JPS6232676B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for encoding relatively simple color figures or figures having halftones with good quality using run length information.

Conventionally, when encoding a figure, the figure is first converted into an electrical signal, regardless of the target figure, and then analog-to-digital conversion (hereinafter referred to as A/D conversion) is performed.
Finally, a method was used to convert digital information into a fixed code format. As a result, relatively simple shapes,
For example, even if the number of colors used and the shape is simple, such as illustrations and schematic diagrams, the amount of encoded information will not be very small due to uneven coloring of the figure, noise when inputting the figure, etc., and the amount of encoded information will not be very small. At the time, the disadvantage was that the decoded figures had noticeable color unevenness and noise.

When the target figure is a relatively simple color figure or a black-and-white figure with halftones, the present invention digitally converts it according to its color, shading, etc. (hereinafter abbreviated as figure form), and encodes it. Detects the edges of a figure with a specific shape from among the target figures, knows the distance from the edge of the screen to the edge, and from the edge of the edge to the edge of the edge,
For each of these sections, the sum of the numbers of consecutive pixels having the same figure form, that is, the sum of the run lengths that is the largest, is determined as the figure form of all pixels existing in that interval, and is encoded and output. By doing so, the graphic form does not change from the edge of the screen to the edge of the border, or from the edge of the border to the edge of the border, in other words, it is not affected by screen smudges or external noise. . A detailed description will be given below with reference to the drawings.

In FIG. 1, a relatively simple figure 2 is formed on a screen 1, the figure 2 and a background 3 are painted in different colors, and the screen 1 is composed of these two colors.
A border 4 of a color different from the above two colors is applied to this figure. Generally, the outline of the figure 2 is drawn and then colored, so in many cases the outline constitutes the border 4, and therefore there is no particular need to add the border 4. If the target figure 2 does not have a border and you cannot modify this original screen, for example, place a piece of transparent paper on top of screen 1, and draw a color that is not used on the screen over the transparent paper. Just apply border 4.

In this way, a screen fringed with colors that are not present on the screen is raster-scanned, converted into an electrical signal, and then encoded.

Although various encoding methods are known, the run-length encoding method is suitable for encoding a screen like screen 1 shown in FIG. The basics of this encoding method is to encode a set of consecutive pixels obtained by raster scanning the screen using their color and number of pixels. For example, one set is encoded in the format shown in Figure 3. . When encoding an actual screen, one set may be encoded with a plurality of codes due to the relationship between the number of bits assigned to the code and the number of consecutive pixels of the same color.

In the following explanation, it will be explained that the screen 1 is encoded using the encoding format shown in FIG. 3, but the present invention is not limited to the encoding format and encoding method explained here as an example. Needless to say. In addition, in FIG. 3, 31 indicates the color of consecutive pixels, and 32 indicates the continuous length.

FIG. 2 shows an embodiment of the present invention, in which the image information on the screen 1 described in FIG. 1 is converted into an electrical signal by the imaging section 5. In FIG. A synchronization signal for controlling the imaging section 5 is received from a synchronization signal generator 6. The output signal of the imaging unit 5 is, for example, the red R on the screen,
Red light, green light according to the shape of green G, blue B,
The signal is supplied to A/D converters 7, 8, and 9 as a green signal. The A/D conversion units 7 to 9 sample the output signal of the imaging unit 5 using the sampling signal generated by the synchronization signal generator 6, and digitize it. In this embodiment, each signal is digitalized with 1 bit. It has become A/
The digital information A/D-converted by the D converters 7 to 9 is counted by counter groups 11 to 13, which constitute pixel number detection means, respectively. Each of the counter groups 11 to 13 includes a counter that counts the number of states that the A/D converters 7 to 9 can take, in this example two states, "0" and "1", and the occurrence of the state. The number of pixels is counted. These counter groups 11 to 13 are set by a signal from the control section 14, and the outputs of the A/D conversion sections 7 to 9 at that time are set to "0" or "1" for each sampling signal from the synchronization signal generator 6. ”, one of the corresponding counters in each counter group 11 to 13 is incremented by one step.

A length counter 15 is provided, which is reset by the control section 14 and counts the sampled signal from the synchronization signal generator 6, regardless of the states of the A/D conversion sections 7-9. Furthermore, a border detection section 17 is provided, and since the color of the pixel of the figure on the screen 1 being scanned can be determined from the state of each output of the A/D conversion sections 7 to 9, the arrangement of bits indicating a preset border is detected. (For example, when black is represented by R = 0, G = 0, B = 0, it is represented by "000", and true red is represented by R = 1, G = 0, B = 0.
(When expressed as "100"), it is detected whether the pixel being scanned indicates border 4 or not, and the control unit 14 is notified of this.
The control unit 14 instructs the encoding unit 18 to perform encoding when the border detection unit 17 starts detecting a border, when the detection ends, and when line scanning for the screen 1 ends, and then the counter group 1
1 to 13 and the length counter 15 are reset respectively. At this time, the imaging unit 5
When not encoding the output of the synchronizing signal generator 6, the control unit 14 detects the retrace period using the synchronizing signal from the synchronizing signal generator 6, does not output a signal that instructs the encoding unit 18 to encode, and controls the counter group. 11~
13. The counter 15 may be maintained in a reset state.

