TW318923B - - Google Patents

Description

318923 V. Description of the invention (1 invention scope) The present invention is related to the computer image generation (CIG) system, in particular, for the display of those objects in the 5V image system at ⑽P; it is used in the display process The special color determination device and the special method for processing the image data describing the surface of the object. 'Background of the invention a: t ^ ^ The image system is designed to provide various simulation systems, such as the preparation of tank forks and the image reproduction of the flying simulator. This kind of simulator is used as a simulator to prepare soldiers without having to go out on the battlefield and use the actual combat system to practice exercises, such as crew members or tank guns, to maintain and improve their skills without practical training. Relevant costs. Therefore, the video display realism that the users of the instrument will benefit from the training and practice purposes is often helpful. P In a typical real-time computer image generation system, such as a flight simulation system, employees of the Central Standards Bureau of the Ministry of Economic Affairs Printed by the consumer cooperative ~ The production can be decomposed into three separate processing stages: controller, geometry processing state, and display 7F processor. This Each of the three processing stages or paragraphs independently processes data representing or related to one of the three consecutive scenes to be displayed. The controller processes in a fixed time, which can be a picture of 16 67 milliseconds: time or pen seconds One of the frame time, usually the former, the scene or image data, and then pass the processed data to the geometry processor. This feature: the screen and frame time is equivalent to the typical TV format display time, that is, three per second Ten frames with two interlaced scan frames per frame. The geometry processor has equal time to do calculations and the result is sent to the display processor at the end of the time. The display processor always processes the video display in one frame time -4 -

A7 A7 Printed by the Beigong Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economy V. Description of the invention (2) — An interlaced scan screen. In each time segment, all three processors continue to work, therefore, at any time t, the processor processes three separate scenarios. Each stage processes input and produces output in a time period, so data passes through these three stages in a sequential manner. The computational load is distributed into the knife-worker's processing stage. This allows each screen or frame time to display the new scene even if the result is delayed by three screens or frames. For example, in a flight simulation system, the pilot ’s input is received by the controller, and after being processed by the controller, geometry processor, and display processor in sequence, the color scene output is sent to the pilot to see in the simulator. The information seen by the staff of the video display 1 starts with a simple control input applied to the environment definition and is converted into pixel video display information. In the controller, the pilot's input is used to calculate the new position and direction of the aircraft ', thereby calculating the mathematical rotation matrix used to rotate objects from the reference coordinate system to the display coordinate system. Any other information related to the aircraft, target position and other aircraft is also calculated in the controller. In some applications, the controller is a general purpose computer. The geometry processor reads from the database object descriptions of scenes that may be seen in the stored three-dimensional digital representation. The read object is rotated into the display coordinates using the rotation matrix calculated in the controller. If the edges of these 3D objects exceed the window boundaries, they are clipped by the geometry processor. The geometry processor maps this 3D data to the 2D display window. The surface of the object beyond the limit of the display window is cropped in the two-dimensional image space. In addition, the geometry processor calculates (using a depth buffer or similar) which object is before or after which other object and stores this information in the priority list. Each object processed is composed of individual surfaces. Each surface is defined by a straight edge ------------ iT (please 53-read first Note on the back t fill in this page) -5- 318923 A7 B7 Printed by the Central Prototype Bureau of the Ministry of Economic Affairs, Consumer Labor Cooperatives V. Invention description (3 polygonal forms 3 priority order 6 people, I mouth% Jingzhong For the order of all surfaces, the surface in the table lists the most preferred surface in order, and the last surface in the table lists #lowest priority surface. When two surfaces reappear on the display, < ask priority surface It will be visible 'and the lower priority surface will be obscured. Finally, the geometric processing is calculated to define: the display endpoint of the edge of the dimension table (iv), and also calculate the Jugu; + ", two m ^ He Mi linearity of these boundary edges The coefficients of the equation. Obviously, it should not be regarded as a receiving-dimensional surface description, as well as surface color, surface attenuation information, "surface priority order list, basic unit structure, detailed level of blending" transparency "diagonal shadow, etc., and use this information to Data description of each pixel in the output scene Cut color display. Each color display in this search usually consists of 1024 visual information, each line has ⑻ "fixed color points or pixels. But the combination of other numbers of lines and pixels per line, It is also possible. The display processor must resolve for each pixel what it should be, & may appear on many surfaces on the pixel, and some parts of the surface may be covered by other surfaces. Because it can be on the display There are 1024 by 1024 pixels, and all these pixels must be calculated in 1 / 6〇 seconds, so the processing load in the display processor is very high. To calculate the multi display system (a single system contains at most 丨 4 Video), the display processor must perform more than 10 billion calculations per second. Due to the huge processing load, a display processor can only drive a few displays. In some systems, a single controller and geometry processor It can be used to drive several display processors and each display processor is connected to several displays. Therefore, the display processor represents the real-time computer image generation system. Most treatment, thus improving (Please read the notes and then fill in the back of this page) • installed · Order

318923 V. Description of the invention (4) The CIG system's effort and realism, the technology is not in the scope of the processor. In the anti-aliasing method, deal with the pixels of the fate. Table shock & shirt strength-it is a problem. Among the elements that the Dagger Pixel King currently wants to prevent the inconsistency of video data components ("missing") and 勮 ϋ and "" when 4 colors or transparency is seen Appropriate high-resolution edge effects can be seen. SUMMARY OF THE INVENTION According to the present invention, the display processor in the real-time computer image generator includes: an edge occlusion computing device, which receives a 毐 -Gamma μ% _ describing at least three vertices of each polygon to be displayed. The input data signal of a parent's display space (I, j) coordinates to determine which edge of each sequential pixel of the display raster, if any, will be cut by the polygon edge and on the determined edge The cutting of 2 sets; the edge cross data signals of all the polygons forming each pixel are stored in the frame buffer on the basis of each pixel, and the color data of each surface occupying any part of the pixel is also stored together: And a post-processing device, receiving data signals describing the coordinates of 12 edges of four pixels adjacent to each pixel to be displayed, to analyze the edge relationship and mix polygon color intensity data to determine the output of the video data received by CIG The last visible color of each pixel displayed by the display device. Economy 'Deng Central Standard Falcon Bureau Employee Consumer Cooperative Printed ---------, installed ------ tT-. (Please read the precautions on the back before filling out this page) The method of displaying the last visible color of each pixel displayed by the display device of the video data output of the image generator includes the following steps: 'First receive the display space describing each of the at least three vertices of each face to be displayed (I , J) the input data signal of the coordinate: then use this vertex data to determine which polygon edge, if any, will cut one of the pixels in a sequence that constitute the display raster; then decide -7- This paper scale is applicable to China Standard (CNS) A4 specification (210X297mm) 318923 V. Description of the invention (5 In the relevant image ♦; hi al, the cut on the building. The edges X of all the polygons forming each pixel are also cut, —,. Xunjie is stored in the frame buffer T ’on a per-pixel basis, and the color data of each surface of any part of the $ child pixel is also stored together: JS.