JP2015219351A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve image processing, such as a blur without providing a work memory in an image processor of a sprite system.SOLUTION: Pre-linking is performed with a vertically long image of the double image size of a sprite of a display object as an absolute sprite, with only α data of the sprite as a first relative sprite, and with only the same color data as a second relative sprite. A storage area for arranging the absolute sprite is secured in a sprite buffer to arrange the absolute sprite, the first relative sprite is arranged in the first half of the storage area, and then, the second relative sprite is arranged in the latter half. Then, filter processing is performed to an area where writing is recently performed, write processing is repeated in the other area as many as the predetermined number of times, and image data of an area where writing is performed last is drawn in a frame buffer, thereby achieving a blur.

Description

  The present invention relates to an image display technique, and more particularly, to a sprite type image display technique.

  FIG. 9 is a block diagram illustrating a configuration of a sprite image display system. The sprite method is a method of expressing an image for one frame by superimposing sprites (small images) prepared in advance. As shown in FIG. 9, the image display system includes an image processing device 2 such as an image processing LSI (Large Scale Integration), a pattern ROM 3, a display device 4, and a host CPU (Central Processing Unit) 5. include. The pattern ROM 3 stores compressed encoded data obtained by compressing and encoding sprite image data. The image processing apparatus 2 reads the compressed encoded data of the sprite from the pattern ROM 3 in accordance with an instruction given from the host apparatus (in the example shown in FIG. 9, the host CPU 5), and decodes (decompresses) the read compressed encoded data. The image data is restored to the image data before compression and is developed in a sprite buffer inside the image processing apparatus 2. Next, the image processing device 2 reads the image data of the sprite from the sprite buffer, performs rendering processing such as enlargement, reduction, rotation, and deformation, and draws it in the frame buffer inside the image processing device 2. The image processing device 2 reads out the image data drawn in the frame buffer in accordance with the horizontal / vertical scanning timing to be output to the display device 4 such as a liquid crystal display, and outputs it to the display device 4. As a result, an image of each frame is displayed on the display device 4. Patent Documents 1 to 3 are cited as prior art documents related to a sprite image display system.

Japanese Patent No. 3861820 Patent 39471702 JP 2001-013952 A

  The sprite type image display system is easy to display even with a low-performance CPU compared to other types of image display systems using polygons, etc., so it is adopted for pachinko and other gaming machines and video games. It was often done. In addition, paying attention to the advantage that images can be displayed with a small number of resources, it has been studied to adopt a sprite type image display system for the user interface of white goods such as refrigerators and washing machines and in-vehicle display devices.

  When a sprite type image display system is used in the user interface unit of an in-vehicle display device or white goods, blurring or glowing is prevented so that the display image is not inferior to other image display systems using polygons. It is preferable that image processing that produces a sense of depth and movement of the image, such as drop and shadow, can be performed. Note that blur is a process of blurring the outline of a sprite to be displayed. Glow is a process for adding a white and blurred border to a sprite to be displayed. The drop shadow is a process for adding a shadow to a display-target sprite.

  As a specific method for realizing various image processing such as blur, glow, drop and shadow in the sprite image display system, the sprite to be displayed is temporarily stored in the work memory, and various filter processing such as Gaussian filter processing is performed. By applying this method, a method for realizing the above-described various image processing is conceivable. However, although it is possible to realize various types of image processing with a large-capacity memory and a high-performance CPU, it is desired to realize it with a limited memory and a low-performance CPU without applying a load.

  The present invention has been made in view of the above-described problems. In a sprite type image processing apparatus, various image processings for producing sprite movement and depth can be performed without providing a separate work memory for image processing. It aims to be realized.

  In order to solve the above-described problems, the present invention provides a sprite attribute table for storing sprite attribute data and a first storage unit for writing image data of a sprite to be rendered in accordance with the attribute data of the sprite attribute table. Writing means, second storage means for storing image data representing an image of one frame, and second writing means for writing image data stored in the first storage means to the second storage means The position information of at least one sprite is stored in the attribute data of the sprite, and the position information of at least one other sprite is the relative position from other sprites other than the sprite. Stored in the data, and the first writing means is a sprite to be filtered. A storage area twice as large as the image size is secured in the first storage means, the sprite is written into the storage area according to the attribute data of the at least one sprite, and the storage area is equally divided into the first half and the second half In accordance with the attribute data of the other at least one sprite, the at least one other sprite is written, and a filtering process is performed on the most recently written area of the first half area and the second half area. The process of writing to the other area is repeated a predetermined number of times, and the second writing means stores the image data of the most recently written area out of the first and second half areas from the first storage means. There is provided an image processing apparatus characterized by reading and writing to the second storage means.

