512283 經濟部智慧財產局員工消費合作社印製 A7 B7 五、發明說明（1 ) 相關專利申請之交叉參考 此專利申請之内容係關於以流水號AT9-98-492來辨識之 共同未決之專利申請，名稱爲"System and Method for Displaying a Three-Dimensional Object Using Motion Vectors to Generate Object Blur··，並與此專利申請同時歸檔；且係關 於以流水號ΑΤ9_99·059來辨識之共同未決之專利申請，名 稱爲"Image Processing to Achieve Motion Blur”，並與此專利 申請同時歸檔。前述之二專利申請是指配給此專利申請之 相同權利人，且在此提及該等專利申請以供參考。 背景 1. 發明領域 本發明概言之係關於電腦圖形之領域，且更明確地説係 關於以一種高效率方式來達成移動模糊及其他與時間有關 之效果之技術。 2. 相關技術歷史 三維（3D)圖形系統可用於多種應用，其中包含電腦輔助 繪圖，建築設計，飛機與其他交通工具之模擬訓練員，分 子模型化，虛擬實境應用，與電動遊戲。三維系統也常建 構於工作站與個人電腦，其中可包含或可不包含3D圖形硬 體。在包含3D圖形硬體之系統中，一圖形加速卡一般可方 便圖形影像之產生與顯示。 一軟體應用程式產生一 3D圖形景物，且提供該景物以及 明屬性至一應用程式設計介面（API)。目前之API包含 OpenGL，PHIGS，與Direct3D。一 3D圖形景物包含一些多邊 -4- 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公爱） --!----ϊ--------------訂--------—線 (請先閱讀背面之注意事項再填寫本頁) 512283 A7512283 Printed by A7 B7, Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the Invention (1) Cross-Reference to Related Patent Applications The content of this patent application is a co-pending patent application identified by serial number AT9-98-492, The name is " System and Method for Displaying a Three-Dimensional Object Using Motion Vectors to Generate Object Blur ... and is filed with this patent application at the same time; and it is a co-pending patent application identified by the serial number ΑΤ9_99 · 059, The name is "Image Processing to Achieve Motion Blur" and is filed with this patent application. The aforementioned two patent applications are assigned to the same rights holders of this patent application, and these patent applications are mentioned here for reference. Background 1. Field of the Invention The outline of the present invention relates to the field of computer graphics, and more specifically to a technique for achieving motion blur and other time-related effects in an efficient manner. 2. Related Technology History Three-dimensional (3D ) Graphics systems can be used in a variety of applications, including computer-aided drawing, architectural design, flying Simulation trainers for aircraft and other vehicles, molecular modeling, virtual reality applications, and video games. Three-dimensional systems are also often built on workstations and personal computers, which may or may not include 3D graphics hardware. Including 3D graphics hardware In a system, a graphics accelerator card can generally facilitate the generation and display of graphic images. A software application generates a 3D graphic scene, and provides the scene and its attributes to an application programming interface (API). The current API includes OpenGL, PHIGS, and Direct3D. A 3D graphic scene contains some multilaterals. -4- This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 public love).-! ----! ------ -------- Order ---------- Line (Please read the precautions on the back before filling this page) 512283 A7
經濟部智慧財產局員工消費合作社印梦 五、發明說明（2 ) 形，且該等多邊形是藉由/些頂點集合來界定。該等頂點 受到結合以形成更大之基元，例如三角形或其他多邊形。 該等三角形（或多邊形）受到結合以形成一些面，且該等面 受到結合以形成一物件。每/頂點關聯於一屬性集合，其 中一般包含：1)材質顏色，以描述該頂點所屬之物件之顏 色；2) —正交向量，以描述該面在該頂點朝向之方向；與 3) 一位置，其中包含三直角座標X，y，與Z。每一頂點可 隨意關聯於紋理座標及/或一阿爾法（亦即透明度）値。除 此之外，景物一般具有一屬性集合，其中包含：1) 一周園 顏色，一般是用以描述周園光之數量；與2)—或更多之個 別光源。每一光源具有一些關聯於其之特性，其中包含一 方向，一周圍顏色，一擴散顏色，與一反射顏色。 圖形系統運用顯像來產生3D圖形景物之二維影像投影以 顯示於一顯示器或其他顯示裝置。一般而言，顯像包含藉 由依所需執行下列運作之一或更多來處理幾何基元（例如 點’直線，與多邊形）：轉換，截割，選擇，照明，模糊 計算’與紋理座標產生。顯像進一步包含處理基元以決定 顯示裝置之個別像素値，且此程序通常特別稱爲光域化。 在一些3D應用中，例如，電腦動畫與模擬程式，3D圖形 景物之物件可在移動。在該等情形之下，最好模擬在移動 t物件之移動模糊。如果無移動模糊，則在移動之物件可 能顯得是以抖動方式通過螢幕。 爾杈备疋％景I深度時，類似之技術也常用以模糊物件。 位於，|視野"内之物件維持未^到模糊，而較近或較遠之物 本紙張尺度顧中關家辟（C^S)A4祕-----_____ (請先閱讀背面之注意事項再填寫本頁) ---------^---------, 512283 A7 B7 經濟部智慧財產局員工消費合作社印製 巧氏張尺度家鮮 五、發明說明（3 件則根據他們與照相機（亦即觀看者）之距離來受到模糊。 一用以模擬物件模糊之以前技術方法包含使用一累積緩 衝區。累積緩衝區是一用以累積一系列之影像，當該等影 像受到顯像時，之非顯示缓衝區。一整個景物（亦即該景 物之每一物件，或基元）在一系列之時片中重複傳送至累 積緩衝區。該整個景物因此累積於累積緩衝區，且接著拷 貝至一畫面缓衝區以供在一顯示裝置上觀看。 一使用一累積緩衝區來模擬物件模糊之以前技術方法展 示於圖1。如圖1所示，一時間區間分割成爲"n"時片（步 驟100)。時間區間是一景物可在一顯示裝置上看見之時間 里，且類似於一視訊照相機快門之曝光間隔，或快門速 度。一較長之快門速度對應於較多數量之模糊，而一較短 (快門速度對應於較少數量之模糊。一時片計數是設定成 爲-(步驟⑽）。接著，選擇_物件（亦即基元）以進行顯像 (步驟綱）。針對此特定時片來計算物件之每一頂點之位 置，顏色，與所有其他每_頂點値（步驟1〇6)。該物件接著 傳送至-顏色緩衝區（步驟⑽）。檢查以決定是否受到顯像 (物件是該景物之最後一物件(步驟n〇)。如果答案是否定 的，則程序回到步驟104，且針對該景物之每一物件來重 複。 如果該景物之最後一你丛1 t 物件已雙到顯像（亦即步驟11〇之問 = ^fyes")’則該景物受到累積（步驟112)，而此意 :、又“§放(例如鈿小成爲1/n)且拷貝至累積緩衝區。時 計數受到檢查以查看是U等於η(步㈣4)。如果答案 11.---f- - - ------------訂----------線 (請先閱讀背面之注音？事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 A7 五、發明說明（4 ) 疋否定的，則時片計數受到增加（步驟116)。程序接著回到 步驟104，且針對每一時片來重複。如果時片計數等於 ?亦即步驟m之問題之答案是”yes”），則累積緩衝區受到 、、猫放，且拷貝至畫面緩衝區（步驟120)及顯示於一顯示螢幕 (步驟122)。 … <因屙整個景物(亦即景物之每-物件)在每-時間區間皆 受到顯像"n"次，所以如圖1所述使用-累積緩衝區是一在 計算上很昂貴之程序。因此，最好具有一系統與方法以更 有效率地模擬三維圖形環境之物件模糊。 發明概要 本發明藉由圖形顯示方法，裝置與系統來解決前述之問 題，其中頂點或其他基元最好是藉由使用移動向量來指定 ^需要模糊。在本發明之第—實例中，對於需要模糊之 母-基兀，-顯示該基元在一指定時間間隔之時變參數變 化的板糊幾何丈到產生，且該模糊幾何接著附加至該基元 j一對應邊緣。該圖形基元與附加之模糊幾何接著儲存於 -晝面緩衝區’以顯示於一圖形系統。在較佳實例中，需 要棱糊（基元之指定與隨後辨識是藉由關聯至少-移動向 量與要模糊之基元之每—頂點來達成，其中移動向量之方 向與大小顯示關聯之頂點之方向與位移。在重要之線性移 =形中，模糊幾何是藉由圖形基元之邊緣與一些直線來 滅：四邊形’其中该等直線是藉由邊緣頂點之移動向量 來定義。最好，模糊幾何之產生包含利用對應之圖形基元 i屬性來推導模糊幾何屬性。最好’模糊幾何鄰接對應圖 —j----1--------------^---------^ (請先閱讀背面之注意事項再填寫本頁) 512283 經濟部智慧財產局員工消費合作社印製 本紐尺度適用中國國家標準 A7 B7 五、發明說明（5 形基元邊緣之區之屬性値實質上等效於圖形基元邊緣之屬 性値。_顯示時變參數之變化速率之漸隱因數是用以達成 該等模糊幾何屬性之漸隱，以致遠離對應邊緣之模糊幾何 邵份之屬性値小於鄰近對應邊緣之模糊幾何部份之屬性 値。模糊幾何可進一步包含反射影像，且該等反射影像是 由圖形基元之二或更多反射影像之一組合所組成。在—實 例中，只有當一邊緣受到辨識及決定成爲在整個指定時= 間隔中皆爲可見之基元之一落後邊緣時，模糊幾何方附加 於該邊緣。 本發明進一步構思用以顯示電腦圖形之第二方法，襞 置，與系統，其中一移動向量關聯於一圖形基元之每—頂 點’其中，如前所述’該移動向量顯示關聯之頂點之—時 又參數的變化，且一漸隱因素關聯於該圖形基元。圖形基 元頂點之光域資料接著針對圖形基元所定義之一像素集： 來進行内插，以產生一光域資料集合，其中該光域資料二 合（每-構件關聯於該像素集合之—對應構件。圖形基元 頂點之移動向量資料也針對相同之像素集合來進行内插以 產生一對應之像素向量集人。 木口 接耆利用漸隱因數，光域資 料术與像素向量集合，夢 " 人拉、 猎由根據漸隱因數沿對應像素 向1所顯示之一方向内插光域 ^ 合。在移動模糊之情形中，_由^來產生一和動直線集 本. 猎由挺糊來模擬之時變參數包 含圖形基凡心位置。在一實 其〜、 、 進仃内插之像素集合是 集合包含圖形基元定,、=像素在另一實例中，像素 _ 我又所有像素。較佳實例是藉由下列 iirl!r------ΦΜ------——訂--------άφ— (請先閱讀背面之注音？事項再填寫本頁) 512283 A7 五、發明說明（6 ) 方式來解決内破移動與爆炸移動之特殊情形·在内破移動 之情形中，以移動三角形來置換移動_，且在爆炸料 之情形中，以強度値除以重疊移動直線之數目。 附圖簡短詔 -旦閱讀下列詳細説明並參看附圖，本發明之其他目標 與優點將變得顯而易見，其中： $ 圖1是根據以前技術來達成移動模糊之一方法之流程圖· 圖2是根據本發明之一實例之一動態圖形基元與其之附 加模糊幾何的圖形； 圖3是根據本發明之—實例之_w形顯示方法的流程圖； 圖4是根據本發明之-實例之—圖形顯示方法的流程圖； 圖5是根據本發明及涵蓋非線性移動之_移動模糊技術 之圖形； 圖6a與6b展示本發明之其他實例，其中模糊幾何選擇性 或全面性附加至一動態圖形基元之邊緣； 圖7a與7b展示本發明之一實例，該實例包含物件之高效 率時間模糊，且該等物件包含紋理對映之反射影像； 圖7c是反射於圖7&與几之物件之影像的圖形； 圖8展·;π —圖形基元與其之附加模糊幾何，其中包含圖 7a與7b之反射影像之一複合影像； 固疋根據本發明之第二實例及使用移動直線來達成 移動模糊之流程圖； (請先閱讀背面之注意事項再填寫本頁) -------t--------- 經濟部智慧財產局員工消費合作社印製 圖10是圖9之模糊技術之一圖形 之每一像素關聯於一移動直線； 其中一基元之一邊緣Yin Meng, an employee consumer cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. The description of the invention (2), and the polygons are defined by the set of vertices / sets. The vertices are combined to form larger primitives, such as triangles or other polygons. The triangles (or polygons) are combined to form faces, and the faces are combined to form an object. Each / vertex is associated with a set of attributes, which generally include: 1) the color of the material to describe the color of the object to which the vertex belongs; 2)-an orthogonal vector to describe the direction in which the face faces the vertex; and 3) a Position, which contains three right-angled coordinates X, y, and Z. Each vertex can be associated with texture coordinates and / or an alpha (ie transparency) 値. In addition, the scenery generally has a set of attributes, which includes: 1) the color of the weekly garden, which is generally used to describe the amount of light in the garden; and 2) —or more individual light sources. Each light source has some characteristics associated with it, including a direction, a surrounding color, a diffuse color, and a reflection color. The graphics system uses display to generate a two-dimensional image projection of a 3D graphic scene for display on a display or other display device. In general, visualization involves processing geometric primitives (such as points' lines, and