Α, Yes 'For each Pixels shown, deal with a few of the selected adjacent pixels, ('丨 such as 12) the arrangement of the edges, to analyze the edge relationship and mix & evening edge shape color / 3 degrees of degree to determine the display pixel Finally, the color is visible. 'Therefore, an object of the present invention is to provide an innovative device and method to determine the color intensity of each pixel on the display of the computer imaging system. The details of the preferred embodiment will be clear and easy to see when referring to the accompanying drawings. Figure overview is a computer image generation (CIG) that can implement the device and method of the present invention. The general block diagram of the system: FIG. 2 is a spatial diagram illustrating the relationship between the object surface polygon in the objective space and the opposite image surface polygon in the image space, and the image elements (pixels) of the display raster that will display the image surface; Printed by the Employee Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs n-I --- -------I m ----- I, ςτ (please read the precautions on the back before filling in this page) Figure 2 a is Figure 2 is shown The enlarged part of the image display near the polygon; Figure 3a is a schematic diagram of a display pixel array developing a pair of polygonal surfaces: Figure 3b is a schematic diagram of a display pixel array positioning a polygonal surface, and also shows Various sampling corners C, pixel edges e and edge distances D necessary for operating the pixel color data determination method according to the present invention; -8-The winter paper scale is applicable to the Chinese National Standards (CNS) to the secret (210x297 public director) S 箄 办 ^ 4 | ^ In the patent application ^ such as the main page (June 86)

Description of the invention (0 Printed by Beigong Consumer Cooperatives, Central Bureau of Standards, Ministry of Economic Affairs *-3 C, pixel corners c 0 and its adjacent corners, the expression of edge intersection and distance, and it is useful for understanding several directions of the invention /, FIG. 4 is a block diagram of the color intensity sub-combination used in the cig display processor according to the principles of the present invention; 5 ′ and 5e are the edge resolutions used in the method of operating the color intensity sub-combination respectively Segment length decision, small polygon filtering and color mixing processing part of the flow chart; Figure 51) 'Display 8_ pixels for each pixel manipulated, 12 • edge segment arrangement and useful for understanding the operation of the method; Figure 5b " shows a direction determined by the length of the line in this method; Figure 5c shows the implementation of mixing colors in a 3 * 3 pixel star arrangement; Figure 6a shows the rest of the group ’s selection of bits for different levels of transparency and different styles The corners of sampling, and Figures 6b-6g show various different transparency operations and conditions, and they are all useful for understanding the "transparent" method of achieving transparency; Figure 7 is based on the strong color of the present invention A summary block diagram of the implementation of a possible innovative post-processor edge register / parser device; and Figures 8a-8g show all possible pixel sets with a pair of edges cut by a single-polygon edge, and all four of a pixel A single possible situation where an edge is cut by an edge formed between four polygons overlapping on the pixel. Detailed description of the exemplary embodiment Referring to FIG. 1 initially, a functional block diagram of a computer image generation system 10, such as an aircraft cockpit simulation system, is shown. Existing CIG skills, including the "Advanced Video Object Generator" of US Patent No. 4,727,365 issued to Bunker et al., February 23, 1988, and i989, April 25, issued -9-This paper size is applicable to China National standard (CNS > A4 specification (210X297 male thin) n ^ i I ^-(please read the precautions on the back before filling in this page) Order 318923 V. Description of invention (, D.MerZ US Patent No. 4,825,391 The "Quote Processing of Depth Buffer Priority of the Immediate Generation System .. and other publications, :: Integrated and here for readers' reference. Yu Youyuan This system includes a controller unit 12, a geometry processor unit Not the processor unit 16. The display processor unit 16 displays information to determine and display the movement of the aircraft on the scene. 3 The recruit enters the input control interface. 2 2 Connects to the computer image generation system. Feed to the input control interface: the block line control interface cries related flight data such as aircraft position: input jitter and yaw posture are transferred from the pilot to the controller 丨 2. Yi, the Ministry of Economy ordered Central Standards Bureau employees to eliminate Printed by Fei Cooperative '' seed coat —-(please read the notes on the back before filling in this page) In the controller 12, the pilot input is used to calculate the new and direction of the aircraft ' It is well-known in the mathematical rotation matrix used by the reference: two systems to display the rotating objects of the coordinate system. Snatching corpses, other information about the position of the target or other aircraft are calculated in the controller ^ generated in certain computer images. The controller unit is a general computer.-The controller unit 12 is regularly updated with the latest aircraft orientation. This orientation is reached by the pilot through the input control interface 22 to reach the controller 12 including the rollover, bumping and yaw of the aircraft, and position Etc. The controller 12 processes the scene in a frame or frame time and then transfers the data to the geometry processor M. The geometry processor 14 reads from the scene database memory 24 in the three-dimensional digital representation of the stray ground The description of the objects that may be seen such as towers, roads, Hesanpeng Factory, etc. The objects are read out from the scene database and transferred to the display coordinates using the rotation matrix of ## calculated in the controller 丨 2. The geometric processor 14 cuts all the edges beyond the window boundary in the objective two rooms. It counts the three-dimensional object data in 10 paper scales. Guxi ox29 ^ V. Invention description (8) A7 B7 Printed by the Employee Consumer Cooperative of the Central Standards Bureau of the Ministry of Economy The system is mapped to the two-dimensional display window. The surface of the object beyond the boundary of the window is cropped in the two-dimensional image workshop. The dimensional object contains points, lines, closed convex polygons, or a combination thereof. Each closed convex polygon is the surface Because of &, each object processed consists of individual surfaces, where each surface is bounded by a straight edge. Once the surface is calculated from the three-dimensional object in the geometry processor 14, this data is passed to the display processor i 6. In addition, the geometry processor ^ 4 uses known methods to calculate which surface is before or after which surface and stores this information in the priority list. The priority list contains the order of all the surfaces in the scene, with the first surface in the list as the highest priority surface ’and the last surface in the list as the lowest priority surface. When the two surfaces overlap on the display, the higher priority surface will be seen, and the overlapping part of the lower priority surface will be blocked. After calculating the priority list, the geometry processor 14 calculates the display end points of the line segments defining the two-dimensional surface and calculates the coefficients of the linear equations describing these boundary edges. Geometric processing 14 Calculate the endpoint and slope of each edge of the surface defined by the island. These resources are then passed to the display processor 16. In Fig. 1, a display processor 16 for driving the video display 18 is shown. In some systems, a single controller and geometry processor 1 * drive more than one display processor. The display processor 16 represents most of the processing and hardware in the real-time computer image generation system. Therefore, efforts to reduce the size of hardware and processing time are mostly concentrated on the display processor. The display processor is responsible for processing the information displayed on the video display 18. The display processor 丨 6 must analyze what color it should be for each pixel on the video display 18, and give many surfaces that may appear on a pixel, and the -11 paper size on the surface. The scale is suitable for China National Building Standard ( CNS) a4 specification (210X297 mm) --------- A —-. (Please read the notes on the back before filling in this f) Order * n I · I f —. 