  The at least one sprite corresponds to the “absolute sprite” described in the detailed description of the present invention, and the at least one other sprite corresponds to the “relative sprite”. According to the present invention, in a sprite image processing apparatus, a predetermined number of times of filtering processing are performed on an image obtained by overlaying relative sprites on absolute sprites without providing a separate work memory for image processing. It is possible to produce the movement and depth of the image.

  For example, if the image data of a sprite is composed of color data representing the color of each pixel constituting the sprite and the transmission processing coefficient of each pixel, only the relative sprite color data is filtered. Then, it becomes possible to realize blur. A specific example of the filter processing in this case is Gaussian filter processing.

  It is also conceivable to perform filter processing only on the transmission processing coefficient in the image data of relative sprites. In this case, the transmission processing by the transmission processing coefficient stored in the most recently written area of the first half area and the second half area of the image of the same color as the image represented by the image data of the relative sprite is performed. When the image obtained by writing is written in the second storage means and the second writing means executes the process of overwriting by reducing or enlarging the image represented by the color data of the relative sprite, the glow is realized. Drop shadows are realized by causing the second writing means to execute the process of overwriting the image represented by the data while shifting the center. In these cases, a Gaussian filter process may be employed as the filter process.

1 is a block diagram illustrating a configuration example of an image display system including an image processing apparatus 11 according to an embodiment of the present invention and a configuration example of an image processing apparatus 11. FIG. It is a figure for demonstrating the prelink on a sprite buffer, and the postlink on a frame buffer. It is a figure which shows the example of storage of a sprite attribute table. 3 is a flowchart showing a flow of a decoding / development sequence executed by the image processing apparatus 11; 3 is a flowchart showing a flow of a drawing sequence executed by the image processing apparatus 11. 6 is a diagram for explaining an operation example of the image processing apparatus 11. FIG. It is a figure which shows an example of the filter coefficient in a Gaussian filter process. 6 is a diagram for explaining an operation example of the image processing apparatus 11. FIG. It is a figure which shows an example of the conventional image processing apparatus 2 of a sprite system.

Embodiments of the present invention will be described below with reference to the drawings.
(A: Configuration)
FIG. 1 is a block diagram illustrating a configuration example of an image display system including an image processing device 11 according to an embodiment of the present invention and a configuration example of the image processing device 11. In FIG. 1, the same components as those in FIG. 9 are denoted by the same reference numerals. As is apparent from a comparison between FIG. 1 and FIG. 9, the image display system of this embodiment is different from the image display system shown in FIG. 9 in that an image processing device 11 is provided instead of the image processing device 2. The image processing apparatus 11 is an image processing LSI as with the image processing apparatus 2. However, the image processing apparatus 11 is different from the image processing apparatus 2 in that it has the following two functions. A first function is to associate a plurality of sprites with each other on the sprite buffer 18 or the frame buffer 23 (hereinafter referred to as “link”) and handle them as one object. Second, image processing that gives motion and depth to the image displayed on the display device 4 (the above-described image processing for realizing blur, glow, and drop / shadow) is executed using the first function. It is a function.
Hereinafter, the image processing apparatus 11 that remarkably shows the features of the present embodiment will be mainly described.

  As shown in FIG. 1, the image processing apparatus 11 includes a sprite attribute table 13, a decode control circuit 14, a ROM interface 15, a decoder 16, a sprite buffer interface 17, a sprite buffer 18, a rendering control circuit 20, a rendering engine 21, and a frame buffer. It has an interface 22, a frame buffer 23, a display control circuit 25, and a host interface 26.

  A host CPU 5 is connected to the host interface 26. The host CPU 5 registers the sprite attribute data in the sprite attribute table 13 via the host interface 26. The sprite attribute data includes various information related to sprites to be displayed on the display device 4 (storage destination address indicating the storage location of the sprite compression-encoded data in the pattern ROM 3, parameters for determining enlargement, reduction, rotation, deformation, etc. of the sprite) Arrangement position data indicating the position).

  The decode control circuit 14 reads the sprite attribute data written in the sprite attribute table 13 by the host CPU 5. Then, the decode control circuit 14 controls the process of reading out the compressed encoded data indicated by the storage destination address included in the sprite attribute data read from the sprite attribute table 13 from the pattern ROM 3, and the decoding process of the compressed encoded data Control.

  More specifically, the decode control circuit 14 gives the ROM interface 15 the storage address included in the sprite attribute data read from the sprite attribute table 13. The ROM interface 15 outputs the storage address given from the decode control circuit 14 to the pattern ROM 3 as a read address, and outputs the compressed encoded data read from the pattern ROM 3 to the decoder 16. The decoder 16 decodes (decompresses) the compressed encoded data read from the pattern ROM 3 via the ROM interface 15 to restore the image data before compression encoding (hereinafter referred to as sprite pattern data). The decoder 16 outputs the restored sprite pattern data to the sprite buffer interface 17. The sprite buffer interface 17 is writing means for writing the sprite pattern data output from the decoder 16 to the sprite buffer 18. Further, the sprite buffer interface 17 receives the instruction from the rendering engine 21, reads the sprite pattern data in the sprite buffer 18, and outputs it to the rendering engine 21.