polygons) by performing one or more of the following operations as needed: transformations, cropping, selection, lighting, fuzzy calculations, and texture coordinate generation. . Imaging further includes processing primitives to determine individual pixels of the display device, and this process is often referred to as light zoning. In some 3D applications, for example, computer animation and simulation programs, 3D graphics objects can be moved. In these cases, it is best to simulate the motion blur of a moving object. If there is no motion blur, moving objects may appear to shake the screen. Similar techniques are often used to obscure objects when preparing for% scene I depth. The objects located in the | field of view remain undisturbed, and the objects nearer or farther away from the paper scale Gu Zhongguan Jiapi (C ^ S) A4 secret -----_____ (Please read the back Please fill in this page again for the matters needing attention) --------- ^ ---------, 512283 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs (Three pieces are blurred based on their distance from the camera (i.e. the viewer). A prior art method to simulate object blurring included the use of an accumulation buffer. The accumulation buffer is used to accumulate a series of images, When these images are developed, the non-display buffer. An entire scene (ie, each object or primitive of the scene) is repeatedly transmitted to the accumulation buffer in a series of time slices. The entire scene Therefore, it is accumulated in the accumulation buffer and then copied to a picture buffer for viewing on a display device. A prior art method of using an accumulation buffer to simulate object blurring is shown in Fig. 1. As shown in Fig. 1, A time interval is divided into " n " time slices (step 100). The interval is the time during which a scene can be seen on a display device, and is similar to the exposure interval, or shutter speed, of a video camera shutter. A longer shutter speed corresponds to a greater amount of blur, and a shorter ( The shutter speed corresponds to a smaller number of blurs. A tick count is set to-(step ⑽). Then, select _object (ie primitives) for development (step outline). Calculate the object for this specific time slice The position, color, and all other vertices of each vertex 1 (step 106). The object is then transferred to the -color buffer (step ⑽). Check to determine whether it is imaged (the object is the object of the scene The last object (step n0). If the answer is no, the process returns to step 104 and is repeated for each object of the scene. If the last object of the scene is 1 t the object has been doubled to the image ( That is, the question of step 11〇 = ^ fyes ") 'then the scene is accumulated (step 112), and the meaning is: "§ put (for example, 钿 小 becomes 1 / n) and copied to the accumulation buffer. Hour count Checked to see if U In η (step 4). If the answer 11. --- f--------------- order ---------- line (Please read the phonetic on the back? Please fill in this page again) Printed A7 by the Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (4) 疋 If not, then the time slice count is increased (step 116). The program then returns to step 104, and for each time If the time slice count is equal to? That is, the answer to the question of step m is "yes"), the accumulation buffer is received, and placed, and copied to the picture buffer (step 120) and displayed on a display screen. (Step 122) ... < Because the entire scene (i.e., each-object of the scene) has been developed " n " times in every-time interval, so the use-accumulation buffer is described in FIG. Computationally expensive program. Therefore, it is better to have a system and method to more effectively simulate the blurring of objects in a three-dimensional graphics environment. SUMMARY OF THE INVENTION The present invention solves the foregoing problems by means of a graphic display method, device, and system, in which vertices or other primitives are preferably specified by using motion vectors. In the first example of the present invention, for the mother-base unit that needs to be fuzzy, the baffle geometry that shows the time-varying parameter change of the primitive at a specified time interval is generated, and the fuzzy geometry is then appended to the base. Element j corresponds to the edge. The graphics primitives and the additional fuzzy geometry are then stored in a -day-plane buffer 'for display in a graphics system. In the preferred example, ambiguity is required (the designation and subsequent identification of the primitives is achieved by associating at least-the movement vector with each-vertex of the primitive to be blurred, where the direction and size of the movement vector show the associated vertices Direction and displacement. In the important linear shift = shape, fuzzy geometry is eliminated by the edges of the graphics primitives and some straight lines: quadrilaterals, where the straight lines are defined by the movement vectors of the edge vertices. Best, fuzzy The generation of geometry includes the use of the corresponding graphic primitive i attribute to derive fuzzy geometric attributes. It is best to 'fuzzy geometric adjacent correspondence graphs—j ---- 1 -------------- ^- ------- ^ (Please read the notes on the back before filling this page) 512283 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs This standard is applicable to the Chinese national standard A7 B7 V. Description of the invention (5-shaped primitive The attributes of the edge regions are essentially equivalent to the attributes of the edges of the graphics primitives. _ The fade-out factor showing the rate of change of time-varying parameters is used to achieve the fade-out of these fuzzy geometric attributes, so as to stay away from the blur of the corresponding edges The Properties of Geometry Attributes smaller than the part of the fuzzy geometry adjacent to the corresponding edge. The fuzzy geometry may further include reflection images, and these reflection images are composed of a combination of two or more reflection images of graphics primitives. In the example, only When an edge is identified and determined to be a trailing edge that is one of the primitives that are visible throughout the interval, a fuzzy geometric square is added to the edge. The present invention further contemplates a second method for displaying computer graphics, 襞And the system, where a motion vector is associated with each of the vertices of a graphics primitive, where, as mentioned previously, the motion vector shows the associated vertices and the parameters change, and a fading factor is associated with the Graphic primitives. The light domain data of the vertices of the graphics primitives are then interpolated for a set of pixels defined by the graphics primitives: to interpolate to generate a light domain data set, where the light domain data is two-in-one (each-component is associated with The corresponding component of the pixel set. The movement vector data of the vertices of the graphics primitives are also interpolated for the same pixel set to generate a corresponding pixel direction. Mukou uses the fade factor, light field data collection and pixel vector collection, and the "people pull and hunt" interpolate the light field in one of the directions shown by the corresponding pixel to 1 according to the fade factor. In the case of motion blur, _ is used to generate a set of moving straight lines. The time-varying parameters simulated by quite paste include the position of the basic center of the graphic. The set of pixels interpolated by ~,, and 是 is The set contains graphics primitives, and = pixels. In another