318923 A7 ~-__B7 5 2. Description of the invention (9) Some parts that may be covered by other surfaces. After resolving the contradiction of priority order in pixels, the area of each visible surface in the pixel is multiplied by the color of its surface 'and the sum of all these weighted colors is output as the pixel color. The display processor 16 receives the two-dimensional surface description from the geometry processor 14 == surface color, surface priority list, and uses this information to output the scene to the color visual display 丨 8. Each color display in the system can be composed of 1024 or more video & signals, and each line usually has 1 color point or pixel (picture element). Since there may be more than one million pixels on the display 'and all these pixels must be calculated in 1 / 6O seconds, the processing load in the display processor is very high. Each pixel may receive 12-bit red data signal, 12-bit margin color data signal 'and 12-bit blue data signal as part of the color intensity information. To describe pixel colors. Printed by the Employee Consumer Cooperative of the Central Standard Falcon Bureau of the Ministry of Economic Affairs (please read the precautions on the back before filling in this page) Referring to Figure 2, the observer's eye 20 is looking at the objective space 30 'on the window at the viewing point Vp Polygon sampling surface 30. Only the surface 30 that is closest to the viewer 20 'along the line of sight 32 will be seen; the visible surface is compared with the point P of the viewing point vp by using the depth buffer technique known to the artist The range distance r between f is determined. Therefore, the visible polygon 30 (in this case, the surface identified by the vertices Va, Vb, and Vc of the triangle) is the polygon with the smallest range Rmn along the line of sight 32. The objective space 30 'is projected onto the image space 34, the middle window 34; the polygonal surface 30 is projected to display the polygon 36, the individual vertices Va, Vb, and Vc correspond to the vertices Vi, V2, respectively. With V 3. The display window is often the raster of the display. Each pixel is light in a sequential manner along the horizontal line (in the J direction). This paper size is applicable to China National Skirt Standard (CNS & A4 specifications (210X297mm). Ο丄 8923 Α7 Β7 V. Description of the invention (10) Scanning from the left edge of the grid / window (J = J.) To the right edge (max), and from the yellow edge of the light / window (at the 0 edge) to the bottom edge (iv) Line scan with increasing distance (in 1 direction). Therefore, the Pth line Lp is scanned from the first (left hand) pixel Pi to the last (right hand) pixel Pr by scanning 3 pixels pp by the line of sight 32 passing through the window plane 34, so the pixel pp includes the image point , And the point Pf3 that is relatively mapped to the polygon. The polygon 36 will include all pixels along the straight line Lp and the leftmost pixel p0 at the position J1 that crosses the left straight line to the rightmost pixel Pq at the parental position JR of the right line. Enlarge the part of the image space under consideration, as shown in Figure 2a, the pixel rows 38 along the scan line 34 "are scanned sequentially in the j direction, starting with the most adjacent pixel 3 o 'then operate the next pixel 38i' and After the pixel reaches the pixel on the scan line that is on the left edge of the polygon and is partially covered by the polygon 36. This pixel is regarded as the edge of the polygon and the line 34 " the leftmost intersection position JL. Continue along this line and pass other The pixel j 8 that is completely covered by the polygon reaches the pixel 38r that is on the right edge of the polygon on the scan line and is covered by the polygon 36 6: this pixel is regarded as the rightmost parent fork position JR with the edge of the polygon. With the j-line 34 " advancing to reach the rightmost pixel 38v of the scan line. Without considering the transparency and chroma of the further polygon, it is completely covered by the polygon 36, which is completely in it, the property of the pixel 3 8 It is the properties of the polygon surface (color, color intensity, transparency, etc.), and when the pixels (such as pixels 3 8 丨 and 3 8r) are partially covered by the polygon 36, these specific many The attribute of the shape only represents the part of the pixel that is covered by the eye. Therefore, the attributes of the rest of the pixel must be obtained from other information and based on the various types of pixels covered by each polygon / color information source. Standards are applicable to China National Standard (CNS) A4 (2 丨 Οχ297mm) Yi Yongxian® Read the notes on the back and then fill out this page) The Central Skirt Bureau of the Ministry of Economic Affairs of the Ministry of Economic Affairs Cooperative Printing

Binding ------- __ 318923 A7 B7 The Ministry of Economic Affairs Central Standards Bureau Employee Consumer Cooperative printed five. The description of the invention (11 percentages are mixed. This can be understood with reference to Figure 3 a, where flute, ^ ,, A first A polygon 36a comes before another polygon 3 6 b (as shown by Ren Fu τ & 1 1T well-known depth buffer processing), as seen in image space. Polygon 36 has multiple vertices (such as four centroids W, V2 , V3 and V4) are connected by the same polygon edge% as edges 39a, 39b, 39C and 39d) through the pixel boundary at the intersection point 40. The image screen is in the form of a rectangular array of pixels, along the horizontal lines of the horizontal edges (for example, pixel lines ... L'L-2 'L-〗, L, L + 1, l + 2, ...) are sequentially arranged in the vertical block The lower boundary of each pixel line L of the next higher number (to be scanned next) on the bit (eg, field ..., p_2, p_1, p, p + 1, p + 2,). In this way, each pixel can be seen in the order of its column L and column P number 3. Some pixels can be seen, such as the pixel (L, P) 'is only completely covered by a polygon (polygon 36 &) and therefore only the attributes of the polygon: Other pixels share / extract multiple polygon attributes. According to one direction of the present invention, as shown in FIG. 3b, we prefer to consider the entire pixel without having to process the color information of multiple sub-regions (or sub-pixels) of each pixel to determine its color attributes, even if the aforementioned U · S. Patent No. 4,825, 39 1 depth buffer processing architecture (whether or not using wingspan / aileron wings processing method). It should be known that each information item processed in our method is a signal word block of electronic digital data; and we may discuss these signals using our own names (such as corners, edges, polygons, etc.), although we still use these The noun refers to the data signal used in each individual geometric concept (for example, "corner" refers to the data word group describing the corner position of the two-dimensional pixel, and "edge" is the data signal group representing the edge line of the relevant polygon. Paper standard suitable financial standard (CNS) ~ secret (2iGX297 mm) I. m 1 ^ 1--I: · --- 1 In) '* ^ I -I 1 ---- II--1 — «^ 1 Τ», τ (please read the precautions on the back before filling in this page) A7 B7 5. Description of the invention (12) "Polygon" refers to the digital data structure equivalent to a specific polygon, etc. Our processing process The end result is to provide the final sequence of video signals (digital video, or analog video deduced by digital) so that the color and intensity of individual pixels on the video display device i 8 are visible to the observer 20. Polygon 36 is sampled in the pixel interval ; Here, the sampling point Are the four corners C of each pixel. For each pixel (1, P), a 4 × 4 array of corners can be used (for example, corners C11 to C14 are on edge E-2, and corners C21 to C24 are on edge E-1 Upper, corner C3u, C34 on the edge [upper corner ^ 丨 to ^ 4 on the edge E + 1) to define a 3 x 3 array of pixels (then consider the 8 adjacent pixels surrounding the central pixel L, P): Therefore, the pixel array includes pixels (L-1, P · υ, (Lq, P), (L], p + 1) along the straight line (1 ″), and pixels (L, P-1) along the straight line (1), (L, p), (L, P + 1), and the pixels α + 1, P-1), (L + i, P), (L + 1, Yang) along the line (L + 1). The color that falls on the corner (sample) point C ij (where i is the line name of the relevant pixel and j is the stop name) closest to the viewer ’s polygon 36 ”is stored for each pixel of the display. The formed color data signal is stored in the video memory containing the color information of each pixel that constitutes the display. The color information of the sampling point C in each corner of the video memory is calculated by the traditional method familiar to those skilled in CIG. We As long as the color information is obtained from one corner of each pixel (for example, the first touch of each pixel in the I'j display, or the upper left corner); when the data for the upper left corner C 〇 of all adjacent pixels When you are ready, you can determine the% full color information of any specific pixel. The area of each pixel represented by the color attribute of a corner can pass through the intersection of the relevant polygon edge and the point 4 〇 falls at the corner_-15 · The size of this paper · Mianguanjia County (CNS) M inspection (21QX297 Gongchu-~ '' ~~~~~~-I ^ i I —1 .—installed-(please read the notes on the back to take, item and then Fill in this page)