  The sprite buffer 18 is storage means that functions as a FIFO (first-in / first-out memory). The sprite buffer 18 has a storage capacity capable of storing a plurality of sprite pattern data. In this embodiment, ARGB 32-bit image data is used as image data before compression encoding, and the sprite buffer 18 stores this image data. ARGB 32-bit image data means that the pixel value of each pixel constituting a sprite is a transmission coefficient α (8 bits), an R component intensity (8 bits), a G component intensity (8 bits), and a B component intensity. This is image data expressed in (8 bits). Hereinafter, data representing the intensities of the R component, the G component, and the B component may be referred to as “color data”.

  The rendering control circuit 20 reads the sprite attribute data in the sprite attribute table 13 and outputs the read sprite attribute data to the rendering engine 21. The rendering engine 21 performs the rendering process (enlargement, reduction, rotation, deformation, etc.) of the sprite pattern data given from the sprite buffer interface 17 in accordance with an instruction given from the rendering control circuit 20. The rendering engine 21 outputs the sprite pattern data for which the rendering process has been completed to the frame buffer interface 22. The frame buffer interface 22 is writing means for writing the sprite pattern data output from the rendering engine 21 into the frame buffer 23. Further, the frame buffer interface 22 receives the instruction from the display control circuit 25, reads the sprite pattern data from the frame buffer 23, and outputs it to the display control circuit 25.

  The frame buffer 23 is a storage unit that stores image data for one frame. The frame buffer 23 has a double buffer configuration in which writing and reading can be switched alternately. The display control circuit 25 generates various timing signals for image display and outputs them to the display device 4. The display control circuit 25 outputs the sprite pattern data read from the frame buffer 23 via the frame buffer interface 22 to the display device 4 in synchronization with the timing signal.

  In the present embodiment, the PRE bit and NOREL bit are included in the sprite attribute data as information for linking a plurality of sprites on the sprite buffer 18 or the frame buffer 23 so as to be handled as one object. Hereinafter, linking a plurality of sprites on the sprite buffer 18 is referred to as “pre-link”, and linking a plurality of sprites on the frame buffer 23 is referred to as “post-link”. If the value of the NOREL bit included in the sprite attribute data is “0”, it means that there is another sprite linked to the sprite corresponding to the sprite attribute data, and if the NOREL bit is “1”, the link is established. Means there are no other sprites to be played. If the value of the PRE bit included in the sprite attribute data of another sprite to be linked is “0”, it means that the other sprite is post-linked, and if the PRE bit is “1” Sprite means to be prelinked. The sprite attribute data includes an END bit. Normally, the END bit is “0”. When there is a sprite to be linked, if there is sprite attribute data whose END bit is “1”, it means that there is no subsequent linked sprite.

  FIG. 2A is a diagram for explaining prelink. Pre-linking is a method of overwriting sprite pattern data of other sprites linked to sprite pattern data stored in the sprite buffer 18 on the sprite buffer 18 so that a plurality of sprites having a link relationship can be combined into one. It means that it can be handled as an object. In the following, the overwritten sprite (that is, the backmost sprite) is called an absolute sprite (or parent sprite), and the overwritten sprite is called a relative sprite (or child sprite). When a relative sprite is arranged in a region exceeding the absolute sprite image size, the portion belonging to the region is not developed but masked (dotted line portion of the sprite 4 in FIG. 2A). FIG. 2A illustrates a case where four sprites of sprites 1 to 4 are pre-linked to the sprite 0. When the prelink shown in FIG. 2A is completed, the sprite pattern data of sprite 0 (data in which the sprite data of sprite 1 to sprite 4 is partially overwritten by the prelink) is read from the sprite buffer 18 thereafter. Thus, sprite 0 to sprite 4 can be handled as one object.

  FIG. 2B is a diagram for explaining the post link. Post-linking refers to overwriting sprite pattern data of relative sprites linked to the absolute sprite on the frame buffer 23 with sprite pattern data of absolute sprites stored in the frame buffer 23. As the absolute sprite in the prelink or the postlink, a normal sprite or a rectangular image painted with a transparent color can be used.