example, pixels_I have all pixels. A better example is by the following iirl! R ------ ΦΜ ------—— order- ------- άφ— (Please read the note on the back? Matters before filling out this page) 512283 A7 V. Description of the invention (6) Ways to solve the special situations of internal breaking movement and explosive movement In the case, the movement triangle is replaced by the movement triangle, and in the case of explosive materials, the intensity 以 is divided by the number of overlapping movement lines. Brief description of the drawings-Once reading the following detailed description and referring to the accompanying drawings, other objects and advantages of the present invention will become apparent, among which: Figure 1 is a flowchart of one method for achieving motion blur according to the prior art. Figure 2 is According to an example of the present invention, a dynamic graphics primitive and its additional fuzzy geometric figures are shown; FIG. 3 is a flowchart of the _w-shaped display method of the example of the present invention; FIG. 4 is an example of the present invention. Flow chart of a graphic display method; Figure 5 is a graphic according to the present invention and a _moving blur technique covering non-linear movement; Figures 6a and 6b show other examples of the present invention, where the fuzzy geometry is selectively or comprehensively added to a dynamic graphic Edges of primitives; Figures 7a and 7b show an example of the present invention, which includes high-efficiency time blurring of objects, and these objects include reflected images of texture mapping; Figure 7c is the object reflected on Figures 7 and 7 Figure 8 shows the image; π — the graphics primitive and its additional fuzzy geometry, which contains a composite image of one of the reflection images of Figures 7a and 7b; The second example and the flowchart of using moving straight lines to achieve moving blur; (Please read the notes on the back before filling this page) ------- t --------- Intellectual Property Bureau of the Ministry of Economic Affairs Printed by Employee Consumer Cooperative Figure 10 is one of the blurring techniques of Figure 9. Each pixel of the graphic is associated with a moving line; one of the edges of one of the primitives
A7 B7 五 、發明說明（ 圖11疋圖10之模糊技術之一圖形，其中基元之每一像素 關聯於一移動直線； 圖12疋用於圖9方法之内破移動情形之圖形； 圖13是用於圖9方法之爆炸移動情形之圖形；且 圖14是用於圖3，4，與9之方法之一電腦系統的簡化方 塊圖。 雖然很谷易對於本發明進行各種修改與運用替代型態， 附圖之特定實例是做爲範例以供展示，且在下文中將受到 詳細説明。但是，應可瞭解本文所展現之附圖與詳細説明 未意謂限制本發明於所揭示之特定實例，而是相反地，本 又意欲涵蓋屬於本發明之精神與範疇之所有修改，等效 物，與替代物，其中本發明之精神與範疇是由附加之申請 專利範圍來定義。 主羞二較佳實例之詳細説明 現在請參看附圖，圖2，3，與4包含一方法之圖形與流 程圖’其中該方法以-有效率之方式顯示圖形影像來模擬 2動，物件型態，視野深度，或相關之時間導向效果。如 刖所討論’一應用程式可產生一圖形景物，其中該景物包 含在移動（或隨時間，距離，某一其他變數而變化）之一或 更多動態物件’與-或更多靜態物件。受到考慮之顯示系 、:通常包含一週期性受到更新或復原之顯示螢幕。顯示螢 幕受到更新之速率是籍由顯示硬體來決定，且通常不受應 用程式設計者之㈣。藉由使用常見之60赫茲之顯示榮幕 更新率，大約每I6·7毫秒（更新週期）會展現一新螢幕。在 ---i----ΓI -----------1--t---------^ (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 JW83 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明（8 ) 涉及移動或其他時間導向變化之 <奇多應用中，在更新週期 中動態物件劇烈變化（在位置或A e 4其他万面）。模糊技術最好 疋用以模擬一物件在一指定時間 把、 吁間間搞所經歷之變化，且該 和足之時間間隔通常至少與更4 44_ 更新週期一樣長。傳統之模糊 技術，如前所討論，通常包含— , 私序，其中整個圖形景物 疋在-些預先決定之間隔受到複製，且累積於一專屬缓衝 S ’直到累積I影像最終轉移至一畫面緩衝區以供顯示爲 止。本發明構思-種技術以改善用以達成移動模糊之傳統 累％技術〈效率。圖3之流程圖描述_種根據本發明之一 實例來達成移動模糊之方法3⑼。在第—步驟搬，需要模 叙圖形景物之每-物件或圖形基元（亦即—動態基元）適 當地受到指定，以辨別動態物件與其餘之靜態物件。靜態 與動態物件之此種指定可發生於應用層次，或另外可發生 、爲特足API之伤。在靜態與動態物件之指定在步驟 302發生之後，模糊技術3〇〇包含第二步驟卯*，其中一或更 多模糊幾何附加至在步驟3〇2中指定成爲動態基元之每一 基元。模糊附加幾何之位置，大小與屬性是用以模擬動態 物件在扣足之時間間隔中所經歷之移動或其他變化。相較 於以則用以達成動態物件之模糊之累積技術，決定適當之 模糊▲何更爲迅速地發生，且需要遠較少之記憶體。在附 加適當t模糊幾何至動態物件之後，圖形基元與他們之附 加杈糊幾何儲存於（步驟3〇6) 一畫面缓衝區，或適合隨後顯 示於一顯示螢幕之其他儲存位置。 圖2是本發明之一實例所構思之一模糊技術的簡化圖 本紙張尺㈣财目目涵 -11 - ^297公釐） (請先閱讀背面之注咅？事項再填寫本頁) -J9 訂—-----線· 經濟部智慧財產局員工消費合作社印製 512283 A7 B7__ 五、發明說明（9 ) 形，以描繪圖形景物200，其中包含靜態周圍背景201内之 一動態圖形基元202a與其之附加模糊幾何210。雖然描繪之 景物200只包含單一動態基元202a，應可理解本文所述之技 術能夠延伸至多個此種動態基元。動態基元202a，如圖3 所示，是包含一頂點集合206a，206b與206c(通稱爲頂點206) 之三角形。極多圖形顯像法運用三角形基元來表示較大圖 形物件之某些部份。每一對頂點定義基元202a之一獨特邊 緣204a5 204b與204c(統稱爲邊緣204)。一先前基元202b是利 用虛線來展現，以表示基元202a之一先前位置。因此，如 果基元202a表示動態基元在時間T之位置，則先前基元202b 表示動態基元在時間T - DELTA之位置，其中DELTA是一預 先決定之時間間隔。此預先決定之時間間隔之一適當選擇 通常至少與景物200可受到顯示之一顯示螢幕之更新週期 一樣長。在此實例中，基元在指定區間之位置變化表示， 至少，基元在景物200受到顯示期間之位置變化。 如先前參照圖3之討論所示，受到考慮之實例所構思之 圖形顯示方法300包含辨別動態基元，例如基元202a，與靜 態基元，例如構成景物200之靜態背景201之基元。在一實 例中，動態基元之指定是藉由使用移動向量來達成。在較 佳實例中，動態基元202a之每一頂點206關聯於至少一移動 向量。在一實例中，每一頂點204可指定成爲一獨特移動 向量。在另一實例中，每一頂點204可繼承來自動態基元 202a所指定之一移動向量之移動向量。在描述之實例中， 關聯於頂點206a與206b之移動向量是分別利用參考號碼208a -12- 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公釐） ------：---------------訂 ---------線 (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 512283 A7 B7 五、發明說明（10) 與208b來表示（雖然關聯於頂點206c之移動向量受到隱藏而 看不見）。移動向量208顯示他們之關聯頂點206之一時變參 數的變化。在受到考慮之時間導向效應是移動之實例中， 例如，移動向量208之大小與方向顯示他們之關聯頂點206 在指定時間間隔DELTA期間之方向與位移。藉由使用第一 移動向量208a與第一頂點206a之座標，藉由參考號碼212a表 示於圖2之先前第一頂點可受到決定。換句話説，第一移 動向量208a自第一頂點206a指向第一先前頂點212a，其中第 一先前頂點212a表示第一頂點在DELTA時間間隔之開端之 位置。同樣地，第二移動向量208b自第二頂點206b指向第 二先前頂點212b。因此藉由觀察圖2，應可理解頂點206之 移動方向相反於他們之對應移動向量208之方向。 移動向量208可在應用程式層次關聯於指定之頂點或基 元，或做爲應用程式介面之一功能。關聯於非零大小之向 量之基元或頂點是指定成爲動態，而未關聯於移動向量 (或關聯於零大小之向量）之基元或頂點是指定成爲靜態。 對於線性移動之情形，如圖2所描繪，單一移動向量即足 以描述其之關聯頂點在時間間隔DELTA中之路徑。現在請 簡短參看圖5，動態基元202a之非線性移動情形受到展 示。非線性移動是藉由延伸線性移動之情形以包含關聯於 每一頂點206之多個移動向量來模擬。在圖5中，例如，第 一頂點206a關聯於移動向量208a，208c，208e，與208g，而第 二頂點206b關聯於移動向量208b，208d，208f，與208h。延伸 本發明以涵蓋非線性移動之情形類似於爲眾所知之使用雲 -13- 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公釐） --Γ--J---------------訂---------線 (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 512283 A7 B7_ 五、發明說明（11) 線來表示曲線。在最簡單之實例中，模糊幾何210是四邊 形帶來表示，其中第一四邊形211是以移動向量208a與208b 來界定，第二四邊形213是以208c與208d來界定，依此類 推。 現在請回到圖2，一模糊幾何210如圖所示附加至動態基 元202a之一對應邊緣204a(統稱爲邊緣或邊緣204)。模糊幾 何210之大小與屬性顯示動態基元202a之一時間相依參數之 變化。在模糊是用以顯示移動之情形中，例如，模糊幾何 210之大小與屬性顯示，至少，動態基元202a在指定之時間 間隔DELTA中之位置變化。對於受到模糊之動態基元是三 角形之情形，例如圖2之動態基元202a，每一對頂點206定 義基元202a之一對應邊緣204。在本發明之較佳實例中，模 糊幾何210附加至動態基元202a之一對應邊緣。在線性移動 之情形中，較佳實例構思四邊形模糊幾何，例如描繪之模 糊幾何210。在此種情形中，模糊幾何210是藉由下列來定 義：附加模糊幾何210之動態基元202a之邊緣204a，以及起 源於第一頂點206a及終止於第一先前頂點212a之第一直線 209a，起源於第二頂點206b及終止於第二先前頂點212b之 第二直線209b，與在第一先前頂點212a及第二先前頂點 212b間延伸之第三直線209c。一旦分別給定第一與第二頂 點206a與206b之座標與第一及第二移動向量208a及208b之方 向及大小，則可輕易決定此四邊形，且此四邊形可在顯像 程序之各種階段受到決定。在一適用於達成迅速模糊幾何 產生且需要最少之應用程式設計虛耗之實例中，模糊幾何 _-14-__ 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公釐） —：—：---------------訂---------線 (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 512283 A7 ---- 五、發明說明（12 ) 產生是利用包含於電腦圖形系統之圖形轉接器之專屬電路 在光域層次受到完成，如下文更詳細所討論。 模糊成、何210之屬性控制與模糊在較佳實例巾是藉由關 聯-漸隱因數與動態基元2〇2a來達成。最好，鄰接其之對 應邊緣204a之模糊幾何21〇的屬性（例如顏色，紋理座標， 與透明度）匹配邊緣2G4a本身之屬性。漸隱因數接著是用於 較佳實例以隨著與邊緣204a之距離之增加來減少模糊幾何 210之屬性値。合成之模糊幾何21〇因此自邊緣扣如至第三 直線209c所表示之後邊緣在外觀上逐漸隱沒。因此，模糊 幾何210鄰接邊緣204a之顏色匹配邊緣2〇知之顏色（此顏色 接著是分別利用第一與第二頂點2〇仏與2〇汕之顏色屬性來 推導）。一指配給基元202a之漸隱因數（藉由應用程式或做 爲API管線之一部份），且是用以，例如，減少遠離邊緣 2 〇 4 a之幾何部份之模糊幾何2 i 〇的不透明度（或增加透明 度）。因爲藉由使用如前所述之關聯漸隱因數來達成漸隱 放果符合當觀看在迅速移動之物件所感受之漸隱效果，所 以本發明提供一表示移動物件之高效率機制。整個圖形基 元之屬性資訊可利用專屬電路，基於動態基元2〇2&之屬性 値’模糊幾何210之大小，與漸隱因數，來迅速達成。使 用本文所述之模糊幾何210與模糊技術所達成之模糊，藉 由無需在一大型累積缓衝區中累積景物2〇〇之多個快照， 可使用較少之時間與遠較少之記憶體來產生一模糊效果。 前面段落所述之程序是以圖4之流程圖來表示，且該流 程圖詳細描述圖3之流程圖之一特定實例400。更明確地 ____ -15· >紙張尺度適用中國國家標準（CNS)A4規格⑵0 x 297公爱) -~ - l· --------訂-------!線· C請先閱讀背面之ii意事項再填寫本頁) A7 A7 經濟部智慧財產局員工消費合作社印製 五、發明說明（13) 說，程序早 ,,% .. —外迴圈，其中針對一圖形景物之每一物 件來執行該迴圍曰、 驟搬），且^傻ΓΓ物件是根據—顯像法來顯像（步 Μ ·項像法包含，除了傳統之顯像技術以外，每 一動悲基元血一彡々私 一 /、 杪動向里及漸隱因數之關聯。在步驟 104/Λ移動向|接著最好是在光域層次受到使用，以決 疋疋°又到考慮之物件是靜態或動態（亦即在移動）。如果 孩物件是靜態，則該物件無需額外考量。但是，如果受到 考慮（物件疋動態’則針對該物件之每—邊緣來執行—内 迴圈。在内迴圈Φ —一 _中 關駟於母一頂點之移動向量資訊是與 每-頂點之位置，，員色，與其他屬性一起用以計算(步驟 4〇6)孩等先前頂點（亦即在時間T-DELTA之該等了員點）之位 置，顏色，與其他屬性資訊。先前頂點資訊接著是用以產 生主在步驟408 ’四邊形型態之一模糊幾何(對於線性移動 4形）°在本發明之—實例中，對於動態物件之每-邊 緣皆重複内迴圈。名i仙每 在/、他K例中，如下文所討論，模糊幾 何八附加至動怨物件之選定邊緣。無論是何種情形，對於 景物之每_物件皆重複整個程序，直到所有想要之模糊幾 何皆已附加至他們之對應邊緣並儲存畫面緩衝區爲 止0 圖6a與6b描述本發明之其他實例，其中針對每—動態物 件202 a所產生之模糊幾何21〇之數目有所不同。在圖^所描 ^巾針對u基π 2Q2a之每—邊緣删來產生模糊 幾何。