具有高度之『z』座標値。 而 、首先依據溝渠式的區間分割,將輸入的地形資 :::=0之各模型點31)切割成若干個區間以成為電 也形模型30 ’每—區間的大小成為(2η+ι)χ(2η+ι)維 度,η為自然數,而且區間的邊界不相互重疊,以一類似 溝渠的區域隔開著,如第2圖所示區間41的大小皆是5χ5 維度,第2圖顯示四個區間41 ;另區間41與區間4ι之間形 成溝渠區域42,且溝渠區域42與溝渠區域42交叉處形成 溝渠交又區域43。另外可將每—區間41繼續劃分四塊之 子區間411,而子區間可繼續發展其子區間,最精細可到 直到最底層之子區間為3維度。 另外電腦地形模型30更包括複數之特徵點35,該特 徵點35係依照觀看電腦地形模型3〇時視角及遠近之不同 而由複數之模型點3 1中選取出來,且複數之特徵點3 5在 區間41,溝渠區域42以及溝渠交叉區域43構成複數之三 角形5 0,亦即每一二角形5 〇都由三個特徵點3 $所形成。 譬如第2圖在右上方之區間41具有較多之特徵點”,因此 第2圖在右上方之區間41有較多之三角形切,亦代表著電 腦地形模型30在右上方之區間41所顯示之解析度較高。 I_______7 本紙張尺度適用中國國豕標準(CNS)A4規格(210 X 297公爱)-------- Order --------- line I 1245233 A7 B7 V. Description of the invention (Battle) The original data of the specially generated terrain model is a commonly used mathematical terrain model 2 called al Terrain Model, DTM) 0 As shown in Figure i, in X Tuqian: From the perspective of the model point 31, a plurality of model points 31 can be framed to form a quadrilateral mesh 4 female one-pointed point 31 having the height "z" coordinate 値. Firstly, according to the trench-type interval segmentation, the input terrain data ::: = 0 of each model point 31) is cut into several intervals to become an electric shape model 30 'the size of each interval becomes (2η + ι) χ (2η + ι) dimension, η is a natural number, and the boundaries of the intervals do not overlap each other, separated by a ditch-like area. As shown in Figure 2, the size of interval 41 is 5χ5, as shown in Figure 2 Four sections 41; another section 41 and a section 4m form a ditch area 42, and the intersection of the ditch area 42 and the ditch area 42 forms a ditch intersection area 43. In addition, each interval 41 can be further divided into four sub-intervals 411, and the sub-interval can continue to develop its sub-intervals, the finest can be until the bottom-most sub-interval is 3 dimensions. In addition, the computer terrain model 30 further includes a plurality of feature points 35. The feature points 35 are selected from the plurality of model points 31 according to the viewing angle and the distance between the computer terrain model 30 and the feature points 35 of the plurality. In the interval 41, the ditch area 42 and the ditch crossing area 43 constitute a plurality of triangles 50, that is, each of the two triangles 50 is formed by three characteristic points 3 $. For example, the second figure in the upper right section 41 has more feature points. "Therefore, the second figure in the upper right section 41 has more triangular cuts, which also represents the computer terrain model 30 in the upper right section 41. The resolution is high. I_______7 This paper size applies to China National Standard (CNS) A4 (210 X 297 public love)

I t Ϊ245233 經濟部智慧財產局員工消費合作社印製I t Ϊ245233 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs

1245233 五、發明說明(b ) 步騾A : 尋找測試點p位於下列哪一特定區域: 右疋位於溝渠交叉區域4 3進行步驟b ; 若是位於溝渠區域42進行步騾c ; 若是位於區間4 1進行步驟D ; 要找出測試點p位於下列哪一特定區域很簡單,因為 只要快速比較測試點piX座標及γ座標與電腦地形模型… 30 (在Χ-Υ平面投影) 步驟Β : 其中之一 由溝渠交叉區域4 3所構成之兩個三角形5 〇中尋找出 測試點Ρ位於哪一個三角形50中,因為溝渠交又區域43_ 定只有兩三角形5〇,因此測試點ρ—定位於這兩三角形5〇 角形5 0。 步驟C : 請參考第4圖,先由溝渠區域4 2兩側之特徵點3 5中找 經濟部智慧財產局員工消費合作社印製 出每侧與測試點p最接近之兩個特徵點3 5所組成之兩個三 角形50,再尋找出測試點p位於哪一個三角形50中。譬如 第4圖之溝渠區域4 2 —側之特徵點3 5為T 1,T 3,T 4, T5 ’ T9,而另一側之特徵點35為B1〜B9,則只要比較測 試點P之X座標,即可迅速找出特徵點T5,特徵點T9,特 徵點B7,特徵點b8,所以涵蓋測試點p之兩個三角形5〇a 50b即可被找出。 本紙張尺^規格⑵〇 X 297公爱) 1245233 A7 B7 五 、發明說明( 步驟D : Φ 在進行尋找可能涵蓋測試點p之兩個三角形5〇時需先 為區間41 (譬如以第&圖為例)建立一如第5圖之方位導 向四元樹(ooq)85,該方位導向四元樹(〇〇q)85 與階層式相依圖(HDG)資料結構8〇建立方式類似,方 位導向四元樹(00Q) 85建立之方式如下·· 第一層為區間4 1之中央模型點,即模型點丨2 ; 訂 第二層為區間4 1之中央模型點丨2之東、南、西、北 方向且位於該區間41邊緣之模型點,即模型點1〇,模型 點2,模型點14及模型點22,其中區間41之中央模型點 1 2有四個指標各指向第二層之四個模型點丨〇,模型點2, 模型點14及模型點22。 第二層為區間41之下之四個子區間411之中央模型 點,即模型點1 6,模型點6,模型點8及模型點1 8,其中 第二層四個模型點10,模型點2,模型點14及模型點22各 具有兩個指標各指向最相鄰斜角方向(東南,西北,西南 或東北方向)之第三層兩個模型點,譬如第二層模型點i 〇 指向第三層之模型點16以及模型點6 〇 第四層為區間4 1之下之四個子區間4 1 1之中央模型點 第一層般各發展出東、南、西、北方向且位於該子區間 4 1 1邊緣之模型點。譬如第三層模型點丨6發展出第四層模 型點1 5 ’模型點1 1,模型點1 7及模型點2 1,而且模型點 1 6指向這四個模型點0 本紙張尺度適用中國國豕標準(CNS)A4規格(210 X 297公髮 1245233 A7 五、發明說明(δ ) 由於第3 a圖顯示一 5 X 5維度區間4 1,因此方位導向 四元樹(〇〇 Q ) 8 5發展到此為止,而譬如一 9 X 9維度區 間則可繼續如上述之原則發展出第五層及第六層。因此方 位導向四元樹(00Q ) 85之原則可歸納如下: 『單數層為紀錄區間4 1以及子區間4 1 1之中央模型 點,而雙數層為紀錄區間4 1以及子區間4 1 1之中央 模型點發展出東、南、西、北方向且位於區間4丨或 子區間4 1 1邊緣之模型點3 1,且單數層每一模型點 31皆有四個指標各指向下一層發展出東、南、西、 北方向且位於该子區間4 1 1邊緣之模型點3 1,且雙 數層每一模型點3 1皆有兩個指標各指向下一層最相 鄰斜角方向之兩個模型點3 1』。 訂- 建好方位導向四元樹(00Q) 85後即可開始尋找涵 盍測試點p之兩個三角形50,以下請一併請參考第6圖, 第6圖係顯示一測試點p在一區域之例子,其偵測之方式稱 為『碰撞偵測』。 S tep 1 ; 我們從方位導向四元樹(00Q) 85的根模型點 開始搜先檢查模型點是否為特徵點35,如果不是 點35的話則立即返回’因為表示所有的子(讣丨⑷模J 點也不是特徵點35,亦即涵蓋測試點p之兩個三角二 ^ v2G V24&AVg V4 V24 ;接著比較測量點卩和^ 厘標値,依據p在Vl2東方的比較結果(p x>Vi2.x) I_______ 11 I紙張尺度_ fi^^(CNS)A4 祕(2ι“^1245233 V. Description of the invention (b) Step 骡 A: Find which of the following specific points is located at the test point p: Right 疋 is located at the intersection of the ditch 4 3 Go to step b; If it is located at the ditch area 42 go to step 骡 c; If it is located at interval 4 1 Proceed to step D; It is easy to find out which of the following test points p is located in a specific area, because you only need to quickly compare the test point piX and γ coordinates with the computer terrain model ... 30 (projected on the X-Υ plane) Step B: One of them The two triangles 50 formed by the trench crossing area 43 are found in which triangle 50 the test point P is located. Because the trench intersection area 43_ is determined to have only two triangles 50, the test point ρ is located in these two triangles. 50〇 角 50。 50. Step C: Please refer to Figure 4. First, find the characteristic points 3 5 on both sides of the trench area 4 2 and find the two characteristic points 3 5 on each side closest to the test point p. The two triangles 50 formed are used to find out which triangle 50 the test point p is in. For example, in the trench area 4 2 in FIG. 4, the characteristic points 35 on the side are T 1, T 3, T 4, T5 ′ T9, and the characteristic points 35 on the other side are B1 ~ B9. X coordinate, you can quickly find the feature point T5, feature point T9, feature point B7, feature point b8, so the two triangles 50a 50b covering the test point p can be found. This paper ruler ^ Specifications ⑵〇X 297 公 爱) 1245233 A7 B7 V. Description of the invention (Step D: Φ When searching for two triangles 50 which may cover the test point p, the interval 41 (for example, the first & Figure as an example) The azimuth-oriented quaternary tree (ooq) 85 as shown in Fig. 5 is established. The azimuth-oriented quaternary tree (OOq) 85 is similar to the hierarchical