  FIG. 3 shows an example of storing sprite attribute data for the absolute sprite and relative sprite shown in FIG. The sprite attribute table 13 continuously stores sprite attribute data of absolute sprites constituting each frame, and the sprite attribute data of relative sprites is the same in a storage area separated from the storage area of the sprite attribute data of absolute sprites. Relative sprites linked to absolute sprites are stored so that they are continuous. The sprite attribute data of the absolute sprite to which one or a plurality of relative sprites are linked includes the storage destination address of the attribute table of the head relative sprite. Since the data size of the sprite attribute data of one sprite is constant, the storage address of the sprite attribute data of the next relative sprite can be calculated from the storage address of the sprite attribute data of the first relative sprite. As described above, the relative sprite whose END bit value of the sprite attribute data is “1” is the last relative sprite linked to the absolute sprite. Therefore, the storage address of the sprite attribute data is calculated from the storage address of the attribute table of the first relative sprite included in the sprite attribute data of the absolute sprite to the relative sprite whose END bit is “1”, and the storage address thereof By reading the sprite attribute data indicated by (2), it is possible to access the sprite attribute data of all relative sprites linked to the absolute sprite. When the link between the absolute sprite and the relative sprite is a prelink, the arrangement position data included in the sprite attribute data of the relative sprite is the sprite most recently developed in the sprite buffer 18 (the relative sprite is the first relative sprite. If present, it represents a relative position to the absolute sprite (the position of the upper left corner of the relative sprite when the upper left corner of the most recently developed sprite is used as the origin). That is, the prelinked relative sprite is not drawn alone in the sprite buffer 18 and the frame buffer 23.

  In this embodiment, in order to realize each of blur, glow, and drop shadow using pre-link and post-link, the sprite pattern data is a pattern in the following manner for the sprite to which blur, glow, and drop shadow is applied. Registered in the ROM 3 in advance.

  For sprites to be blurred, vertically long image data representing a transparent color having an image size twice that of the sprite is compressed and encoded in the pattern ROM 3 as sprite pattern data of absolute sprites. As a result, a vertically long area that is twice the image size is secured in the sprite buffer 18. Then, image data consisting only of the α data of the contour portion of the sprite to be blurred is compressed and registered in the pattern ROM 3 as sprite pattern data of the first relative sprite, and only the color data of the sprite is registered. Is registered in the pattern ROM 3 as the sprite pattern data of the second relative sprite in the pattern ROM 3. “1” is set in the PRE bit of the sprite attribute data of the first and second relative sprites. That is, the first and second relative sprites are prelinked with respect to the absolute sprite. The sprite attribute data of the first relative sprite includes arrangement position data indicating that the first relative sprite is arranged in the first half area (hereinafter referred to as odd area) obtained by dividing the absolute sprite into two equal parts in the horizontal direction. It is. On the other hand, in the sprite attribute data of the second relative sprite, arrangement position data indicating that the second relative sprite is arranged in the latter half area (hereinafter, even area) obtained by dividing the absolute sprite into two equal parts in the horizontal direction. It is included.

  For sprites that are the subject of glow or drop / shadow, vertically long image data representing a transparent color having an image size twice that of the sprite is compressed and registered in the pattern ROM 3 as sprite pattern data of the absolute sprite. . As a result, a vertically long area that is twice the image size is secured in the sprite buffer 18. Then, image data consisting only of α data of the outline portion of the sprite that is the target of glow or drop / shadow is compressed and encoded and registered in the pattern ROM 3 as sprite pattern data of the first relative sprite. Furthermore, image data consisting only of the color data of the sprite that is the target of glow or drop / shadow is compressed and registered in the pattern ROM 3 as the sprite pattern data of the second relative sprite.

“1” is set in the PRE bit of the sprite attribute data of the first relative sprite. That is, the first relative sprite is pre-linked to the absolute sprite. The sprite attribute data of the first relative sprite includes arrangement position data indicating that the first relative sprite is arranged in the even area. On the other hand, “0” is set in the PRE bit of the sprite attribute data of the second relative sprite. That is, the second relative sprite is post-linked to a new sprite obtained by pre-linking the first relative sprite to the absolute sprite. The sprite attribute data of the second relative sprite with respect to the sprite to be glowed is centered with respect to the parameter indicating that the second relative sprite is to be reduced (or enlarged) and the new sprite. Arrangement position data indicating that the elements are arranged so as to coincide with each other. On the other hand, in the sprite attribute data of the second relative sprite for the sprite that is the target of drop / shadow, a parameter indicating that the second relative sprite is to be reduced and the second relative sprite are added to the new relative sprite. Arrangement position data indicating that the center is shifted with respect to the sprite is included.
The above is the configuration of the image processing apparatus 11.

(B: Operation)
Next, the operation of the image processing apparatus 11 will be described with reference to the flowcharts shown in FIGS.
(B-1: Flow of decoding / development sequence)
FIG. 4 is a flowchart showing the flow of the decoding / decompression sequence executed by the image processing apparatus 11. The decoding / decompression sequence refers to a process of reading and decoding the compression-encoded data stored in the pattern ROM 3 and writing the sprite pattern data as a decoding result in the sprite buffer 18. In the decoding / development sequence, the decoding control circuit 14 first accesses the sprite attribute table 13 and acquires sprite attribute data (step Sa1). Next, the decode control circuit 14 outputs the storage address of the compressed encoded data included in the sprite attribute data acquired in step Sa1 to the ROM interface 15 and instructs to read one block. Here, one block is, for example, 16 × 16 dot data, and the sprite pattern data is usually composed of n (n is an integer larger than 1) block data. Receiving the above instruction, the ROM interface 15 reads one block of compressed encoded data from the pattern ROM 3 and outputs it to the decoder 16. The decoder 16 decodes the compressed encoded data (step Sa2).