在此實例中’每—模糊幾何加是根據前述之方 法，基於-頂點對，來產生，其中該頂點對定義附加模糊 ---^---------------I--^--------- (請先閱讀背面之注音？事項再填寫本頁)A7 B7 V. Description of the invention (Figure 11 之一 Figure 10 is one of the blurring technology graphics, in which each pixel of the primitive is associated with a moving line; Figure 12 图形 Figure for the internal breaking movement of the method of Figure 9; Figure 13 FIG. 14 is a diagram of an explosion movement situation used in the method of FIG. 9; and FIG. 14 is a simplified block diagram of a computer system used in one of the methods of FIGS. 3, 4, and 9. Although it is easy to make various modifications and application substitutions to the present invention Types, specific examples of the drawings are examples for illustration, and will be described in detail below. However, it should be understood that the drawings and detailed descriptions presented herein are not meant to limit the invention to the specific examples disclosed On the contrary, the present invention is intended to cover all modifications, equivalents, and alternatives belonging to the spirit and scope of the present invention, wherein the spirit and scope of the present invention are defined by the scope of additional patent applications. For a detailed description of a good example, please refer to the drawings. Figures 2, 3, and 4 contain a method's graphics and flowchart. 'The method displays graphic images in an efficient way to simulate 2 movements, object types. , Depth of field, or related time-directed effects. As discussed in 'An application can produce a graphical scene, where the scene contains one or more of moving (or changing with time, distance, some other variable). "Dynamic objects" and-or more static objects. The display system under consideration: usually includes a display screen that is periodically updated or restored. The rate at which the display screen is updated is determined by the display hardware and is usually not affected by For application designers. By using the common 60Hz display glory update rate, a new screen will be displayed approximately every I7 · 7 milliseconds (update period). --- i ---- ΓI --- -------- 1--t --------- ^ (Please read the precautions on the back before filling out this page) Printed by JW83 A7 B7 Employee Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs Printed by the Intellectual Property Bureau's Consumer Cooperatives V. Invention Description (8) In mobile applications or other time-oriented changes < Qiduo applications, dynamic objects change drastically during the update cycle (in locations or Ae 4 other faces). Fuzzy techniques are best used to simulate The changes that an object undergoes at a specified time, and the time interval between the sums is usually at least as long as the 44_ update period. Traditional fuzzy techniques, as previously discussed, usually include —, private sequence In which, the entire graphic scene is copied at some predetermined intervals and accumulated in a dedicated buffer S 'until the accumulated I image is finally transferred to a picture buffer for display. The present inventive concept-a technology to improve the use of In order to achieve the traditional cumulative technology of moving blur <efficiency. The flow chart of Figure 3 describes a method 3 to achieve moving blur according to an example of the present invention. In the first step, it is necessary to model each object of the graphic scene or Graphic primitives (that is, dynamic primitives) are appropriately specified to distinguish dynamic objects from the rest of static objects. This designation of static and dynamic objects can occur at the application level, or it can also occur, which is an API-specific injury. After the assignment of static and dynamic objects occurs in step 302, the blurring technique 300 includes a second step 卯 *, in which one or more fuzzy geometries are appended to each primitive designated as a dynamic primitive in step 302. . The position, size, and attributes of the fuzzy additional geometry are used to simulate the movement or other changes experienced by a dynamic object during the deduction time interval. Compared to the accumulation technology used to achieve the blurring of dynamic objects, determining the appropriate blurring can happen more quickly and requires far less memory. After adding the appropriate fuzzy geometry to the dynamic object, the graphics primitives and their additional paste geometry are stored in (step 306) a picture buffer, or other storage location suitable for subsequent display on a display screen. Figure 2 is a simplified diagram of a fuzzy technique conceived as an example of the present invention. Paper size of the paper is -11-^ 297 mm) (Please read the note on the back? Matters before filling out this page) -J9 Order —----- line · Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 512283 A7 B7__ V. Description of the invention (9) Shape to depict the graphic scene 200, which contains one of the dynamic graphic primitives in the static surrounding background 201 202a and its additional fuzzy geometry 210. Although the depicted scene 200 contains only a single dynamic primitive 202a, it should be understood that the techniques described herein can be extended to multiple such dynamic primitives. Dynamic primitive 202a, as shown in FIG. 3, is a triangle containing a set of vertices 206a, 206b, and 206c (collectively referred to as vertices 206). Most graphics development uses triangular primitives to represent some parts of a larger graphic object. Each pair of vertices defines a unique edge 204a5 204b and 204c (collectively referred to as edge 204) of the primitive 202a. A previous primitive 202b is shown with a dashed line to indicate a previous location of one of the primitives 202a. Therefore, if primitive 202a indicates the position of the dynamic primitive at time T, the previous primitive 202b indicates the position of the dynamic primitive at time T-DELTA, where DELTA is a predetermined time interval. An appropriate selection of this predetermined time interval is usually at least as long as the update cycle of one display screen where the scene 200 can be displayed. In this example, the change in the position of the primitive in the designated interval indicates that, at least, the position of the primitive during the display of the scene 200 changes. As previously discussed with reference to FIG. 3, the graphical display method 300 conceived by the considered example includes identifying dynamic primitives, such as primitive 202a, and static primitives, such as primitives that form the static background 201 of the scene 200. In one example, the assignment of dynamic primitives is achieved by using motion vectors. In the preferred example, each vertex 206 of the dynamic primitive 202a is associated with at least one motion vector. In one example, each vertex 204 may be designated as a unique motion vector. In another example, each vertex 204 may inherit a motion vector from one of the motion vectors specified by the dynamic primitive 202a. In the example described, the movement vectors associated with the vertices 206a and 206b are respectively using the reference numbers 208a -12- This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) ------: --------------- Order --------- line (Please read the precautions on the back before filling out this page) Printed by the Employee Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 512283 A7 B7 V. Invention description (10) and 208b (although the movement vector associated with vertex 206c is hidden and cannot be seen). The movement vector 208 shows the change of one of their associated vertices 206 over time. In instances