dependency graph (HDG) data structure 80. The way to guide the quaternary tree (00Q) 85 is as follows: · The first layer is the central model point of the interval 41, that is, the model point 丨 2; The second layer is the central model point of the interval 41, and the east and south of 2 The model points located in the west, north, and north of the section 41, that is, model point 10, model point 2, model point 14, and model point 22, of which the central model point 12 of section 41 has four indicators each pointing to the second Four model points of the layer 丨 0, model point 2, model point 14 and model point 22. The second layer is the central model point of the four sub-intervals 411 below the interval 41, that is, the model point 16, the model point 6, the model Point 8 and model point 18, of which there are four model points 10, model point 2, model point 14 and model point 2 in the second layer. 2 each has two indicators pointing to the third layer two model points in the most adjacent oblique direction (south-east, northwest, southwest, or northeast direction), such as the second layer model point i 〇 pointing to the third layer model point 16 and Model point 6 〇 The fourth layer is the central model point of the four sub-intervals 4 1 1 below the interval 41. The first layer generally develops models in the east, south, west, and north directions and is located on the edge of the sub-interval 4 1 1. For example, the third layer model point 丨 6 develops the fourth layer model point 1 5 'model point 1 1, model point 17 and model point 21, and model point 16 points to these four model points 0 paper scale Applicable to China National Standard (CNS) A4 specification (210 X 297, issued 1245233 A7 V. Description of the invention (δ) As Figure 3a shows a 5 X 5 dimensional interval 41, the orientation-oriented quaternary tree (〇〇Q ) 8 5 has developed so far, and for example, a 9 X 9 dimensional interval can continue to develop the fifth and sixth layers as described above. Therefore, the principle of orientation-oriented quaternary tree (00Q) 85 can be summarized as follows: " The singular layer is the central model point of the recording interval 4 1 and the sub-interval 4 1 1, and the even layer is Recording the central model points of interval 41 and sub-interval 4 1 1 develops model points 3 1 in the east, south, west, and north directions and located at the edge of interval 4 丨 or sub-interval 4 1 1 and each model point 31 in the singular layer There are four indicators each pointing to the next layer to develop model points 3 1 in the east, south, west, and north direction and located on the edge of the sub-interval 4 1 1 and each model point 3 1 in the even layer has two indicators each pointing The two model points 3 1 ″ in the direction of the nearest adjacent oblique angle on the next layer. Order-After building the azimuth-oriented quaternary tree (00Q) 85, you can start looking for the two triangles 50 that contain the test point p. Please refer to Figure 6 below. Figure 6 shows a test point p in one For example, the detection method is called "collision detection". S tep 1; We search from the root model point of the azimuth-oriented quaternary tree (00Q) 85 to check whether the model point is the feature point 35. If it is not the point 35, it immediately returns' because all the sub (讣 丨 ⑷ The J point is not the characteristic point 35, that is, the two triangles covering the test point p ^ v2G V24 & AVg V4 V24; Then, the measurement points 卩 and ^ centimeter 値 are compared, according to the comparison result of p in Vl2 East (p x > Vi2.x) I_______ 11 I paper size_ fi ^^ (CNS) A4 Secret (2ι "^