* 1T Printed by the Employee Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs Printed by the Consumer Labor Cooperative of the Central Bureau of Accreditation of the Ministry of Economy 18 ^ 3 A7 _ __B7 V. Description of the invention (13)

The landing point C 〇 is determined by the knowledge of the arrangement of the line segments in advance: here we use 4 line segment arrangements to pass through " 5 corner arrangements (the central corner C 〇 and the four corners about the central pixel corner C 〇 Point). According to the clockwise position of the 12 o'clock position of the corner C 〇, the four line segments are arranged: the upper line segment between the corner C 〇 and the CT on the previous edge (E-2) immediately above it ST; the right line segment SR between the corner C 〇 and the CR immediately to the right on the scanning edge (E-1): between the corner C 〇 and the CB immediately below the next edge (E) Line segment SB: and the upper line segment SL between the corner C 〇 and Cl on the same edge (E-1) immediately to its left. The distance D, where X is T, R, B or L, is determined by the pixel corner C 〇 to the relevant intersection point 40, the upper, right, lower, and left line segments are determined by the processing and methods well known in the art, and the distance Dx data signal is stored in the video memory. In addition, for each The transparency identification bits (which can be defined as clear == 〇 and opaque = 1) of the corner sampling point C 〇 are also added to the video memory Referring now to FIG. 3c, if the same polygon covers the pixel sampling point C, the edge of the polygon is used to determine the position 40x between each vertical and horizontal edge passing through the pixel sampling point and its four adjacent sampling corners CX arrangement. In this way, the edge crossing positions 40t, 40r, 40b, and 401 are determined and all relative distances DT, DR, DB, and DL are calculated and stored in the video memory. However, if the edge crossing distance Dx is greater than the distance between corner sampling points Then, the distance Dx is constrained to the distance between sampling points. The distance between the corners of adjacent arrays can be set to a normalized value 1.0: then the positive distance Dx is normalized to a value between 〇〇 and 1 _ 0. As shown, the distance DT from the central corner C 〇 to the edge crossing position of the middle and upper corner CT is 40t, which is less than 10 (corner to corner -16- clothing paper size is applicable to China National Standard (CNS) A4 specification (210X297 Mm) -------- '1 installed ------ ordered (please read the precautions on the back before filling in this page) A7 B7 Ministry of Economic Affairs 4-Central Bureau of Standards, Male Worker Consumption Cooperation Du Yin System V. Description of the invention (14 complete distances) and count The calculated distance e and distance are stored as a data signal. The intersections at the central corner Co and the opposite right and left sides are the same. The distance DR and DL are also the same as 40. Because each normalized distance is less than the corner to The distance of the corner is "but it is not the case for the intersection and position under the adjacent corner cB" and therefore has a distance greater than-: the distance value is constrained to 10, indicating that the polygon exceeds the corner C (^ CB. That is, if The distance 0 is equal to the distance between the corner sampling points, and the polygon on the central corner point is also regarded as covering the corresponding adjacent sampling corner. The number of bits used in the data distance Dx is between 3 and 5 bits, but it can be combined because = and unified. Therefore, each corner sample point C is the address of a separate location in the video memory. The location contains data describing the color, transparency markings, and distance data related to the left, right, and fork positions. If the polygon at the sampling point is transparent, the transparency flag is also set.

Since the screen distance generally increases from left to right and from top to bottom, we currently prefer to assume that the distance between the corners of the sampling arrangement is 1 · 〇 and the intersection and position of the right and bottom if the intersection and position are 10 or left and above If the intersection distance is 0.0, the polygon covers adjacent sampling points. Therefore, simple distance data word truncation and constraints will produce a distance between 0.0 and 1.0 from left to right and top to bottom; if the top or left intersection and the position is 0.0, the polygon covers the top or left phase Adjacent corner points, and if the lower or right intersection and the position is 1.0 ', the polygon covers the lower or right corner points. This is just a simple example, and it can vary depending on the specific use if needed. Various methods for calculating the distance between edges and positions are widely known in computer graphics and must be widely used by artists who are accustomed to this. D • 17 This paper size is applicable to the Lai transfer standard (CNS) A4 ^ ⑺Qx297 (public) IIII-I-I '--I-n II (please read the notes on the back and fill in the text) A7 A7 Printed by the Employee Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs V. Description of invention (15) π Refer to the figure now 4. It shows the demonstration of the color intensity of the video output data signal for processing the top-point data signal input to request the display π to the viewer 20. σ 16a, in which the procedure of the flowchart in FIGS. 5a-5e is used. The first input Maq of the sub-combination is paired with the first input 42 & of the edge occlusion computing device 42, and the remaining display processor 丨 6 receives the vertex data of the display polygon in the (1 ,;) coordinates Signal: The output 42b of the computing device is a set of normalized edge intersection data words for the specific pixels that are then processed. Pixel (Lp 'Pp) is described by the device input number 42c of the pixel input 42c from the sub-combination input i6a_2, and the display line L p number data of the device input from the sub-combination input 6a_3 2d 3 appears in the frame The first input of the buffer (mass memory) device is a 4-bit edge data signal, and there are countless pixels on the display, each with a storage location of many bits deep: each position consists of 44c and 44d lines and pixels The field data word group is uniquely specified (that is, by the line J〇s Lp s' μ and the field Xia. Guest PP s Lax). The frame buffer is also coupled to the input 44e to receive the polygon color data signal and the sub-combination interface 16a-4, and to receive the transparency 'shell material signal and S1〗 combination interface 16a-5 at the input 44f. The frame register input stage 44a 'stores all information such as the color / transparency of each pixel, the intersection of the edge and the logo on a per-pixel basis. This data is also provided to the frame register output 44b on a per-pixel basis: It must be known that since the current buffer device displays only a single set of pixel designation inputs 44c / 44d, there is only the next pixel to be addressed It will receive input data and supply data at output 44b. At the same time, it is necessary to provide multiple groups of pixels when the stored data is taken out and output by one (or more) different pixel positions independently described.-18 · n -I-i: I-.-IIJ, J,--I — I—-n. U3., T (please read the precautions on the back before filling this page)