where the time-directed effect under consideration is movement, for example, the magnitude and direction of the movement vector 208 shows their direction and displacement of their associated vertices 206 during the specified time interval DELTA. By using the coordinates of the first motion vector 208a and the first vertex 206a, the previous first vertex shown in FIG. 2 by the reference number 212a can be determined. In other words, the first motion vector 208a points from the first vertex 206a to the first previous vertex 212a, where the first previous vertex 212a represents the position of the first vertex at the beginning of the DELTA time interval. Similarly, the second motion vector 208b points from the second vertex 206b to the second previous vertex 212b. Therefore, by looking at FIG. 2, it should be understood that the moving directions of the vertices 206 are opposite to the directions of their corresponding moving vectors 208. The motion vector 208 may be associated with a specified vertex or primitive at the application level, or it may be a function of the application interface. Primitives or vertices that are associated with a non-zero size vector are designated as dynamic, while primitives or vertices that are not associated with a motion vector (or a vector of zero size) are designated as static. For the case of linear movement, as depicted in Figure 2, a single motion vector is sufficient to describe the path of its associated vertex in the time interval DELTA. Referring now briefly to Fig. 5, the non-linear movement of the dynamic primitive 202a is shown. Non-linear motion is modeled by extending the linear motion scenario to include multiple motion vectors associated with each vertex 206. In FIG. 5, for example, the first vertex 206a is associated with the motion vectors 208a, 208c, 208e, and 208g, and the second vertex 206b is associated with the motion vectors 208b, 208d, 208f, and 208h. Extending the present invention to cover the case of non-linear movement is similar to the known use of cloud-13- This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) --Γ--J --- ------------ Order --------- line (please read the notes on the back before filling out this page) Printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economy 512283 A7 B7_ 5 (11) Line of the invention description curve. In the simplest example, fuzzy geometry 210 is a quadrilateral band representation, where the first quadrilateral 211 is defined by the motion vectors 208a and 208b, the second quadrilateral 213 is defined by the 208c and 208d, and so on . Returning now to FIG. 2, a fuzzy geometry 210 is attached to one of the dynamic elements 202a corresponding edge 204a (collectively referred to as edge or edge 204) as shown. The size and attributes of the fuzzy geometry 210 show changes in one of the time-dependent parameters of the dynamic primitive 202a. In the case where blur is used to show movement, for example, the size and attributes of the fuzzy geometry 210 are displayed, at least, the position of the dynamic primitive 202a at a specified time interval DELTA changes. For cases where the blurred dynamic primitives are triangular, such as the dynamic primitive 202a of FIG. 2, each pair of vertices 206 defines one of the primitives 202a corresponding to the edge 204. In the preferred embodiment of the present invention, the fuzzy geometry 210 is attached to a corresponding edge of one of the dynamic primitives 202a. In the case of linear movement, a better example contemplates a quadrilateral fuzzy geometry, such as the blurred geometry 210 depicted. In this case, fuzzy geometry 210 is defined by: the edge 204a of the dynamic primitive 202a of the additional fuzzy geometry 210, and the first straight line 209a originating from the first vertex 206a and terminating at the first previous vertex 212a, origin A second straight line 209b between the second vertex 206b and the second previous vertex 212b, and a third straight line 209c extending between the first previous vertex 212a and the second previous vertex 212b. Once the coordinates of the first and second vertices 206a and 206b and the directions and sizes of the first and second motion vectors 208a and 208b are given, the quadrilateral can be easily determined, and the quadrilateral can be subjected to at various stages of the imaging process. Decide. In an example suitable for achieving rapid fuzzy geometry generation and requiring minimal application design waste, fuzzy geometry _-14 -__ This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) —: —: --------------- Order --------- line (please read the precautions on the back before filling out this page) System 512283 A7 ---- V. Description of the invention (12) The production is performed at the optical domain level using the exclusive circuit of the graphics adapter included in the computer graphics system, as discussed in more detail below. The attribute control and fuzzing of fuzzy formation and Ho 210 are achieved in a better example by the correlation-fade factor and dynamic primitive 002a. Preferably, the attributes (e.g., color, texture coordinates, and transparency) of the fuzzy geometry 21o adjacent to the corresponding edge 204a match the attributes of the edge 2G4a itself. The fade-out factor is then used for a better example to reduce the properties of the fuzzy geometry 210 as the distance from the edge 204a increases. The resulting fuzzy geometry 21 is therefore gradually obscured in appearance from the edge buckle as indicated by the third straight line 209c. Therefore, the color of the adjacent edge 204a of the fuzzy geometry 210 matches the color of the edge 20 (this color is then derived using the color attributes of the first and second vertices 20 顶点 and 20 汕, respectively). A fade factor assigned to the primitive 202a (either by the application or as part of the API pipeline), and used to, for example, reduce the fuzzy geometry 2 i 〇 of the geometric part away from the edge 〇 04 a Opacity (or increase transparency). Because the fading effect is achieved by using the associated fading factor as described above, the fading effect is consistent with the fading effect experienced when viewing a rapidly moving object, so the present invention provides a highly efficient mechanism for representing moving objects. The attribute information of the entire graphic element can be achieved quickly using a dedicated circuit based on the size of the fuzzy element 210 of the dynamic element 202 & and the fade factor. Using the fuzzy geometry 210 and fuzzy technology described in this article, by eliminating the need to accumulate multiple snapshots of the scene 200 in a large accumulation buffer, less time and far less memory can be used To produce a blur effect. The procedure described in the previous paragraph is represented by the flowchart of FIG. 4, and the flowchart describes a specific example 400 of the flowchart of FIG. 3 in detail. More specifically ____ -15 · > Paper size applies Chinese National Standard (CNS) A4 specification (0 x 297 public love)-~-l · -------- Order -------! Line · C, please read the meanings on the back before filling in this page) A7 A7 Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs V. Invention Description (13) Said that the procedure is early,%. Each object of a graphic scene performs the encirclement and removal), and the object is developed according to the imaging method (step M · item image method includes, in addition to the traditional imaging technology, each A movement of sorrowful primitive blood, a sacrifice, a sacrifice, and a fading factor. At step 104, Λmoving direction | is then preferably used at the level of the