Every time the encoding unit 18 receives an encoding instruction signal from the control unit 14, it first detects the counter with the larger count value for each of the counter groups 11 to 13.

In this embodiment, each counter group has a counter for which the output value of the A/D converters 7 to 9 is "0" and a counter for "1". A/D converter 7 that selects a counter and that counter should count
.about.9, that is, "1" or "0", are regarded as the outputs of the A/D converters 7 to 9 for all pixels in that section.

As a result, one state has been selected for each color component of one section, and information indicating which color that section is can be obtained by combining these states. On the other hand, information indicating the length of the section, that is, the number of pixels in the section, can be obtained from the contents of the length counter 15. The encoding unit 18 creates a code for the section from these two types of information as shown in FIG. 3, and outputs it. Therefore, regarding the scanning line 1 _i shown in FIG. 1, from the edge P ₀ of the screen 1 to the beginning P ₁ of the border, there is information indicating the color of the background 3, and the length between these P ₀ and P ₁ . A code consisting of the information indicating the border is output, and from the start of the border P ₁ to the end P ₂ of the border is the color of border 4, and the code of the length between them is the end of the border.
From _P2 to the beginning of the border _P3 , the color of figure 2 and the code of the length between them are output.

If it is inconvenient to decode and represent the border 4 in the state shown in FIG. Either the encoding is performed by giving an encoding instruction signal to the encoding unit 18 and the reset signal for the counter groups 11 to 13 and the counter 15, respectively, or the encoding is performed including the edges, and when decoding, the edges are replaced with the edges. Just replace it with the previous or next color. In the above description, each color signal is distinguished into two types, "0" and "1", in the A/D converters 7 to 9, so three types of color components can take two states, and 2 ³ = Eight types of colors can be selected, but by adding more bits to distinguish more levels of each color signal and providing a group of counters corresponding to the number of possible states, it is possible to select more colors. can.

In addition, in the above description, colors are defined by combining the states of the counters of each set using three sets of counters 11 to 13, each set of two counters, but for example, if eight counters are provided, It can also be configured to count the color information itself. In the case of such a configuration, the one with the largest count value among the eight counters may be selected, and the color or shade determined by that counter may be defined as all pixel information in that section.

Furthermore, if a plurality of gray levels are displayed so that halftones occur not only on a color screen but also on a black and white screen, the output of the imaging section 5 is converted into a digital signal by one A/D conversion section, and at least three depending on the size of the digital signal. One of the two counters may be incremented, and a code consisting of density information of a level corresponding to the maximum count of the counter in one coding section and information on the length of the section may be output as the coded output.

As described above, according to the present invention, even if there are spots or noise on the screen in one encoding section, that is, from one edge of the screen to the edge of the border, or from the edge of the border to the edge of the border, Such noise does not affect the output since it is small within one coding interval and appears as a small count value in the counters in each of the counter groups 11-13. In the above-mentioned encoding, when looking at one encoding section, the one having the largest sum of the number of pixels representing graphic forms such as various colors and shadings is set as the encoded output. Therefore, for relatively simple figures, the amount of information in the encoded output is small and efficient encoding can be performed.

In the above, the main scanning direction (scanning line direction.
In the figure, we have described an encoding method that removes the color unevenness of figures and the influence of external noise in the horizontal direction (horizontal direction).
Also in the vertical direction (in the figure), the encoded output of adjacent scanning lines may differ due to noise etc. for pixels having the same color or shading information, so this also applies in the sub-scanning direction. By applying similar processing to the graphics, it is possible to encode the graphics with better image quality.

That is, the output code obtained from the encoding unit 18 is decoded and developed on the image memory, and the main scanning direction of the figure is shifted by 90 degrees (in Fig. 1, the main scanning direction is changed to the vertical direction). performs a read,
The output signals may be given in place of the output signals of the A/D converters 7 to 9 for re-encoding.

Note that when re-encoding in this way, the main scanning direction is shifted by 90 degrees, so it is necessary to encode again with the main scanning direction shifted 90 degrees. If scanning is performed, encoding can be omitted once.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing an example of a figure with borders, Fig. 2 is a block diagram showing an embodiment of the figure encoding method according to the present invention, and Fig. 3 is a diagram showing an example of a run-length code. be. 1...Color screen, 2...Graphics, 3...Background,
4... Border, 5... Imaging section, 6... Synchronization signal generator, 7-9... A/D conversion section, 11-13...
...Counter group, 14...Control unit, 15...Length counter, 17...Edge detection unit, 18...Encoding unit.

Claims (1)

[Claims] 1. Scanning and inputting a figure with a border in a predetermined specific color and/or shading level, and detecting the border existing in the figure based on the specific color and/or shading level. a border detection means for detecting a border; a section identification means for identifying, by the border detection means, a section from an edge of the screen to an edge of the border, and a section from one edge of the border to the end of the border; means for detecting the number of pixels existing in an interval for each interval identified by the interval identifying means; /or check the shading level, count which of the plurality of classification categories according to preset colors and/or shading levels each pixel belongs to, and as a result of counting, the counted value among the plurality of classification categories is determined. an in-section color/shade level determination means for determining that all pixels existing in the section have the color and/or shading level of the classification section that has become the largest; and an encoding means for encoding and outputting an image signal based on the number of pixels obtained and the color and/or gray level obtained by the intra-section color/shade level determining means. Graphical encoding method.