Sj e p 2 ; 比較測量點P和模型點V . ^ ^ 南方的比較v> v &値,依據^Vl4 作為下_、ΜΦ1 l4.y)挑選子(ehiId)模型點V, 為.a,θ上哥的起使點。重複Stepl的搜尋方法,而因 = ΐν,8的西方,所以挑選模型點、西方的子 C 1 模型點V丨7作為下一次的搜尋。Sj ep 2; Compare measurement point P and model point V. ^ ^ South comparison v > v & 値, select ^ Vl4 as the next _, MΦ1 l4.y) and select the sub-point (ehiId) model point V, as .a, Brother's starting point on θ. Repeat Stepl's search method, and since = ΐν, 8 is the west, the model point and the western child C 1 model point V 丨 7 are selected as the next search.

Step 3 : 田〜’J量點P與模型點V17比較時,我們除了知道p在 V,7的北方外,還清楚地知道V|7是方位導向四元樹 (00Q) 85的葉結點,所以模型點v丨7東北方的三角形八 VI2VI7V,8極可能涵蓋測量點p。另外,程式此時需返回 上一層的父(parent)模型點(Vis)進行另—方向的搜 尋,亦即比較測量點P和模型點¥18的丫座標値,依據p在 V1S北^的比較結果,挑選子(chUd)模型點作為下 -次搜尋的起使點,由於模型點Vi3不是特徵點35,因此 △ v12v:8v14極可能涵蓋測量點p。因此涵蓋測試點p之兩 個二角形被找出後,即進行計算測試點p是落於哪一三角 形。 、一 1245233 A7 B7 五、發明說明(⑺) 之前已處理過測量點和模型點V】8的X座標値比較, 現在還需比較測量點p和模型點^^8的γ座標値,。從比較 、、。果可以彳于知P在V丨8的北方(P · y v 18 · y),所以挑選模 型點V!8北方的子(child)模型點V"進行下一次的搜 尋。但是,由於Vls不是特徵點35,所以Vis北方的區域 就是由vu北方的三角形△义2%8%4所構成,因而測量 點P很可能就是由此三角形所涵蓋。 S^te p 5 : 經過上述的搜尋後,測量點p保證一定會落在△ Vi2Vi7v18 *AVi2Vi8vi4兩個三角形中的其中一個。 訂 ,上述之步騾之重點在於先預設蓋測量點P的兩個三角 开y 5 0之四個特徵點3 5為區間4丨最外圍之四個特徵點V 〇, V4 V24,V2〇,而若在方位導向四元樹(〇〇q) 85中尋 找到其他新的特徵點,則將新的特徵點取代原四個特徵胃占 <其中一個。上述之步騾當由單數層向下尋找時,尋㈣ 試點所位於『東西向”戈『南北向』之模型點”,若找到 有特徵點35,餐新找狀特徵點35取代距離測試%最 紅原預設之特徵點35,並則進入雙數層尋找,當『東西 向』或:南北向』都曾尋找過則停止尋找;另外當由雙數 層向下尋找時,尋找測試點所位於方向之模型點,若:到 Ϊ特徵點35/則將新找到之特徵點35取代距離測試點 =原預叹 < 特徵點3 5,並進入則進入單數層尋找力 有找到特徵點35,則回去上層之單數層。 右,又 L----- 13 本纸張尺^^家標準 1245233 A7 " _ 一~— p7_______ 五、發明說明(丨丨) ‘上所陳’本創作無論就目的、手段及功效,在在均 ’項不其迥異於習知技術之特徵,為一大突破,懇請貴審 厂文員明祭,早日賜准專利,俾嘉惠社會,實感德便。惟 肩/王意’上述實施例僅為例示性説明本發明之原理及其功 效,而非用於限制本發明之範圍。任何熟於此項技藝之人 士均可在不達背本發明之技術原理及精神下,對實施例作 修改與變化,譬如在建立方位導向四元樹(〇〇Q)。時 可於步驟D之前即進行,甚至在步驟A之前即將所有區間 之方位導向四元樹(00Q ) 85建立好。本發明之權利保 護範圍應如後述之申請專利範圍所述。 (請先閱讀背面之>t意事項再填寫本頁) 訂---------: 經濟部智慧財產局員工消費合作社印製 本纸張尺度適用中國國家標準(CNS)A4規格(210 X 297公爱)Step 3: When Tian ~ 'J is compared with model point V17, in addition to knowing that p is in the north of V, 7, we also know that V | 7 is the leaf node of the azimuth-oriented quaternary tree (00Q) 85. Therefore, the triangle point VI2VI7V, 8 in the northeast of the model point v 丨 7 most likely covers the measurement point p. In addition, the program now needs to return to the parent model point (Vis) of the previous layer to search in the other direction, that is, compare the measurement point P with the model coordinate ¥ 18 of the model point 値, according to the comparison of p at V1S North ^ As a result, the chUd model point is selected as the starting point for the next search. Since the model point Vi3 is not the feature point 35, Δv12v: 8v14 is likely to cover the measurement point p. Therefore, after the two polygons covering the test point p are found out, the triangle in which the test point p falls is calculated. 1. 