light field to determine 又 ° and consider it. The object is static or dynamic (that is, moving). If the object is static, the object does not need to be considered. However, if it is considered (object dynamic), the inner loop is performed for each edge of the object. In the inner circle Φ — a _ in the parent vector, the movement vector information of the vertex is related to the position of each vertex. , Together with other attributes, is used to calculate (step 406) the position, color, and other attribute information of the child's previous vertex (that is, at the time T-DELTA). The previous vertex information is then used to In step 408, one of the quadrangular shapes is generated (for a linearly moving 4-shape). In the example of the present invention, the inner loop is repeated for each and every edge of the dynamic object. In the example K, as discussed below, the fuzzy geometry is attached to the selected edges of the object. In any case, the entire process is repeated for each object of the scene until all the desired fuzzy geometry has been attached to them. Figures 6a and 6b describe other examples of the present invention, in which the number of fuzzy geometries 21 generated for each dynamic object 202a is different. The description in Figure ^ Each basis of u-base π 2Q2a is deleted to generate fuzzy geometry. In this example, 'each-fuzzy geometry addition is generated according to the method described above based on-vertex pairs, where the vertex pair defines additional blurring --- ^- -------------- I-^ --------- (Please read the note on the back? Matters before filling out this page)
512283 經濟部智慧財產局員工消費合作社印製 A7 B7__ 五、發明說明（14) 幾何210之邊緣。因此，第一模糊幾何210a附加至第一邊緣 204a(藉由第一與第二頂點206a與206b來定義），第二模糊幾 何210b附加至第二邊緣204b(藉由頂點206b與206c來定義）， 依此類推。此實例具有簡單之好處，但是當受到產生之幾 何最終受到隱藏而無法看見時，可導致效能與品質之下 降。圖6b之實例藉由下列方式來克服此潛在缺點：關聯一 額外資訊於每一邊緣204，且運用此資訊來決定是否應爲 該邊緣產生一模糊幾何210。更明確地説，圖6b所描述之實 例構思領先邊緣/落後邊緣與動態基元202a之每一邊緣204 之關聯。爲説明本文起見，一落後邊緣是定義成爲在整個 時間間隔DELTA中皆爲可見之一邊緣204。在圖6b之特定情 形中，例如，204a是一落後邊緣，而邊緣204b與204c是領先 邊緣。一般而言，因爲附加之模糊幾何在顯示之影像中受 到隱藏而無法看見，所以無需且最好不要產生領先邊緣， 例如邊緣204b與204c，之模糊幾何。因此，在模糊幾何之 產生期間忽略領先邊緣較佳且更爲有效率（在辨別一領先 邊緣所需之時間少於產生一領先邊緣之模糊幾何所需之時 間的合理假設下）。但是即使包含領先邊緣/落後邊緣資訊 實際上會減緩附加模糊幾何至動態基元202a之程序，最妤 選擇性產生領先邊緣之模糊幾何以改善最終影像之品質。 影像品質可能由於應看不見之模糊幾何之存在而受損。此 概念之一延伸是藉由隱藏基元600與其之關聯模糊幾何610 來進一步表示於圖6b。在得知圖形景物之每一基元之深度 資訊之後，隱藏物件，例如物件600，可自晝面緩衝區中 _-17-_ 本紙張尺度適用中國國家標準(CNS)A4規格（210 X 297公釐) --：----r--------------訂---------線 (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 Μ氏張尺度適用中國國家標準（CNS〉A4規格（210 X 297公釐了 A7 五、發明說明（15) 刪除，以自顯示之景物中去除不要之組件。 現在清參看圖7a，7b，7c，與8，關於本發明之紋理對映反 射4形受到考慮。動態物件與他們之組成基元未總是維 1相同〈紋理外觀。-物件之表面之反射可能隨著物件之 考夕動而改交。在一使用累積之傳統模糊方法中，針對用以 產生杈糊效果（圖形景物之每一快照重新計算反射之影 像。極少需要相對於反射之此種位準之詳細度。本發明之 一實例構思一種用以達成模糊幾何之反射影像之方法，其 中刪必而產生之反射顯像之數目。此實例適用於支援犯 紋理之顯像系統，其中使用”s”與，,t"來表示犯紋理座標， 而r來表π第二座標。在此實例中，，，s "與”丨”紋理座標 是以傳統方式來指示紋理，但是使用"r，，索引來表示時 間而非i間。在此方法中，在多個時間點多個反射影像 受到產生。在線性移動之情形中，如圖8所示，一對反射 〜像又到產生。雖然爲簡化起見，描繪之範例展示只使用 二反射影像之本發明之一實例，應很容易理解可使用多於 二個反射影像。第一反射影像，如圖7a之動態基元2〇沘所 示，表示來自圖7c描繪之反射影像7〇1之反射，且在指定之 時間間隔DELTA之開端存在於動態基元2〇為之表面，而圖 7b之第二反射影像202a表示在指定之時間間隔之末端存在 於動態基元202a表面之反射（亦即目前存在於基元2〇2&之表 面之反射）。額外位準之準確度可藉由在指定時間間隔全 程間之距離相等之間隔顯像額外之反射影像來建構。該等 反射⑦彡像接著依序載入3D影像。模糊幾何21 〇之” s "與"玄” -18- l·----------------------^ (請先閱讀背面之注意事項再填寫本頁) 512283 A7 B7 經 濟 部 智 慧 財 產 局 員 工 消 費 合 作 社 印 製 五、發明說明（16) 紋理座標接著是根據傳統之反射對映來定義，而在表示指 足時間間隔之開端之模糊幾何部份，” r "座標是定義成爲 0.0，且在表示指定時間間隔之末端之模糊幾何部份，"r，， 座標足義成爲1·〇。在具有四邊形模糊幾何之線性移動之情 形中’沿基元202a之邊緣204a r= 1.0，而在邊緣209c r=0.0。接著藉由下列方式來產生模糊幾何21〇之其餘部份 t反射影像：設定紋理混合運作，以使用雙線性内插來在 最初反射影像與最終反射影像之間平順混合反射影像之顏 色。 現在請參看圖9,該圖展現一描述一方法或影像處理技 術900之流程圖，其中影像處理技術9〇〇是根據本發明之另 一實例運用移動向量來達成移動模糊。在圖9之流程圖所 表示之實例與伴隨之圖⑺-丨3之説明中，定義動態基元之每 一頂點之移動向量，與屬性値是藉由使用漸隱因數在基元 之全部（或部份）受到内插，以產生一移動直線集合。參照 圖9更明確地説，移動向量與漸隱因數，分別在步驟9〇2與 904，以前述之方式關聯於一動態基元之每一頂點。在步 驟906，頂點光域資料接著在動態基元2〇仏所定義之一像素 集合中進行内插。該頂點光域資料包含位置，顏色，紋理 座標，透明度，與其他適當資訊。光域資料受到内插之該 像素集合是決足於實例。在圖10所描述之實例中，由於其 之高效率效能（亦即快速顯像），該像素集合可局限於沿基 元202a之邊緣204之像素。在圖〗丨所描述之實例中，由於其 之卓越品質（亦即更準確之顯像），進行内插之該像素 ——l·--------------訂---------線 (請先閱讀背面之注意事項再填寫本頁)512283 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7__ V. Description of the Invention (14) Edge of Geometry 210. Therefore, the first fuzzy geometry 210a is attached to the first edge 204a (defined by the first and second vertices 206a and 206b), and the second fuzzy geometry 210b is attached to the second edge 204b (defined by the vertices 206b and 206c) , So on and so forth. This example has the advantage of simplicity, but when the resulting geometry is eventually hidden from view, it can lead to a decrease in performance and quality. The example of FIG. 6b overcomes this potential disadvantage by associating an additional information to each edge 204 and using this information to determine whether a fuzzy geometry 210 should be generated for the edge. More specifically, the example described in Figure 6b contemplates the association of leading edges / lagging edges with each edge 204 of the dynamic primitive 202a. For the purposes of this document, a trailing edge is defined as an edge 204 that is visible throughout the entire time interval DELTA. In the particular case of Fig. 6b, for example, 204a is a trailing edge and edges 204b and 204c are leading edges. In general, because the additional fuzzy geometry is hidden from view in the displayed image and cannot be seen, it is not necessary and best not to produce a fuzzy geometry of leading edges, such as edges 204b and 204c. Therefore, it is better and more efficient to ignore leading edges during the generation of fuzzy geometry (under a reasonable assumption that the time required to identify a leading edge is less than the time required to generate a fuzzy geometry of a leading edge). But even if including leading edge / lagging edge information will actually slow down the process of adding fuzzy geometry to the dynamic primitive 202a, it is best to selectively generate blurring geometry of the leading edge to improve the quality of the final image. Image quality may be impaired by the presence of blurry geometry that should be invisible. One extension of this concept is further shown in Figure 6b by hiding primitives 600 and their associated fuzzy geometry 610. After learning the depth information of each element of the graphic scene, hidden objects, such as object 600, can be retrieved from the daytime buffer _-17-_ This paper size applies to China National Standard (CNS) A4 (210 X 297) (Mm)-: ---- r -------------- Order --------- line (please read the precautions on the back before filling this page) Economy The Ministry of Intellectual Property Bureau's Consumer Cooperatives printed the M-sheet scales in accordance with Chinese national standards (CNS> A4 specifications (210 X 297 mm A7) V. Description of invention (15) Deleted to remove unnecessary components from the displayed scene. Now referring to Figs. 7a, 7b, 7c, and 8, the texture antireflection 4 shape of the present invention is considered. Dynamic objects and their constituent elements are not always the same as dimension 1 <texture appearance.