  Next, the decode control circuit 14 determines whether or not it is necessary to develop the block decoded in step Sa2 into the sprite buffer 18 (step Sa3). For example, when the block decoded in step Sa2 is a relative sprite block prelinked to an absolute sprite, and all of the sprite pattern data of the block belongs to an area exceeding the image size of the absolute sprite. On the other hand, the decode control circuit 14 determines that the expansion of the block is not necessary, and determines that the expansion is necessary in other cases. If the determination result of step Sa3 is “No” (that is, if it is determined that expansion is not required), the process of step Sa4 is executed. Conversely, if the determination result of step Sa3 is “Yes”, the process of step Sa5 is performed. Processing is executed.

  In step Sa4, which is executed when the determination result in step Sa3 is “No”, has the decoding control circuit 14 completed decoding of all the blocks constituting the sprite corresponding to the sprite attribute data acquired in step Sa1? Check whether or not. When the check result in step Sa4 is “No” (that is, when decoding of all the blocks is not completed), the decode control circuit 14 instructs the ROM interface 15 to read the next block. The compression encoded data read by the ROM interface 15 is decoded by the decoder 16 (step Sa2). Thereafter, the above-described operation is repeated, and when the decoding of all the blocks constituting the sprite is completed without expanding the most recently decoded sprite pattern data (step Sa4: Yes), the decode control circuit 14 again performs step Sa1. The sprite attribute data of the next sprite (or the first relative sprite if it is an absolute sprite to which one or more relative sprites are linked) is read from the sprite attribute table 13. Thereafter, the decoding process for the next sprite is performed in the same manner as described above.

  In step Sa5, which is executed when the determination result in step Sa3 is “Yes”, the block decoded in step Sa2 is output from the decoder 16 to the sprite buffer interface 17, and an access sequence to the sprite buffer 18 is started. The As described above, the sprite buffer 18 is a FIFO (first-in / first-out memory), and the access sequence to the sprite buffer 18 is a process of sequentially developing sprite pattern data in the sprite buffer 18 in units of blocks.

  In the access sequence to the sprite buffer 18, first, the sprite buffer interface 17 determines whether the development target block is a relative sprite block prelinked to the absolute sprite (step Sa6). If the determination result in step Sa6 is “No”, the sprite buffer interface 17 checks whether or not the sprite buffer 18 is in the FULL state (step Sa7), and if not, executes the expansion (step Sa8). If the sprite buffer 18 is in the FULL state (step Sa7: Yes), the sprite buffer interface 17 repeatedly executes the process of step Sa7 until the determination result of step Sa7 becomes “No”, and the sprite buffer 18 is free. Wait for you to do.

  If the determination result of step Sa6 is “Yes”, the sprite buffer interface 17 executes the process of step Sa8 without performing the check of step Sa7. The reason why the check of step Sa7 is not performed when the determination result of step Sa6 is “Yes” is that an area for storing the sprite pattern data of the relative sprite is secured when the absolute sprite is expanded.

When the development of one block is finished, the decode control circuit 14 checks whether or not the development of all the blocks is finished (step Sa9). If the development of all the blocks has not been completed (step Sa9: No), the process returns to step Sa5, waits for the decoding of the next block to be completed, and the processes after step Sa6 are executed for that block. Further, when the development of all the blocks has been completed (step Sa9: Yes), the processing after step Sa1 is executed again. However, the sprite pattern data that has been expanded into the sprite buffer 18 by the above processing is the sprite pattern of the last relative sprite prelinked to the absolute sprite, and is formed by linking the absolute sprite and the relative sprite. When it is instructed to apply blur, glow or drop shadow to the object, the sprite buffer interface 17 executes image processing for realizing these. Details of this image processing will be made clear later.
The above is the flow of the decoding / decompression sequence executed in the image processing apparatus 11.

(B-2: Flow of drawing sequence)
Next, a drawing sequence for writing sprite pattern data to the frame buffer 23 will be described. This drawing sequence is executed asynchronously with the decoding / decompression sequence described above. FIG. 5 is a flowchart showing a flow of a drawing sequence executed by the image processing apparatus 11. In this drawing sequence, first, sprite attribute data such as parameters for determining enlargement, reduction, rotation, deformation, etc. of the sprite pattern to be written to the frame buffer 23 and data indicating the display position are read from the sprite attribute table 13 by the rendering control circuit 20. (Step Sb1). At this time, the sprite attribute data for the relative sprite prelinked to the absolute sprite is excluded from the read target. This is because it is not necessary to draw the relative sprite prelinked to the absolute sprite in the frame buffer 23 as described above. However, the attribute data of the relative sprite to be post-linked is read, and the drawing is performed after the drawing of the absolute sprite.