318923 Α7 Β7 5. Description of the invention (16) Name input allows input data to be loaded into one (or more) pixel locations. The frame buffer output 4 1 b provides the digital data signal of the selected anti-aliasing corner and edge arrangement (here are 12 edges of the 4 adjacent pixels of the central pixel (L p, pp) that are processed next) One input 46a 'of the post-processing device 46' row line and pixel field data signals are provided to other inputs 46c and 46d. The edge register is accessed by the edge parser 46_2 to determine the portion of the currently processed pixel covered by the color / transparency properties of a polygonal surface: this data is then used by the color mixer 46_3 to set the current pixel (LP'P p) the final mixed color and pass the data by post-processing output 46b 'δ into a sub-combination output 16 & · 6 video output data signal (can be displayed or further processed if necessary). Printed by the Ministry of Economic Affairs, Central Bureau of Standards and Staff's Consumer Cooperatives-(Please read the precautions on the back before filling in this page) The processing steps performed by the edge occlusion calculator are shown in the flow chart in Figure 5a. The data signals of L p and the rest P p are received in step 50 and the vertex data signals of polygon 36 are received in step 51. In step 52, at the beginning of the display line L loop, the calculator determines which display line L from the top scan line (where ι = Ι〇) to the bottom scan line (where I = Imax) is at least one pixel Shao Fen appears a polygon; these scan line numbers L are temporarily recorded and the first number LP of these scan lines is set. The scan line number is entered in step 53, where the left and right polygon edge column lines of the scan line are found to block JL and Jr; these limits are also temporarily stored. Step 54 starts the nested pixel loop from the boundary (left to right) found in step 53, considering each pixel j p established along the scanning line l p in step 52. Steps 55, 56, 57 and 58 can be found inside the nested loop. Do the following: find the four cross positions c χ and the related four positive -19s arranged in this corner of the pixel's upper left corner C 〇 The Zhang scale applies the Chinese National Standard (CNS) Α4 specification (210 × 297 mm). Printed by the Employee Consumer Cooperative of the Central Standard Falconry Bureau of the Ministry of Economic Affairs. 20. 5. Description of the invention (17) The standardized distance Dx; limit the distance to the maximum pixel value of 1, If the distance a DX is greater than 1 pixel distance, then set the "overlay" mark (step 5 5) 叱 = polygon is transparent, then operate the pixel data in certain directions (transparency test: will be discussed in detail below); corner The color of c〇 is regarded as multi-sided: color, texture effect (color, frequency modulation, and the like: the function of the difference is calculated (step 57); then the relevant data signal is temporarily stored in the video memory (frame buffer 44). Then, go to step 59; Judgment: Will the domain of next-order pixels exceed the right boundary Jr? If not, you can perform further pixel processing on the Lp scan line: use the output and go to step 60, and loop back to Bu +1. Return to step 54. If the current pixel is in JR_h, the next pixel does not contain the current polygon, & output do not exit the loop to continue processing-scan line L = (Lp + l). Enter step Η and scan line Loop judgment: Next—Will the sequential scan line exceed the lower scan line limit of the polygon, "? If not, you can do this. Advanced pixel processing on the scan line: use output 61a and go to step 62 so that the scan line loop returns to step 5 2 with L = L + 1; then it will pass the new scan line The pixel loop of "If the scan line just completed is on the bottom of the polygon, down-scanning, the spring does not contain the target polygon, and the output 6 1 b exits the loop to continue processing the next polygon 36. When the current polygon processing is completed, the step is entered and the polygon loop determines whether all polygons in the current window 34 have been considered? If there are still other polygons. To process, select the exit 63a, advance the polygon index in step 6 and grab the set of polygon vertices (step 5 1). If other polygons must be considered, the display frame does not require further pixel processing in a part of this system; exit from output 63b

The size of this paper is (CNS n —1 nnnnn --- 'i ^-—— II! I 丁 * ye (please read the precautions on the back before filling in this page) A7 B7 V. Invention description (18) to continue Process the data signal of the next image window frame and start the calculation of the edge occlusion of the pixel data of the display frame. Transparency The polygon surface 36 has a certain, known transparency. We normalize the transparency so that it can be used The number between 〇〇 (completely transparent) and 丨 〇 (completely unknown) represents this transparency. Transparency is achieved by ignoring or disabling the sampling point of a particular pixel corner, so that even if the polygon falls on (ie, obscures) the sampling corner point In the above, the polygon characteristic data will not be written to the video memory of the corner of the pixel: the edge intersection position related to this ignored corner will also not be written to the memory. According to the transparency number, the sampling point is invalidated by the way of invalidating the sampling point. You can see the polygon visually behind the transparent polygon. The more sampling points that fail, the more transparent the polygon. This is like poking a hole in a transparent polygon See what's behind. A style selection code and transparency value will be assigned to the polygon. The transparency value indicates how many sampling points to invalidate and the style selection indicates which sampling points to invalidate. Central Bureau of Economy, Ministry of Economic Affairs, Beigong Consumer Cooperative印 袈 I! -I II! II-»1! '' 士 ^ II--I II 丁 U3. .-¾ (please read the precautions on the back before filling in this page). For failure treatment, please refer to Figure 6a. It shows five different degrees of transparency τ for the two style selection (PS) bits. It can be seen that for different degrees of T degrees, the transparency failure style of the bit selection at the low binary degree (ps = 〇) style is In the high binary degree (PS = 1) style, select the bit complement. Although only five levels of transparency T 'are shown, more transparency can be calculated. The number of D-levels is completed by modifying the edge intersection position, as shown in the figure As shown in 6b-6e, just as a function of transparency. The more transparent the polygonal surface, the fewer sampling points assigned in the area. This procedure increases or decreases the polygonal area. 21-This paper scale applies the Chinese national standard Standard (CNS) A4g (210 X 29? Mm) 318923 A7 B7 printed by the Employee Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economy V. Invention description (19) ~~-Sampling corner points in: as seen in Figure 6b, transparent The degree τ is very low (0 < Τ < 1/4) only one pixel corner Ca is in the selected style, while the other corners cb, Cc, Cd of the same pixel are invalid and are not considered; the left-right intersection is moved to w to Sampling point, so the distance 1> 1 ^ and 1 :) 1 ^ become i 〇 and 〇_〇 'and the corrected bottom intersection position distance D, Ba becomes (4 * T * Dh), where DBa is uncorrected The bottom crosses the location 4〇 "to the distance from the fall again. Similarly, the position of the top intersection is 4 °, and the distance DTa is corrected to D, h = 丨 _ (4 * τ * (〇 · DTa)), where D is the distance of the pixel edge segment. When the degree of transparency increases, the edge intersection position is moved away from the corner sampling point, thus increasing the effective area of the polygon surface in the pixel under consideration. · Figure 6c shows the situation when 1/4 s T g 1/2: a pair of pixel corners (here corners (3 and Cc) are used and a pair of corners (Cb and Cd) are invalid and are not considered . For the top corner point Ca, the left and right forks are still moving forward to the sampling point. Therefore, the distances DLa and DRa are still 1.0 and 0.0, respectively: the distance between the top and bottom sense forks DTa and DBa is not modified => The bottom corner point cc, when the corrected right-hand intersection and position distance D, Rc becomes (4 * (T-1 / 4) * DRc), where dRc is the uncorrected corner C c bottom fork position distance, top and The bottom intersection position distances DTc and DBc become 1.0 and 0.0, respectively. Similarly, the left intersection position distance DLc is corrected to D, Lc = l- (4 * (Tl / 4) * (D-DLc)). Fig. 6d shows the situation when each of the 1 / 2ST is 3M: three pixel corners (here, corner Ca, (^ and Cd) are used and only one corner (cb) is invalid and not considered. The