-Reflection of the surface of the object May be changed as the object moves. In a traditional method of accumulation using accumulation, re-calculate the reflected image for each snapshot of the graphic scene. This is rarely needed relative to reflection. Level of detail. An example of the present invention contemplates a method to achieve the inverse of fuzzy geometry The method of radiating images, in which the number of reflections that must be generated is deleted. This example is suitable for imaging systems that support criminal textures, where "s" and, t " are used to represent the texture coordinates, and r is used to represent π The second coordinate. In this example, the "s" and "丨" texture coordinates indicate the texture in the traditional way, but use the "r", index to indicate time instead of i. In this method, between Multiple reflection images are generated at multiple time points. In the case of linear movement, as shown in Figure 8, a pair of reflections ~ images are generated again. Although for the sake of simplicity, the example shown is shown using only two reflection images. As an example of the invention, it should be easy to understand that more than two reflection images can be used. The first reflection image, as shown in the dynamic primitive 20a of FIG. 7a, represents the reflection from the reflection image 701 depicted in FIG. 7c, And at the beginning of the specified time interval DELTA exists on the surface of the dynamic primitive 20, and the second reflection image 202a in FIG. 7b represents the reflection that exists on the surface of the dynamic primitive 202a at the end of the specified interval (that is, the current presence The reflection of the surface of the element 200 &). The accuracy of the extra level can be constructed by imaging the extra reflection images at equal intervals throughout the specified time interval. The reflection artifacts are then sequentially Load 3D image. Fuzzy Geometry 21 〇 ”s " and” Xuan ”-18- l · ---------------------- ^ (please first Read the notes on the back and fill in this page) 512283 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of the invention (16) The texture coordinates are then defined according to the traditional reflection mapping, and indicate the time interval The fuzzy geometric part at the beginning of the "r" is defined as 0.0, and the fuzzy geometric part at the end of the specified time interval, " r, has a coordinate meaning of 1.0. In the case of a linear movement with a fuzzy geometry of a quadrangle, ′ is along the edge 204a r = 1.0 of the primitive 202a, and at the edge 209c r = 0.0. The rest of the blurred geometry 21 is then generated by the following method: Set the texture blending operation to use bilinear interpolation to smoothly blend the colors of the reflection image between the initial reflection image and the final reflection image. Reference is now made to Fig. 9, which shows a flowchart describing a method or image processing technique 900, in which the image processing technique 900 uses motion vectors to achieve motion blur according to another example of the present invention. In the example shown in the flowchart of FIG. 9 and the description of the accompanying figure ⑺-3, the movement vector and attribute 定义 of each vertex defining the dynamic primitive are defined by using the fade factor in all the primitives ( Or part) are interpolated to produce a set of moving straight lines. More specifically, referring to FIG. 9, the motion vector and the fade factor are associated with each vertex of a dynamic primitive in steps 902 and 904, respectively. At step 906, the vertex light domain data is then interpolated in a pixel set defined by the dynamic primitive 20 仏. The vertex light field data contains position, color, texture coordinates, transparency, and other appropriate information. This pixel set where the light domain data is interpolated is a matter of example. In the example described in FIG. 10, due to its high efficiency (i.e., rapid development), the pixel set may be limited to pixels along the edge 204 of the cell 202a. In the example described in Figure 〖丨, because of its superior quality (that is, more accurate imaging), the pixel that is interpolated-l · -------------- order --------- Line (Please read the precautions on the back before filling this page)
M2283 A7 B7 五、發明說明（17) 濟 部 智 慧 財 產 局 員 工 消 費 社 印 製 ，由包含於基元202a之所有像素所組成。無論選定之實例二’内插程序皆產生一内插像素資料集合，其中該集合 I每一構件對應於該像素集合 、 口 口足構件。在步驟908，對 万；夕動向1執行一類似之内插 換句选說，關聯於基元 2〇2at頂點之移動向量在相 ^ 你仰丨』万、艾驟906又内插所使用之像 一合雙到内插，以產生對應於該像素集合之一内插移動 向量集合。雖然圖9顯示内插步驟9〇6與9〇8成爲不同步驟， 應可理解該等步驟基本上可視建構而定受到整合。在頂點 $物與移動向量資訊之内插完成之後，定義之像素集 口:每I素具有一關聯之移動向量與光域資訊。此資訊 接著在移動向量之方向連同漸隱因數受到内插，以產生_ 移動直線集合，其中在移動向量之方向(亦即在相反於行 進方向之方向）逐漸隱沒，且導致關聯之基元而之模糊。 圖12與13描述使用影像處理技術9〇〇以達成移動模糊且需要 特殊考畺之特足情形。圖12表示内破移動之情形，其中基 元202a在時間間隔DELTAi末端之大小小於在該時間間^ (開端之大小。在此種情形中，方法900之步驟910所產生 (私動直線沿不同方向射出，而產生一通常不足以達成所 要目標之反轉星突波效果。爲處理此種情形，本發明之一 只例構思建造移動三角形，其中該等移動三角形是由鄰接 t移動直線對來界定，且在圖12是以參考號碼230來表示。 孩等二角形230可藉由保留來自先前像素之移動直線資訊 來適當建造。相對地，圖13表示爆炸移動之情形，其中基 元202a在時間間隔DELTA之末端之大小大於在該時間間隔 訂 線 I_____ - 20. 本纸張尺度適用中國國家標準（CNS)A4規格（21〇 χ 297公釐 512283 經濟部智慧財產局員工消費合作社印製 A7 五、發明說明（18) 之開端之大小。在此種情形之下’移動直線21 5收斂且可 重疊於移動直線215之終止點之附近。在此種情形之下， 重疊之移動直線可能傾向於不當地放大對應於多於一移動 直線之像素之強度。爲克服此種效應，爆炸移動之情形是 藉由縮放直線以致該直線之寬度小於一像素來處理。因爲 部份像素無法獲得，此效應可藉由縮小移動直線像素之不 透明度1/N來達成，其中1/N表示對應於一給定像素之重疊 直線之數目。如果，例如，一給定像素對應於3移動直線 215，則每一重疊移動直線之内插像素之不透明度將乘以 1/3，以產生一具有適當強度之像素。 現在請參看圖14，一代表性電腦系統140之簡化方塊圖受 到展tf，其中遠電腦系統適合建構參照圖2 - 13所述之模糊 技術。系統14〇包含一或更多處理單元142a，…142η，且每一 處理單元皆耦接至一主機匯流排144。一系統記憶體146可 自主機匯流排144經由一記憶體控制單元145來存取，且其 可建構成爲一離散裝置，或整合至處理單元142或匯流排 橋接器148。匯流排橋接器148位於主機匯流排144與週邊匯 流排150之間，且是用以便利該二匯流排間之通訊。週邊 匯流排150最好相容於多種工業標準週邊匯流排架構之任 何架構，例如在微處理機型電腦系統之領域爲眾所知之 PCI，MCA，或EISA架構。一根據本發明之圖形轉接器152 連接至週邊匯流排15〇，且是用以便利藉由位於系統°記憶 體146之一應用程式所產生之圖形景物的顯像。圖形轉接 器15 2最好包含—内部«裝置或記憶體以及各種内:電 路(未展示於此圖），例如圖形轉接器152經由其來與週： (請先閱讀背面之注意事項再填寫本頁) 訂---------線- -21 - A7 B7 五、發明說明（w) w排150進行通訊之週邊匯流排介面。圖形轉接器Η]進一 (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 7—畫面緩衝區介面，且該介面適用於允許轉接器 152來操縱畫面緩衝區153。畫面緩衝區153是—特殊用途儲 存裝置，用以儲存圖形影像之一數位表示以供顯示於一顯 不裝置156。一視訊控制器154受到調整以利用儲存於畫面 、、爰衝區153之影像來更新顯示裝置156。視訊控制器I%可整 口义圖形轉接器152，或另外可構成一離散裝置。同樣 j，晝面缓衝區153可整合於圖形轉接器152，或视訊控制 器154。圖形轉接器152進一步包含適當裝置以執行參照圖 3，4，與9所述之方法。對於針對圖3與4之模糊幾何方法 t實例，圖形轉接器152可包含專屬之模糊幾何產生電路 或庫人隨，其中该軟體是處於儲存於圖形轉接器儲存裝置之 %細指令之型態且適合供轉接器152之一處理單元來處 理，或電路與軟體之一組合。在此實例中，轉接器152之 模糊幾何產生器是配置成爲可辨識動態基元（亦即需要模 糊义基7L )且，做爲回應，決定及附加適當之模糊幾何至 動悲基7L之所要邊緣。另外，參照圖3與4所述之模糊幾何 方法之相對簡單性允許該等方法適當地嵌入一儲存於系統 記憶體145之應用程式。在針對先前參照圖9所述來達成模 糊之影像處理方法之實例中，圖形轉接器152最好包含移 動向量檢測，内插，與移動直線產生電路以產生適當之移 動直線。雖然此影像處理方法也可想像成爲包含於一應用 程式’應可理解由於此影像處理方法之相對複雜度，其最 好是在像素位準以硬體來建構。 熟悉本技術領域並閲讀本文者應可明瞭本發明構思—種 南效率方法，裝置，與系統以達成時間導向圖形效果，例如 _____ -22- 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公釐） 512283 A7 B7 圍意欲廣義解讀成為涵蓋 修正太 Ί" 第089110324號專利申請案 中文說明書修正頁(91年6月） 五、發明説明（2(3 ) L~-- 移動模糊。應可瞭解在詳細說明與附圖中展示與說明之本 發明之型態只是做為目前較佳之範例。下列之申請專利範 所揭示之較佳實例之所有變型。 件符號說明 厂 Jui .f;潜 4 ,、tv;11r修iE.· 140 電腦系統 1 鬈 142a... 142η 處理單元 i 3 144 主機滙流排 ] 145 記憶體控制單元 Ί 146 系統記憶體 : 148 滙流排橋接器 150 週邊滙流排 152 圖形轉接器 153 畫面缓衝區 154 視訊控制器 156 顯示裝置 200 圖形景物 201 背景 202a、202b 動態圖形基元 204a、204b 、204c 邊緣 206a' 206b 、206c 頂點 208 移動向量 209 直線 210 模糊幾何 211 第一四邊形 212 頂點 213 第二四邊形 215 移動直線 230 移動三角形 600 基元 610 模糊幾何 701 影像 -23- 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公釐)M2283 A7 B7 V. Description of the invention (17) Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs, Consumer Affairs Agency, and composed of all pixels included in element 202a. Regardless of the selected case two 'interpolation procedure, an interpolated pixel data set is generated, where each component of the set I corresponds to the pixel set and the mouth-foot-foot component. In step 908, a similar interpolation is performed on Wan; Xixiang1. The motion vector associated with the vertices of primitive 002at is related to the one used by Wan, Ai 906 and interpolation. Like one-to-two double-to-interpolation to generate a set of interpolated motion vectors corresponding to one of the pixel sets. Although FIG. 9 shows that the interpolation steps 906 and 908 are different steps, it should be understood that these steps are basically integrated depending on the construction. After interpolation of vertex $ object and motion vector information is completed, a defined pixel set: each I element has an associated motion vector and light field information. This information is then interpolated in the direction of the motion vector along with the