1245233 A7 B7 V. Description of the invention (⑺) The X coordinate 値 of the measurement point and the model point V] 8 has been processed before. Now it is necessary to compare the measurement point p with the γ coordinate 模型 of the model point ^^ 8. From comparison,. If we know that P is in the north of V 丨 8 (P · y v 18 · y), we choose the child point V of the model point V! 8 north for the next search. However, because Vls is not the characteristic point 35, the area north of Vis is formed by the triangle Δ2 2% 8% 4 of vu north, so the measurement point P is likely to be covered by this triangle. S ^ te p 5: After the above search, the measurement point p is guaranteed to fall on one of the two triangles △ Vi2Vi7v18 * AVi2Vi8vi4. The main point of the above step is to preset the four characteristic points 35 of the two triangular openings y 5 0 of the measurement point P to be interval 4 丨 the four outermost characteristic points V 〇, V4 V24, and V2. If another new feature point is found in the azimuth-oriented quaternary tree (〇〇q) 85, the new feature point will replace the original four feature stomachs < one of them. The above steps: When looking down from the singular layer, find the model point in the "east-west" direction and "north-south direction" of the pilot. If a feature point 35 is found, the new feature point 35 replaces the distance test%. The reddest preset feature point 35 is entered in the even-numbered layer. When the "east-west" or "north-south" direction has been searched, the search stops. In addition, when you search downward from the even-numbered layer, the test point is located. For the model point in the direction, if you reach the Ϊfeature point 35 /, then the newly found feature point 35 will replace the distance test point = the original sigh < feature point 3 5 and enter the singular layer to find the feature point 35. Then go back to the upper singular. Right, and L ----- 13 paper rulers ^^ Home standard 1245233 A7 " _ 一 ~ — p7_______ V. Description of the invention (丨 丨) 'Shang Chen Chen' This work, regardless of the purpose, means and effect, The features in Zaijun are quite different from those of conventional technology. For a major breakthrough, I implore the clerk of your factory to give a sacrifice to the patent as soon as possible. However, the above-mentioned embodiments are merely illustrative of the principles and functions of the present invention, and are not intended to limit the scope of the present invention. Anyone skilled in the art can make modifications and changes to the embodiment without departing from the technical principles and spirit of the present invention, for example, in establishing an orientation-oriented quaternary tree (OOQ). The time can be performed before step D, and even before step A, the azimuth-oriented quaternary tree (00Q) 85 of all intervals is established. The scope of protection of the rights of the present invention shall be as described in the patent application scope mentioned later. (Please read the > t notice on the back before filling out this page) Order ---------: Printed by the Consumers' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs This paper is sized for the Chinese National Standard (CNS) A4 (210 X 297 public love)