top corner point Ca is about The intersection position distance DLa and DRa are still 1.0 and 0 · 0 '. The top and bottom intersection position distance DTa and DBa are not modified. The left and right intersection point position of the bottom corner point Cc is maintained at DRcNo-22- This paper scale is applicable to China National Standard (CNS) A4 (210X297mm). Packing-f Please read the precautions on the back before filling this page j. Βτ Employee Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs印 褽 318923 A7 A7 * B7 Fifth, the description of the invention (20) When changing, the distance between the top and bottom X fork positions are maintained at 1 _ 0 and 0.0. C d The distance between the corners is corrected. The distance between the right and left cross positions is limited. The value of i 〇 and 0.0. When the bottom parent position distance Dgd becomes (4 * (Tl / 2) * DBd) 'where DB d is the bottom cross position distance that is not corrected. Similarly, the top cross position distance is corrected by DTd Become C, Td = l- (4 * (Tl / 2) * (D-DTd)). Finally, Figure 6d shows the situation when 3/4 contains TS0.999: because the use of all four corners implies T = 1 is opaque, so only three pixel corners (negative drop Ca, (^ and ^) and one corner (Cb) are invalid; only checking the validity of corner C b is to determine that it is fully opaque (T = 1. 〇). The distance DLa and DRa at the top and bottom corner points ca are corrected to D, La = l (4 * (T-3 / 4) * (D-DLa)) and D_Ra = (4 * (T- 3 / 4) * (D-DRa)): The distance between the top and bottom intersections DTa and DBa is not modified. When the left-right intersection and bottom position distance of the lower left corner point Cc is maintained at DLc and DRc, the top and bottom intersection positions The distance is maintained at 1 · 0 and 0 · 0, respectively. When the top and bottom intersection positions of the bottom right corner cd are maintained at their original values DTd and DBd, respectively, their right and left intersection position distances are still limited to i.〇 and 〇.〇, respectively. The effect of the procedure discussed above is to increase or decrease the polygon area that makes up the sampling corner points, as shown in Figure 6f (PS = 〇) and Figure 6g (PS = 1). For PS = 〇 'If the transparency value T is between 0 and 1/4, the area where the sampling angle falls Ca increases as τ increases (more opaque), as indicated by the arrow xa. If 1/4 transport T < 1/2 ', the area of the sampling corner Cc increases when T increases, as shown by the arrow &' If 1/2 S T < 3/4, the sampling corner Cd when T increases The area increase is shown by the arrow Xc. The arrow Xd is the increase of the sampling corner C c in the clockwise direction. Contrary to the increase of the CCW of other arrows, if it is T> 3/4. PS = 1 得 -23- n J-· I «--... * " -II---..... 1 IT * * τ (please read the precautions on the back before filling this page) 1. I--Printed by the Pongong Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs A7 --------- B7___ _ V. Description of the invention (21) ~ To the opposite change, as shown in Figure 6g. The area covered by pixels of each polygon can be randomly assigned a style selection configuration, which will help to infer the color information of the entire pixel based on the transparency, priority order and color of all polygons that violate the child pixels. Post-memory processing After solving the color and transparency considerations of the corners before the frame punch, a sequence of shell material signal correction steps 7 _ 8 〇 in the post-processor-edge analysis for processing 'process shown in Figure 5b. These steps can be performed in a time-independent or time-dependent (preferably interleaved) manner using the edge intersection position / corner lean signal correction step 5 _ 6 5. No matter what kind of pixel information is used in the frame buffer: Therefore, if the occlusion calculation becomes a dilemma or slow, the data signal stored by Xianke will be used a. Edge analysis processing starts at step 7 and is accessed by The frame buffer 4 4 obtains pixels (L p, p magic corner C〇 arranged kidney material signal, which is arranged along the four sequential edges (E + 1) to (E_2) sixteen corners C11_C44 (Figure 5b ,) Related. _ Set of twelve corner-to-corner edge segments S0-SU set in step 71, each generalized pixel edge segment Sx (Figure 5b ") stretches between a pair of corners (^ and 〇 ^ and each The corners have X and Dw distances associated with them. To obtain each edge, first enter the judgment step 72 and make an initial judgment whether both Dv and Dw cover the edge segment sx. If so, normalize the edge segment The length 1 must be 0.5 (step 73); if not, step 74 judges whether it is ~ two distances from: :: less than the distance SX3. If it is, the length of the edge segment is the relative distance of the corner considered (for example, for corner Cv, L = Dv), as set in step 75: if not, go to step 76 to check the two The transparency of the corners. If there is only one corner _ -24- This paper size is applicable + national standard YcNS) A4Di · (21〇 · 〆297 Gongjuan〆 '(please read the precautions on the back before filling this page )-Binding · Order 318923 A7 B7 _ 5. Description of the invention (22) (Please read the precautions on the back before filling in this page) Fork position is transparent, step 7 7 Make the edge segment length 1 equal to its relevant distance and set the transparency sign , But if both intersections and positions are transparent, step 7 8 sets the edge segment length 1 to the average of the distance and Dw. All steps 7 3, 7 5, 7 7 and 7 8 exit to step 7 9, where the edge segment The number Sx (0 SS g 11) is used to check whether the pixel has other edge segments that need to be processed. If yes, step 8 0 increments the edge segment number S and returns to step 71: if not, the edge segment information data signal is prepared Passed to the color mixer. Printed by the Employee Consumer Cooperative of the Central Department of Economics of the Ministry of Economic Affairs. Now referring to FIG. 7, the edge registers and edge parser parts of the post-processor device 46 are 4 6-1 and 4 6-2 and can be used. Special logic devices replace specially designed calculations In the input bus 81, the corner data input signal from the frame buffer is received. The three first-in first-out data registers 82a-82c are connected in sequence so that the four sequential edge data signal strings are temporarily different Appears on each of the bus bars 8 1 a-8 1 d. The data at the corner on the lower edge E of the current pixel scan line L will be on the bus bar 8 1 b, while the register device 82b is already on the bus Row 8 1 c makes the stored corner data signal of the immediately preceding scan line available and thus provides an edge (E-1), and the register device 82c has made the stored data on the bus 8 1 d The data signal of the edge (E-2) before the edge (E-1) can be used. Then the lower-edge (E + i) becomes available at the bus 8 1 a: therefore, it will be seen that each pixel P on the scan line l only the data at the next sequential edge (E + 1) is stored And it will be processed after it can be used, it needs the availability of the next scan line (L + 1). In any case, the corner nodes C 4 4, C 3 4, C 2 4 and C 14 reservations will be input (next clock) to the next current information in the reservation register device 8 4 'eg: temporary The memory device 84a has a corner node C43 -25 at its output. This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) A7 ____B7 5. Information of the invention description (23); the temporary device 84b is in its The data of the corner node C33 is output: the register device 84c outputs the data of the corner node C23; the register device 84d outputs the data of the corner node C13. Two sets of next-sequence registers 85a-85d and 86a-86d retain other pixel data of the next sequence to provide for different corner nodes C42, C32, C22, C12 and C41, C31, C2 1, C1 1 respectively Information signal. Paired with appropriately related corner data signal nodes is one of the twelve edge segment computing devices 88a-881, each of which provides the calculated edge segment length 1 data signal at its associated output 88a'-881 '. Therefore, the 'S〇 edge segment length] _〇' data signal is provided by the computing device 88a connected between the nodes C12 and C22; the S1 edge segment length 1S1 data signal IV is provided by the computing device 88b connected between the nodes C13 and C23: 52 The edge segment length iS2 data signal] _2 'is provided by the computing device 8 8 c connected between the nodes C21 and C22; the S 3 edge segment length 3 data signal L' is provided by the computing device 88d connected between the nodes C22 and C23; and Continue on until the length of the edge segment of the su is the data signal] provided by the nose gauge device 8 81 connected between the nodes C33 and C43. Then all twelve edge segment length data signals of the pixel (l p, p p) currently processed are sent to the color mixing device 4 6-3. Printed by the Beigong Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs-(please read the precautions on the back before filling out this page) In order to display the appropriate color information for each pixel 36, 1 6 (4 X 4) phases The Zheng sampling corners C i! -C44 (see Figure 5c) are taken out of the video memory: the sampling corner points around the pixels (L p 'P p) will be calculated. The final goal is to mix a set of four corners to mix colors B 'such as the upper left (ti) corner ct i mixed color Btl, the upper right (tr) corner Ct # mixed color bu, and the lower left (bl) corner Cbl The color Bbl and the mixed color Bbr of the lower right (br) corner cbr, each of which is sampled from the respective corner Cjj and the color Cij, is calculated by: -26- This paper size is applicable to China National Standard (CNS) A4 specification (210X297mm) A7 318923 5. Description of the invention (24)