fading factor to produce a set of _moving straight lines, in which the direction of the motion vector (that is, in a direction opposite to the direction of travel) gradually disappears and causes the associated primitives to The blur. Figures 12 and 13 describe the special case of using image processing technology 900 to achieve motion blur and requiring special examination. FIG. 12 shows the situation of internal breaking movement, in which the size of the primitive 202a at the end of the time interval DELTAi is smaller than the time ^ (the size of the beginning. In this case, the step 910 of method 900 (private moving lines along different Projected in a direction, which produces an inverse starburst effect that is usually not sufficient to achieve the desired goal. To deal with this situation, one example of the present invention contemplates the construction of moving triangles, where the moving triangles are formed by pairs of moving straight lines adjacent to t. It is delimited and is represented by the reference number 230 in Fig. 12. The cuboid 230 can be appropriately constructed by retaining the moving line information from the previous pixel. In contrast, Fig. 13 shows the situation of the explosion movement, in which the element 202a is in The size of the end of the time interval DELTA is larger than the line I_____ at the time interval. 20. This paper size applies the Chinese National Standard (CNS) A4 specification (21〇χ 297 mm 512283. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs. A7 V. The size of the beginning of the description of the invention (18). In this case, the 'moving line 21 5' converges and can overlap the ending point of the moving line 215 In this case, the overlapping moving lines may tend to improperly enlarge the intensity of pixels corresponding to more than one moving line. To overcome this effect, the situation of explosive movement is by scaling the line so that the width of the line Less than one pixel to process. Because some pixels are not available, this effect can be achieved by reducing the opacity of moving linear pixels by 1 / N, where 1 / N represents the number of overlapping lines corresponding to a given pixel. If, For example, if a given pixel corresponds to 3 moving lines 215, the opacity of the interpolated pixels of each overlapping moving line will be multiplied by 1/3 to produce a pixel with the appropriate intensity. Now refer to FIG. 14, a representative The simplified block diagram of the basic computer system 140 is shown in tf. The remote computer system is suitable for constructing the fuzzy technology described with reference to FIGS. 2 to 13. The system 14 includes one or more processing units 142a, ... 142η, and each processing unit is Coupled to a host bus 144. A system memory 146 can be accessed from the host bus 144 via a memory control unit 145, and it can be constructed as a discrete It is installed or integrated into the processing unit 142 or the bus bridge 148. The bus bridge 148 is located between the host bus 144 and the peripheral bus 150, and is used to facilitate communication between the two buses. The peripheral bus 150 Ideally compatible with any of a variety of industry standard peripheral bus architectures, such as the PCI, MCA, or EISA architecture known in the field of microprocessor-based computer systems. A graphics adapter 152 according to the present invention is connected 15 to the peripheral bus, and is used to facilitate the visualization of graphic scenes generated by an application located in the system ° memory 146. The graphics adapter 15 2 preferably contains-internal «device or memory and Various internal: circuits (not shown in this figure), such as the graphic adapter 152 to communicate with the week: (Please read the precautions on the back before filling this page) Order --------- Line-- 21-A7 B7 V. Description of the invention (w) The peripheral bus interface of the w row 150 for communication. Graphic adapter Η] further (please read the precautions on the back before filling out this page) 7-screen buffer interface printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs, and this interface is suitable for allowing adapter 152 to manipulate the screen Buffer 153. The picture buffer 153 is a special-purpose storage device for storing a digital representation of a graphic image for display on a display device 156. A video controller 154 is adjusted to update the display device 156 with the image stored in the screen and the flushing area 153. The video controller I% can integrate the graphics adapter 152, or it can form a discrete device. Similarly, the daytime buffer 153 can be integrated into the graphics adapter 152 or the video controller 154. The graphics adapter 152 further includes appropriate means to perform the methods described with reference to FIGS. 3, 4, and 9. For the example of the fuzzy geometry method t for FIGS. 3 and 4, the graphics adapter 152 may include a dedicated fuzzy geometry generating circuit or library, where the software is in the type of% fine instructions stored in the graphics adapter storage device. And is suitable for processing by a processing unit of the adapter 152, or a combination of a circuit and software. In this example, the fuzzy geometry generator of the adapter 152 is configured to recognize the dynamic primitives (that is, the fuzzy sense base 7L is required) and, in response, decides and appends the appropriate fuzzy geometry to the dynamic base 7L. Desired edge. In addition, the relative simplicity of the fuzzy geometry methods described with reference to Figures 3 and 4 allows these methods to be properly embedded in an application stored in system memory 145. In the example of the image processing method previously described with reference to Fig. 9, the graphic adapter 152 preferably includes motion vector detection, interpolation, and a motion line generating circuit to generate an appropriate motion line. Although this image processing method can also be imagined to be included in an application 'it should be understood that due to the relative complexity of this image processing method, it is best to construct it in hardware at the pixel level. Those familiar with the technical field and reading this article should understand the concept of the present invention-a kind of efficiency method, device, and system to achieve time-oriented graphical effects, such as _____ -22- This paper standard applies to China National Standard (CNS) A4 specifications ( (210 X 297 mm) 512283 A7 B7 is intended to be interpreted in a broad sense to cover the revision of Tai Chi " No. 089110324 Patent Application Chinese Manual Revised Page (June 91) V. Description of the invention (2 (3) L ~-Blur in motion It should be understood that the form of the invention shown and described in the detailed description and the drawings is only a presently preferred example. All variants of the preferred example disclosed in the following patent application. Symbol description factory Jui .f ; Qian 4, TV; 11r repair iE. · 140 computer system 1 鬈 142a ... 142η processing unit i 3 144 host bus] 145 memory control unit Ί 146 system memory: 148 bus bridge 150 peripheral bus Row 152 Graphics adapter 153 Picture buffer 154 Video controller 156 Display device 200 Graphics scene 201 Background 202a, 202b Dynamic graphics primitives 204a, 204b, 20 4c edge 206a '206b, 206c vertex 208 moving vector 209 straight line 210 fuzzy geometry 211 first quadrilateral 212 vertex 213 second quadrilateral 215 moving straight line 230 moving triangle 600 primitive 610 fuzzy geometry 701 image-23- scale of this paper Applicable to China National Standard (CNS) A4 (210 X 297 mm)