  The rendering control circuit 20 gives the sprite attribute data read from the sprite attribute table 13 to the rendering engine 21. The rendering engine 21 calculates initial parameters based on the sprite attribute data input from the rendering control circuit 20 (step Sb2). Next, the rendering control circuit 20 determines whether the image of the sprite to be drawn has already been developed on the sprite buffer 18 with reference to the status (EMPTY) of the sprite buffer (step Sb3). The rendering control circuit 20 performs a drawing process if the determination result in step Sb3 is “No” (that is, not EMPTY) (step Sb4), and if the determination result in step Sb3 is Yes, the corresponding sprite is developed. Step Sb3 is repeated until is completed.

  In the rendering process of step Sb4, the rendering control circuit 20 first instructs the sprite buffer interface 17 to read sprite pattern data. In response to the instruction, the sprite buffer interface 17 reads out data corresponding to the coordinates (address) calculated by the rendering engine 21 based on the sprite attribute data from the sprite buffer 18 and outputs the data to the rendering engine 21. The rendering engine 21 outputs the data and display position data indicating the display position to the frame buffer interface 22. The frame buffer interface 22 draws the sprite pattern data received from the rendering engine 21 at the address corresponding to the display position data received from the rendering engine 21 together with the sprite pattern data (steps Sb4 and Sb5).

Since drawing processing to the frame buffer 23 includes processing such as rotation and deformation, confirmation of EMPTY of the sprite buffer 18 is determined in units of sprites. If the entire image of one sprite is not developed in the sprite buffer 18, the drawing process cannot be started. When all drawing of the sprite is completed (step Sb6: YES), the process returns to step Sb1, and the rendering control circuit 20 reads the next sprite attribute data from the sprite attribute table 13. Then, the next sprite rendering process is performed based on the read sprite attribute data.
The above is the flow of the drawing sequence.

  As described above, the reading of the sprite pattern data from the sprite buffer 18 is executed asynchronously with the writing of the sprite pattern data to the sprite buffer 18. That is, the sprite buffer 18 is a FIFO having a storage capacity capable of developing a plurality of sprite patterns as described above. The pattern data of the next sprite is not waited until the drawing of the frame buffer 23 is finished unless the FIFO becomes FULL. It can be decoded and expanded in the sprite buffer 18. The rendering control circuit 20 can continue drawing in the frame buffer 23 if the FIFO is not EMPTY. As a result, each process can be performed without considering the time difference between the decoding process of the pattern data in the pattern ROM 3 and the drawing process of the frame buffer 23. As a result, the drawing processing capability can be improved.

  Next, image processing for realizing each of blur, glow, and drop / shadow will be described. As described above, in the present embodiment, sprite pattern data is registered in the pattern ROM 3 by dividing the sprite that is subject to any one of blur, glow, and drop / shadow into an absolute sprite and a plurality of relative sprites. . For this reason, the decoding / decompression sequence processing is executed in the order of the absolute sprite, the first relative sprite, and the second relative sprite for the sprite to be blurred. Similarly, with respect to sprites to be glowed or dropped / shadowed, decoding / development sequence processing is executed in the order of absolute sprite, first relative sprite, and second relative sprite.

  The first and second relative sprites are relative sprites that are pre-linked to absolute sprites when the processing to be performed on the sprites to be processed is blur, and the first relative sprites are glow and drop / shadows. . In the case of the blur, the sprite pattern data of the first relative sprite is developed in the odd area of the absolute sprite on the sprite buffer 18 in accordance with the arrangement position information included in the sprite attribute data for the first relative sprite. On the other hand, the sprite pattern data of the second relative sprite is expanded in the even area of the absolute sprite on the sprite buffer 18 in accordance with the arrangement position information included in the sprite attribute data for the second relative sprite. In the case of glow, drop, and shadow, the sprite pattern data of the first relative sprite is the absolute sprite even on the sprite buffer 18 in accordance with the arrangement position information included in the sprite attribute data for the first relative sprite. Expands to an area.