tl_blend_color BfCCH '+ C ^ + C ^ + CjJM I nnnnn II \ n- n In n II— T * T (Please read the precautions on the back before filling this page) tr_blend_color Btr = (C13' + C14, + C23, + C24,) / 4 bl_blend_color Bbl = (C31, + C32, + C41, + C42,) / 4 br—blend_color Bbr = (C33, + C34, + C43 '+ C44') / 4. Referring to Figures 5c '5d and 5e, the color mixer includes small polygon processing (Figure 5d) followed by color calculation processing (Figure 5e). Small Polygon Filter When most polygonal surfaces 36 are large enough to cover multiple pixels 38, it is possible for the polygon to cover only a single pixel. To some extent, it is possible to make the polygon too small to completely cover one pixel. .Polygons that are smaller than a pixel 'or whose size is less than the distance D of the sampling point of the corner, will blink ‘because at some locations, the polygon falls on the sampling corner and can be seen, while in other positions the polygon falls on the sampling corner Can't see from time to time. Small polygons can be detected and filtered (removed or erased) to reduce flicker. To do this, each of the four corners Ctl, Ctr, Cb, and Cbr of pixels 3 6 must be printed independently by the Employee Consumer Cooperative of the Central Bureau of Standards of the Ministry of Economy. The mixed value is calculated according to the size of the polygon covering the corner sampling point. The mixed color is formed by mixing the colors of the sampling points. The smaller the polygon, the more mixed colors are used to replace the sampling point colors. The length L of the intersection of the edge segments is used to calculate the mixed value of each corner. This can be done using the following procedure, using the corner Ctl as an example of step 90a to calculate the mixed values representing various distances. First, find the vertical length] _vert and horizontal distance] _horz (step 9 1): Then determine the first and second mixing coefficients Blend ”and Blend_2 (step 92), and take the smallest one (step 93) to obtain the Blend_tl coefficient. Compare the Blend_tl coefficient with 1.0 and normalize it (step 9 4), if necessary -27- This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X 297 gram) Ministry of Economic Affairs Central Bureau of Precinct Employee Consumer Cooperative Printing Α7 Β7 5. Description of the invention (25) and limit it to the maximum value (step 95). Step 96 averages the colors of the above four corners, and then Step 9 7 determines that X = 0, 2, 3, 4, 5, or 1 〇, that is, in the edge segment S0, S2, S3, S4, S54S1 () (edge segment If the extension will pass through the corner cu) whether there is any intersection position distance is obtained by the transparent polygon. If yes, the mixed color value is ignored (step 9 8) and set to 0 ”If no transparent polygon is involved, step 9 9 finds the blended tl color B,‘ when it becomes the last tb_blend_color value Btl. This process is repeated in step 90b to find the colors of the other three sampling points of this pixel, that is, to find the final mixed color values Btr, Bb, and Bbr for the upper right corner Ctr ', the lower left corner cbl, and the lower right corner cbr, respectively. Referring now to FIG. 5 e, the final color calculation processing of the color mixer 4 6-3 starts at step 9 2, in which the number of intersection positions where the four sides of the pixel (edge segment% ^, S6 and S8) are not covered is calculated. The edge count (must be zero or at least two) and transparency conditions are used in step 93. If there are 2 edge intersection positions and none of the four sampling corners are covered by transparent polygons (determined by the inspection identified by transparency), it is assumed that a single polygon edge passes through the current pixel: two areas must be calculated (the first polygon covers area A 1 Cover the area A2) with another polygon. This can be found in the consideration of Figures 8a-8f, which shows six possible situations of the position of the 2 edge intersection and the position: In Figure 8a, the edge 3 9 is nearly vertical and the edge segment s3 is cut into two parts (left The normalized length of the part is equal to, since the total length of the normalization of any pixel edge segment is 1, the normalized length of the right part (1-b)); the bottom edge segment $ 8 is cut into two parts (left part The normalized length of the copy ^ and the normalized length of the right part (1_U)-area AIrrO.VL + U and area A2 = 1-A1 (at -28- the paper size is applicable to the Chinese National Standard (CNS) A4 specification ( 21〇χ297mm) f Please read the precautions on the back before filling in this page} Binding · Order A7 B7 5. Description of the invention (26) In all cases), where the pixel color CP is determined by the following formula CP 〇.) (Ctl Color + Cbi color) * Al + 〇5 * (Ctr color + Cbr color in Figure 8b, edge 3 9 is close to horizontal and the left edge segment ~ is cut into two parts (B / known normalization length Is and, The normalized length of the bottom part (Bu Yi)), the right side 'Elephant Se is cut into two parts (the normalized length of the top part u and, the normalized length of the bottom 16 (1 _ I · 6))-region A 1 = 〇 · 5 (^ 5 +] _6), where the pixel color CP is CP-〇. "(Ctl color + Ctr color ) * A1 + 0 ”(Cbi color + Cbr color” A2. Similarly, in the case of FIG. 8 c, area A 1 = 0 · 3 (] _5 * L), pixel color _ CP = (CU color) * Al + l / 3 (Cbl color + Cbr color + Cu color) * A2. In the case of Figure 8d, a1 = 0 5 (1_i5) * l8, CP = (Cbi color) * Al + l / 3 (Ctl color + Cbr color + Ctr color) * Α2. In the case of Figure 8 e, A 1 = 0.5 (1-] _3) *] _6, CP = (Ctr color) * Al + l / 3 (Ctl color + Cbr color + Cbl color) * A2. Finally, for the situation in Figure 8 f, a 1 = 0.5 ((1-] _8) * (1-L)), CP = (Cbr color) * Al + l / 3 (Ctl Color + Cbl color + Ctr color) * A2. If there are no two edge intersections, or if the transparency flag is set at any of the four corner sampling points, the pixel is considered to have four separate areas A 1 'A 2, A 3 and A 4 and 4 cross position processing; must use the regional weighted good paper scale applicable to the Chinese national standard (CNS) Jing Shu (2 丨 os7mm (Please read the precautions on the back before filling in this page)-Packing. Order Printed by the Employee Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 29- 318923 A7 B7 5. Description of the invention (27) Step 95 of calculation. It is displayed graphically in Figure 8g; the calculated area is as follows: A 1 = 1/4 (lj + Is) Α2 = ι / 4 ((1 · [3) +] _ 6) Α3 = ι / 4 ((1-1.5 ) +18) Α4 = 1/4 ((146) + (148)) The final mixed pixel color CP data is: CP = (Ctl color) * Al + (Ctr color) * A2 + (Cbl color) * A3 + (Cbr color) * Α4. The mixed color C Ρ information of the pixel makes the final decision in step 9 4 or 9 5. This pixel color data signal is output to the display to be presented to the viewer. 7 〇3 Here is a detailed description of the decision pixel in the computer image generator Preferred embodiments of the apparatus and method of color content, many changes and modifications will be apparent to those who are accustomed to this. Therefore, we will only use the content of the attached patent application project to limit the scope and not use the specific tools and procedures explained here. n ^ — — ^ 1 ·-I- I-m Ϊ-m ^ i I— I mm-^ (please read the precautions on the back before filling this page) Printed by the Ministry of Economic Affairs Central Standards Bureau Unemployment Consumer Cooperative -30 -This paper scale is applicable to China National Standard (CNS) Α4 specification (210Χ 297mm)

Claims (1)