Claims (1)

1245233 A8 § D8 六、申請專利範圍 1· 一種在電腦地形模型尋找位置之彳法,其中構成電腦 地形模型所需之資料包括複數之模型點,該複數之模 型點在平面Χ — Υ方向形成四方形網狀,而複數模型點 係由依據溝渠式的區間分割成若干個區間,每一區間 的大小成為(2η+1)Χ(2η+1)維度,η為自然數,而且區 間的邊界不相互重疊,區間之下亦可繼續發展四個子 區間,而子區間可繼續發展其子區間,直到最底層之 子區間為3維度;另區間與區間之間形成溝渠區域,且 溝渠區域與溝渠區域交又處形成溝渠交叉區域;另外 電腦地形模型更包括複數之特徵點,該特徵點係依照 觀看電腦地形模型時視角及遠近之之不同而由複數之 杈型點中選取出來,且複數之特徵點在區間,溝渠區 域以及溝渠交叉區域構成複數之三角形,該方法係尋 找某一測試點在電腦地形模型當中位於哪一三角形, 琢方法先進行尋找測試點為於下列哪一特定區域:溝 木X又區域,溝渠區域或是區間,而依照不同之尋找 方式進行尋找· 經濟部智慧財產局員工消費合作社印製 2 ·如申请專利範圍第1項所述之在電腦地形模型尋找位置 之方法,其中: 若是位於溝渠交叉區域進行步驟A ; 若是位於溝渠區域進行步騾B ; 若是位於區間進行步驟C ; 步勝A :由溝渠交叉區域所構成之兩個三角形中尋找出 • 測試點位於哪一個三角形中; 15 1245233 六、申請專利範圍 步驟B ·先由溝渠區域兩側之特徵點中找出每側與測試 點最接近之兩個特徵點所組成之兩個三角 . 形,再尋找出測試點位於哪一個三角形中; 步騾C : 步驟C 1 :為該區間建立一方位導向四元樹 (0 0 Q ) ’將構成該區間之模型點資料建立 於該方位導向四元樹(〇〇Q)中,該方位導 向四元樹(00Q )建立之原則如下: 第一層為區間之中央模型點,可繼續往下建立 树狀結構,直到最底層之子區間為3維度 為止’使得單數層為紀錄區間以及子區 間之中央模型點’而雙數層為紀錄區間 以及子區間之中央模型點發展出東、 南、西、北方向且位於區間或子區間邊 緣之模型點,且單數層每一模型點皆有 四個才曰標各指向下一層發展出東、南、 西、北方向且位於該子區間邊緣之模型 點,且雙數層每一模型點皆有兩個指標 各指向下一層最相鄰斜角方向之兩個模 型點; 步驟C2 :尋找可能涵蓋測試點之兩個三角形之四 個特徵點,其尋找之規則如下·· 預設 < 四個特徵點為形成區間最外圍四個模 型點; 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297^7" 項 頁 線 1245233 六、申請專利範圍 由万位導向四元樹(00Q)第一層向下尋找 取代預設之特徵點;1245233 A8 § D8 6. Scope of Patent Application 1. A method for finding a location on a computer terrain model. The information required to form a computer terrain model includes a plurality of model points. The model points of the plurality form four in the direction of the plane X-Υ. It is a square mesh, and the complex model points are divided into several intervals based on trench-type intervals. The size of each interval becomes the (2η + 1) × (2η + 1) dimension, η is a natural number, and the boundary of the interval is not Overlapping each other, four sub-intervals can continue to be developed below the interval, and the sub-interval can continue to develop its sub-intervals until the bottom-most sub-interval is 3 dimensions; another interval and interval form a trench area, and the trench area and the trench area intersect It also forms a trench crossing area; in addition, the computer terrain model further includes a plurality of feature points, which are selected from a plurality of branch points according to the difference in perspective and distance when viewing the computer terrain model, and the plurality of feature points In the interval, the trench area and the trench crossing area form a complex triangle. This method is to find a test point in the computer terrain model. Which triangle is located in it? Find out which of the following specific areas is the test point: trench wood X area, trench area or interval, and search according to different search methods. • Intellectual Property Bureau, Ministry of Economic Affairs, Consumer Consumption Cooperative Print 2 · The method for finding a location on a computer terrain model as described in item 1 of the scope of the patent application, wherein: if it is located in the cross area of the ditch, perform step A; if it is located in the ditch area, perform step B; if it is located in the interval, perform step C; Backstep A: Find out the two triangles formed by the intersection area of the trench • Which triangle is the test point located in? 15 1245233 VI. Patent Application Step B · First, find each side from the feature points