  When the processing to be performed on the sprite to be processed is blur, the last relative sprite to be prelinked to the absolute sprite is the second relative sprite, and therefore, after the expansion of the sprite pattern data of the second relative sprite is completed The sprite buffer interface 17 executes the processing shown in FIG. On the other hand, when the processing to be performed on the sprite to be processed is glow or drop shadow, the last relative sprite to be prelinked with respect to the absolute sprite is the first relative sprite. After completing the development of the sprite pattern data of the sprite, the sprite buffer interface 17 executes the processing shown in FIG. That is, the sprite buffer interface 17 performs Gaussian filter processing while reading the sprite pattern data of the relative sprite expanded in the even area and writes it in the odd area (FIG. 6: step st1, which is surrounded by a circle in FIG. 6). F represents filtering. The Gaussian filter process is a process of calculating a pixel value of a pixel to be processed by adding a value obtained by multiplying a pixel value of each pixel of a pixel to be processed and its surrounding pixels and a coefficient corresponding to each pixel. . For example, when Gaussian filter processing is performed in units of 5 × 5 matrix (when pixel u33 in FIG. 7 is a processing target), the pixel value of each pixel in the 5 × 5 region is set to the pixel value of each pixel shown in FIG. The result of adding the values obtained by multiplying the coefficients for each pixel is the Gaussian filter processing result of the pixel U33.

  If the relative sprite developed in the even area is image data of only color data, the sprite buffer interface 17 performs Gaussian filter processing only on the color data and overwrites the color data of the image data stored in the odd area. If the relative sprite developed in the even area is image data of only α data, the sprite buffer interface 17 performs Gaussian filter processing only on the α data and overwrites the α data of the image data stored in the odd area. After step st1, the sprite buffer interface 17 performs Gaussian filter processing while reading the data written in the odd area and writes it in the even area (step st2). Thereafter, the sprite buffer interface 17 switches even / odd and repeats the gauss filter cascade processing a predetermined number of times (FIG. 6: step st3 to step st7). The number of repetitions of the Gaussian filter cascade processing may be included in the sprite attribute data of the absolute sprite and given to the image processing apparatus 11, or may be a fixed value determined in advance. FIG. 6 illustrates the case where the cascade processing of the Gaussian filter is performed seven times.

  When the processing to be performed on the sprite to be processed is a blur, the sprite pattern data of the first relative sprite is α data of the contour portion of the sprite to be processed, and the sprite pattern data of the second relative sprite is the processing target This is the color data of the sprite. When the number of times of the cascade processing is an odd number, the final result (that is, image data composed of color data subjected to the Gaussian filter cascade processing and α data not subjected to the cascade processing) is odd. Expands to an area. If the above-described drawing sequence is executed with the stored contents of the odd area as a processing target, the blurred sprite image data is drawn in the frame buffer 23 by the cascade of Gaussian filter processing, and blurring is realized.

  When the processing to be performed on the sprite to be processed is glow or drop shadow, the sprite pattern data of the first relative sprite is image data including only α data of the contour portion of the sprite to be processed. Therefore, when the Gaussian filter cascade processing is performed an odd number of times, the α data of each pixel stored in the odd area is updated to a result of repeated Gaussian filter processing. Glow and drop shadow are realized by causing the frame buffer interface 22 to execute the processing shown in FIG. 8 in the drawing sequence in addition to the processing shown in FIG. In other words, the frame buffer interface 22 uses the Gaussian filtered α data stored in the odd area to perform a fixed color (for example, white for glow and gray for drop / shadow) and transparency processing. Then, it draws in the frame buffer 23 as a new sprite. Then, the frame buffer interface 22 post-links the second relative sprite (in the example shown in FIG. 8, an image of the character “A”) to the new sprite. In FIG. 8, stippled hatching represents a transparent color, hatched hatching represents the fixed color, and horizontal hatching represents a background color.

  If the process to be performed on the sprite to be processed is glow, the second relative sprite is reduced (or enlarged) and then overlapped on the frame buffer 23 with the center aligned with the new sprite. Written. As a result, the outline of the second relative sprite is given the blurred border of the fixed color, and a glow is realized. On the other hand, if the process to be performed on the sprite to be processed is a drop shadow, the second relative sprite is reduced and then the center is shifted with respect to the new sprite on the frame buffer 23. Are overwritten. As a result, a portion of the new sprite that does not overlap with the second relative sprite appears as a shadow, and a drop shadow is realized.

  As described above, according to the present embodiment, when the sprite pattern read from the pattern ROM 3 and decoded in real time is drawn on the frame buffer 23 while being developed in the sprite buffer 18 in units of blocks, the depth of the sprite pattern is drawn. It is possible to realize blur, glow, drop shadows that produce a feeling.

(C: deformation)
Although one embodiment of the present invention has been described above, this embodiment may of course be modified as follows.
(1) In the above embodiment, the absolute sprite has an image size twice that of the sprite to be processed, but may be the same image size as the sprite to be processed. In this case, after securing a storage area for the image size twice as large as the processing target sprite in the sprite buffer 18, an absolute sprite (transparent color image) is arranged in the odd area, and a relative sprite is arranged in the even area. The relative sprite arrangement information is defined as follows. In this case, the relative sprite is expanded outside the area of the absolute sprite, but it may be handled as irregular. According to such an aspect, the image of the odd area that is the final drawing target matches the absolute sprite, and the drawing sequence is simplified.