A8 B8 C8 ------------- D8 printed by the Beigong Consumer Cooperative of the Central Bureau of Economic Affairs of the Ministry of Economic Affairs VI. Application for a patent ® '' — i • A video material in response to a computer image generator The method of signal output to determine the last visible color of each pixel of the pixel array to be displayed includes the following steps: D (a) First, receive each of at least two vertices describing each of at least one polygonal surface to be displayed A set of display space (I,) coordinates of the input data signal, each pair of different vertices defines different edges of the relevant polygon surface: (b) In response to the received vertex data signal, generates a description of each pixel array along the limit display The data signal of the intersection of the relevant edge of each edge segment of each pixel with the edge of each polygon surface, if any: (c) Based on each pixel, store the intersection of the edge segments of all polygon surfaces forming the pixel And the data signal 'and the color data of each polygonal surface occupying any part of the pixel; (d) For each pixel to be displayed, process the selected multiple phases The stored data of multiple different edge segments arranged adjacent pixels to obtain the color intensity data signal of each corner of each displayable pixel: and (e) mixing the color intensity data signals of all corners of the currently processed pixel to determine The last visible color of the display pixel. 2. According to the method in item 1 of the patent application scope, where step (b) includes step knees. • Use a set of four pixels in the selected corner of the moon as the sampling point: extend from the designated one pixel corner to the same pixel The intersection position information of the edge segment of the other corner. · 3. According to the method in item 2 of the scope of patent application, in which step (c) _____. 31-The iron standard of this paper is applicable to Zhongguanjia Standard (CNS) A4 ·· (210X297mm) '——Installed-- ---- 1T ------ 2. (Please read the precautions on the back before filling in this page) A8 Βδ C8 D8 Employee consumption cooperation of the Central Standards Bureau of the Ministry of Economy A transparency value τ between transparent and opaque for each pixel: does not store any data for τ value indicating transparent pixels; increase the number of corner sampling points of the wrongly stored data in response to the increase in transparency τ. 4. According to the scope of the patent application The method in item 3 includes the step of storing the data of all four pixel corners only when the brightness value of the pixel is totally opaque. 5. According to the method in item 3 of the patent application scope, the method further includes steps : For each of the T values whose range is less than the maximum value, provide a complementary pattern of multiple pixel corners for storing data: select a different pattern for any polygonal surface. 6. According to item 5 of the patent application scope The method further includes the step of randomly selecting one of a pair of patterns for each evening shape with a T value within a pre-selected range and associated with a specific display pixel. 7. According to item 丨The method in which the steps include the step of selecting and connecting pixels adjacent to at least one edge segment of multiple pairs of pixel corners. 8_According to the method in item 7 of the scope of the patent application, including the configuration of four adjacent pixels in the process Steps for arranging other pixels 9. According to the method in item 8 of the patent application scope, it further includes the step of selecting other arrangements of pixels immediately above, below, before and after pixels in the process. Gua according to item i of the patent application scope Method, where multiple polygonal surfaces provide the color of the pixel in processing, and further include by mixing the color density of all polygons from any effective corner covering the pixel ------------ --- ST ------; i. (Please read the precautions on the back and then fill in this page) National Standard (CNS) -32- A4 present (2 丨 0x297mm) Six ABCD Hammer Department The Central Bureau of Precinct Employee Consumer Cooperative printed 11.11 to obtain the color of the pixel under processing. The method in Item 10 of the color profit range, further includes responding to the transparency of a specific polygon before the color comes from 2: Steps for correcting the color density data of the twelve-sided revenge for each twelve revenge. There are :: The method in the " item of the patent scope, further including the margin of the step. The method of item 13 in the patent application scope, which is formed by the shape and number, in which step (b) advances to step-by-step removal step. Determines the transparency of each pixel between transparent and opaque, and responds to changes in transparency value T., Change the number of friends in the corner where the data is stored. According to the method in item 13 of the patent application scope, it further includes the step of: in response to an increase in the transparency value of the relevant polygon related to the corner of the pixel, such as the amount of large edge intersection position data, to shift the relevant edge intersection position from the relevant pixel corner Move away. 15 • According to the method in item 丨 4 of the patent application scope, further including the steps • If the distance of the edge intersection in the selected direction is greater than the distance between the sampling points', limit the value of the edge intersection position data to a predetermined maximum value . 16. According to the method in item 15 of the patent application scope, in which the display space is a display grating and includes the steps of: correcting the distance of the edge crossing position in the direction of the first optical cabinet for the selected pixel corners; for the remaining pixel corners, The distance of the edge crossing position is corrected in the second direction, which is usually at right angles to the first raster direction. 17.—The color intensity of the display processor in the computer image generator (CIG) -33- This paper scale is applicable to the Chinese National Standard (CNS> Α4 specification (210X 297 mm) -------- '1 ^ ------ Ιτ ------- ^ (Please read the notes on the back before filling out this page) Printed by the Staff Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 318923 as B8 C8 ______ D8 VI. Application The sub-components of the patent scope include: an edge-occlusion segment computing device that receives an input data signal describing a set of display space (I, J) coordinates of each of at least three vertices of each of a plurality of polygonal surfaces to be displayed, To determine (a) which edge segment of the visible display pixels of each sequential series, if any, is crossed by the polygon edge defined by each pair of different vertex data signals received, and (b) along The intersection position of the relevant pixel edge segment; The frame buffer device, based on each pixel, stores the edge intersection data signals of all polygons forming each pixel, and the color of each polygon surface occupying any part of the pixel data And the post-processor device, for each pixel to be displayed, the frame buffer device receives a plurality of edge data signals describing the arrangement of a plurality of adjacent pixels, to analyze the edge relationship and the mixed polygon color intensity data to determine the display device to receive The CIG's video data outputs the last and visible color data signal of each pixel to be displayed. According to the sub-component in item 17 of the patent application scope, the post-processor device further receives a The data signal of the transparency value of the polygon and corresponding to the received τ-value mixed color information signal. According to the sub-component in item 17 of the patent application, the pixel arrangement includes four adjacent to each currently processed pixel Extra pixels. 2〇. According to the sub-component in the patent application scope item I9, the post-processing device includes: simultaneously providing data signals to the corner data register device along the scanning lines of the raster edge of multiple displays. ; Retain the register device, connect n Ί ^ ~~ ir ------ Γ ^ (please read the precautions on the back before filling this page) A8 B8 C8 D8 patent application scope to receive each edge # line data signal to provide data signals to multiple corners on each of the multiple edge scan lines at the same time; and edge segment computing device connected to the register device To determine the length of the intersection and position on each of the multiple pixels arranged along the edge of the pixel. (Please first fill in this true copy of the legal matters of interest \ 20-Packed · Printed by the Employee Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs -35 This paper standard is applicable to the Chinese National Standard (CNS) Α4 specification (210Χ297 mm)