on both sides of the trench area Two triangles formed by the two feature points closest to the test point, and then find out which triangle the test point is in; Step 骡 C: Step C 1: Create an azimuth-oriented quaternary tree (0 0 Q) 'The model point data constituting the interval is established in the azimuth-oriented quaternary tree (〇〇Q), and the azimuth-oriented quaternary (00Q) The principle of establishment is as follows: The first layer is the central model point of the interval, and the tree structure can continue to be established until the bottommost sub-interval is 3 dimensions' so that the singular layer is the central model point of the recording interval and the sub-interval. 'The double layer is the central model point of the record interval and sub-interval. It develops model points in the east, south, west, and north direction and located at the edge of the interval or sub-interval. Each model point in the singular layer has four labels. Point to the next layer to develop the model points in the east, south, west, and north direction and be located at the edge of the subinterval, and each model point in the bilevel layer has two indicators that each point to the two model points in the direction of the next nearest oblique angle Step C2: Find the four characteristic points of the two triangles that may cover the test points. The rules for the search are as follows: Preset < The four characteristic points are the four model points that form the outermost part of the interval; This paper scale applies to China Standard (CNS) A4 specification (210 X 297 ^ 7 " item page line 1245233) 6. The scope of patent application is from 10,000-bit to quaternary tree (00Q). The generation of a predetermined feature point; 由單,層向下尋找時,尋找測試點所位於 『東西向』或『南北向』之模型點,若 找到有特徵點,則將新找到之特徵點取 代距離測試點最遠之原預設之特徵點, ,則進入雙數層尋找,當『東西向』或 『南北向』都曾尋找過則停止尋找. 當由雙數層向下尋找時,尋找測試點所位於 万向<模型點,若找到有特徵點,則將 =到讀徵點取代距離測試點最遠之 =設之特徵點’並進人則進 :二層若州到特徵點,則回去上; 線 本紙張尺度適用中國國家標準(CNS)A4規格(21〇 X 297 17When searching from the single layer to the lower layer, look for the model point where the test point is located in the "east-west" or "north-south direction". If a feature point is found, the newly found feature point will replace the original preset farthest from the test point For feature points,, enter the double layer search, and stop searching when both "east-west" and "north-south" have been searched. When searching downward from the double layer, find the test point located in the universal < model point, If a characteristic point is found, then replace the farthest distance from the test point to the reading sign point = set the characteristic point 'and then enter it: if the second layer of the state reaches the characteristic point, go back up; the paper size of the paper is applicable to the Chinese country Standard (CNS) A4 specification (21〇X 297 17
TW90131245A 2001-12-17 2001-12-17 Method for locating a position in a computer terrain model TWI245233B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW90131245A TWI245233B (en) 2001-12-17 2001-12-17 Method for locating a position in a computer terrain model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW90131245A TWI245233B (en) 2001-12-17 2001-12-17 Method for locating a position in a computer terrain model