(2) In the above-described embodiment, a description will be given of a case where image processing that produces sprite movement and depth feeling such as blur, glow, drop / shadow, etc. is realized while alternately using odd and even areas secured in the sprite buffer 18. did. However, an Arabic font with complex composition rules can be created by scaling, rotating, and transforming the entire character string while using pre-link and post-link to link multiple sprites on the sprite buffer 18 or the frame buffer 23. It is also possible to draw a character string of a font having a non-uniform width, such as a layout, mixed languages, and the like.

(3) In the above-described embodiment, a case has been described in which image processing that produces sprite movement and depth feeling such as blur, glow, and drop / shadow is realized. However, the application target of the present invention is not limited to these image processing, and can be applied to any image processing that requires repeated filtering processing on sprite pattern data. The filtering processing is Gaussian filtering processing. It is not limited to.

(4) In the above embodiment, the Gaussian filter process is performed only on the color data when realizing the blur. However, the Gaussian filter process may also be performed on the transmission processing coefficient. In short, a sprite attribute table that stores attribute data of each sprite, a first writing unit that writes image data of a sprite to be drawn into the first storage unit according to the attribute data of the sprite attribute table, and one frame Second storage means for storing image data representing an image, and second writing means for writing the image data stored in the first storage means to the second storage means, and at least one sprite Is stored in the attribute data of the sprite, and the position information of at least one other sprite is stored in the attribute data of the sprite as a relative position from other sprites other than the sprite. The first writing means stores twice the image size of the sprite to be filtered. An area is secured in the first storage means, the sprite is written into the storage area according to the attribute data of the at least one sprite, and the storage area is equally divided into the first half and the second half, and the other at least one other Writing the at least one other sprite in accordance with the attribute data of the sprite, performing a filtering process on the most recently written area of the first half area and the second half area, and writing to the other area. The second writing means reads the image data of the most recently written area from the first storage means and repeats the predetermined number of times, and the second storage means Any image processing apparatus that is characterized by writing to the means may be used. In this case, different types of filter processing may be applied to the color data and the transmission processing coefficient. By doing so, it is considered that an effect different from any of blur, glow, and drop shadow can be imparted.

DESCRIPTION OF SYMBOLS 3 ... Pattern ROM, 4 ... Display apparatus, 11 ... Image processing apparatus, 13 ... Sprite attribute table, 14 ... Decoding control circuit, 15 ... ROM interface, 16 ... Decoder, 17 ... Sprite buffer interface, 18 ... Sprite buffer, 20 ... Rendering control circuit, 21 ... rendering engine, 22 ... frame buffer interface, 23 ... frame buffer, 25 ... display control circuit, 26 ... host interface.

Claims (6)

A sprite attribute table for storing attribute data of each sprite;
First writing means for writing image data of a sprite to be drawn into a first storage means in accordance with attribute data in the sprite attribute table;
Second storage means for storing image data representing an image of one frame;
Second writing means for writing image data stored in the first storage means to the second storage means,
Position information of at least one sprite is stored in the attribute data of the sprite, and position information of at least one other sprite is stored in the attribute data of the sprite as a relative position from other sprites other than the sprite. And
The first writing means secures a storage area twice as large as the image size of the sprite to be filtered in the first storage means, and stores the sprite according to the attribute data of the at least one sprite. Write to the area and write the at least one other sprite in accordance with the attribute data of the at least one other sprite to the storage area equally divided into the first half and the second half. Apply the filter process to the area that was most recently written and repeat the process of writing to the other area a predetermined number of times,
The second writing means reads the image data of the most recently written area from the first and second half areas from the first storage means and writes the image data to the second storage means. Image processing device. The image data of the sprite is composed of color data representing the color of each pixel constituting the sprite and a transmission processing coefficient of each pixel.
The image processing apparatus according to claim 1, wherein the first writing unit performs the filtering process on color data. The image data of the sprite is composed of color data representing the color of each pixel constituting the sprite and a transmission processing coefficient of each pixel.
The image processing apparatus according to claim 1, wherein the first writing unit performs the filtering process on a transmission processing coefficient. The second writing means includes
The image of a predetermined color having the same image size as the one of the equally divided areas is subjected to a transmission process using a transmission coefficient stored in the most recently written area of the first half area and the second half area. The image obtained in this way is written in the second storage means, and the image represented by the color data of the image data of the at least one other sprite is overwritten by reducing or enlarging it. The image processing apparatus described.   The image processing apparatus according to claim 1, wherein the first storage unit functions as a first-in / first-out memory. Compression encoded data obtained by performing compression encoding on the sprite image data is read from the pattern ROM in accordance with the attribute data in the sprite attribute table, and the image data before compression encoding is restored and supplied to the first writing means. The image processing apparatus according to claim 1, further comprising a decoder.

Images