Publications (1)

Publication Number Publication Date
TWI245233B true TWI245233B (en) 2005-12-11

Family

ID=37190049

Family Applications (1)

Application Number Title Priority Date Filing Date
TW90131245A TWI245233B (en) 2001-12-17 2001-12-17 Method for locating a position in a computer terrain model

Country Status (1)

Country Link
TW (1) TWI245233B (en)

Similar Documents

Publication Publication Date Title
CN103455480B (en) A kind of position search method based on electronic chart and device
US7822541B2 (en) Map creation device and navigation device
EP1855263A1 (en) Map display device and map display method
EP2320382A1 (en) Bird's-eye image forming device, bird's-eye image forming method, and bird's-eye image forming program
US6449557B2 (en) Device and method for changing map information
RU2298227C2 (en) Method for displaying three-dimensional polygon on screen
KR20130040773A (en) Three-dimensional map drawing system
CN102183261A (en) Method for re-using photorealistic 3D landmarks for nonphotorealistic 3D maps
JP2006349872A (en) Apparatus for creating map information, method of creating map information and program
JPWO2005098792A1 (en) MAP INFORMATION GENERATION DEVICE, MAP INFORMATION GENERATION METHOD, AND MAP INFORMATION GENERATION PROGRAM
KR100634536B1 (en) 3 dimensional transformation method for 2 dimensional overpass data on 3 dimensional graphics environment and apparatus therefor, and 3 dimensional visualization method for 2 dimensional overpass data on 3 dimensional graphics environment and apparatus using the same
Laycock et al. Automatically generating large urban environments based on the footprint data of buildings
JP4619504B2 (en) 3D digital map generator
JP2859004B2 (en) Graphic processing unit
CN101477562B (en) Multi-resolution self-adapting marking method in 3DGIS
JP2001307121A (en) Map information correction device and its method
CN112579715A (en) Data processing method and device, electronic equipment and storage medium
CN112017262B (en) Pavement marker generation method and device, storage medium and electronic equipment
TWI245233B (en) Method for locating a position in a computer terrain model
CN111488411B (en) Road facility construction method and device, rendering method, medium and terminal
JP2006215378A (en) Map data display device, map data display method, navigation device and map data display program
Murata 3D-GIS application for urban planning based on 3D city model
JP2009020906A (en) Map display device, method for specifying position on map, and computer program
JP2007293597A (en) Analysis device, retrieval device and measurement device, and program
KR102635566B1 (en) Apparatus for generating map of navigation system and method thereof

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees