JPH0815311B2 - Color document image processing device - Google Patents

Description

TECHNICAL FIELD OF THE INVENTION The present invention relates to a color document image processing apparatus for inputting a color document image, performing predetermined image processing such as enlargement / reduction conversion, and outputting the image.

[Technical background of the invention] Recently, by improving the efficiency of work by digitizing conventional paper-based general work, especially office work, it is easy to deal with complicated work due to sophistication of work. For this purpose, a document image processing apparatus, such as a facsimile or an electronic file, for electronically processing a document image has been actively developed.

In these devices, generally, a document image is scanned by a scanner and converted into an electric signal, and then compressed and stored, or once stored in an image memory, processed and modified and output to an image printer or a CRT display. Such information processing is performed. In recent years, it has become relatively easy to input a color document image, and it is desired to develop an apparatus that can perform the same processing on a color document image.

[Problems of background art] However, in the processing of color document images, there are various problems that cannot be solved by simply extending the conventional monochrome document image processing, and in particular, a large increase in the amount of information to be handled. It is the biggest problem. That is, even if the image is not an image in which the color tone changes continuously like a photograph, the hues handled by documents are various,
For accurate reproduction, it is necessary to add an information amount of several bits to each hue component, and the information amount becomes enormous. For example, a black and white binary image can be represented by an information amount of 1 bit per pixel, while 8 bits of each of R, G, B (R: red, G: green, B: blue) of a color image are used. When the amount of information is added, it is really 24 times (= 8 ×
The amount of information increases in 3).

If the amount of information to be handled increases in this way, not only the effective capacity of the storage device or the information storage device increases, but also the scale of birdware of each image processing system increases significantly. Therefore, the processing of a color document image becomes extremely expensive, and as an image processing apparatus used in a general office, what has been practically satisfactory has not been developed at present.

[Object of the Invention] The present invention has been made in view of such circumstances.
Sophisticated handling of color document images by easily realizing image processing for color document images with only a small increase in the amount of birdware, and minimizing the effective reduction in storage and storage of color image information. It is an object of the present invention to provide a color document image processing device that realizes the low cost.

SUMMARY OF THE INVENTION The present invention has an input means for inputting n-bit color image data and a plurality of memories in which conversion tables are stored in advance, and the n-bit color input by the input means. Input image data to each of the plurality of memories,
1-bit encoded image data is output from each memory based on the conversion tables stored in the plurality of memories to convert the encoded image data into m-bit (where m <n) encoded image data. Image processing including a conversion unit 1 and a scaling unit that performs a scaling process and a compression / decompression unit that performs a compression / decompression process on the m-bit encoded image data encoded by the first conversion unit. Means and a plurality of memories in which conversion tables are stored in advance, and the m-bit encoded image data image-processed by the image processing means is input to the plurality of memories, respectively, and the plurality of memories are stored in the plurality of memories. Second conversion means for converting to color image data of 1 bit (where l> m) based on the stored conversion table, and the second conversion means. Output means for outputting color image data of bits, and coded image data of m bits converted by the first conversion means are transferred to the image processing means, and image processing is performed by the image processing means. And a data transfer bus having an m-bit width for transferring the m-bit encoded image data to the second conversion means, the image processing for the color document image can be performed with a small amount of hardware. This is easily realized by only increasing the number, and minimizes the reduction in the effective capacity for storing and accumulating color image information, and realizes sophisticated handling of color document images at low cost.

Generally, when each of R, G, and B of color image information is provided with an 8-bit information amount, 24-bit (3 × 8 = 24) information amount is required to represent one pixel. If the input image information is composed of, for example, 16 colors, it is possible to display all of the input image information with a 4-bit (16 = 2 ⁴ ) code.

The color document image processing apparatus of the present invention applies this principle, and converts n-bit color image information input from the outside into m-bit coded image information smaller than n bits by the first conversion means. , By performing image processing on the coded image information with a small number of bits,
The hardware configuration is made smaller.

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram showing the basic arrangement of a color document image processing apparatus according to an embodiment of the present invention.

As shown in FIG.
Input means 1 for inputting color image data consisting of bits
And a first conversion unit 2 for converting the n-bit color image data into m-bit encoded image data.
An image processing means 3 including a memory for performing various kinds of image processing on the encoded image data, and m on which the image processing is performed.
A second converter 4 for converting 1-bit encoded image data into 1-bit color image data, an output means 5 for outputting the 1-bit color image data, and a control for controlling each of these means. And means 6.

FIG. 2 is a block diagram showing the devices constituting this embodiment.

In the apparatus of this embodiment, the input means 1 is composed of a color scanner 11 for reading a document, and the n-bit color image data input from the scanner 11 is an input interface provided with a first conversion means 2. The data is transmitted via 21 to the data transfer bus B having an m-bit width for data transfer between the respective devices.

The data transfer bus B has a temporary storage device 31 for temporarily storing m-bit encoded image data, and an information compression / expansion unit 32, m for compressing / decompressing m-bit encoded image data.
An image processing device such as an enlarging / reducing means 33 for enlarging and reducing bit-encoded image data, a disk memory 7 which is an information storage device for storing m-bit encoded image data or compressed data thereof. Storage means is connected.

Further, on the data transfer bus B, a display device including a display memory 51a for displaying m-bit encoded image data, a display control means 52a and a CRT display 53a, an output interface 51b and an output from the output interface 51b. An output unit such as a color printing device including a color printer 52b for ejecting the image is connected. The display control means 52a and the output interface 51b are provided with second conversion units 4a and 4b for converting m-bit encoded image data into 1-bit color image data, and a CRT display 53a and a color printer 52b.
Respectively display or print 1-bit color image data.

The control means 6 manages the control of this apparatus through a control bus (not shown) and controls by the microprocessor.

In this embodiment, the input color image data is composed of 18 bits (n = 6 × 3 = 18) for each of the color signals R, G and B input from the color scanner 11, 6 bits in total. It is assumed that the number of colors used in the input document is up to 16 colors and 4 bits (16 = 2 ⁴ , m = 4) in the first conversion means 2.
Has been converted to. Further, at the time of output, the second converting means 4 does not add 8 bits to each of the color signals R, G, and B.
It has been converted to 24 bits (l = 8 × 3 = 24).

Next, the flow of color image data in this embodiment will be described.

The color document image data input from the scanner 11 with 6 bits for each of R, G, and B is input to the first conversion unit 2 via the input interface 21, and is converted into 4-bit encoded image data here. This conversion is performed by referring to a conversion table created by a method described later according to the color type and the number of colors used in advance. The converted 4-bit encoded image data is stored in the temporary storage device 31.
Stored in.

Next, the enlarging / reducing means 33 reads out the coded image data from the temporary storage device 31 and converts the coded image data, for example, in the display memory 51a
To enlarge and reduce the image. Since the conversion at this time is performed on encoded 4-bit data, the hardware configuration can be small.

The encoded image data stored in the temporary storage device 31 is information-compressed by the information compression / expansion means 32 and stored in the disk memory 7. This compression is 10 if the compression method is selected.
More than double compression is possible. The coded image data is compressed to 4/18 at the code writing stage. Therefore, a total of 40 to 50 times data compression will be realized without deterioration of image quality, and the system will be extremely practical.

Further, the coded image data stored in the disk memory 7 is converted into 24-bit color image data via the second conversion means 4a and 4b, output to the color printer 53b, and displayed on the display 53a.

As described above, according to this embodiment, 18
All the image data having bits can be handled by 4 bits in the internal processing, and the hardware can be significantly reduced accordingly.

Further, both the first conversion unit 2 and the second conversion units 4a and 4b have a conversion table, and the conversion is executed by referring to this conversion table. Therefore, it is possible to correspond to any color configuration of the scanner 11 and the color printer 53b, and conversely, it is possible to process the birdware in the system regardless of the configuration of the scanner or the color printer. .

Further, in a general color display device, there are many cases in which color data is displayed on a display via a look-up table, but this color document image processing device is very well adapted to this system. That is, the contents of the conventional look-up table can be applied to the apparatus of the present invention by replacing it with a conversion table described later.

In this embodiment, the first conversion unit 2 is configured so that the conversion table can be set by the input interface 21 and the control means 6 while reading an image. Basically, it suffices to identify how many colors are used in the target color document image and assign a code to each of them, and this schematic flow is shown in FIG.

That is, when one pixel of a color document image is input (step 301), it is determined whether the color of the pixel is different from the previous pixel (step 303), and it is determined that the color is different from the previous pixel. Then, the color registered so far is searched (step 305), and it is judged whether or not the color has already been registered (step 307). If not registered, the color is registered as a new color (step 307). Step 309). Step 3
If it is determined in 03 that the input pixel has the same color as the previous pixel, or if it is determined in step 307 that this color has already been registered, it is determined that the processing has ended (step 311). ) Unless otherwise step 301
Return to

In this way, the color used in the target color document image is registered, and a code is assigned to the registered color and the first conversion unit 2 and the second conversion units 4a and 4b are registered.
Create a conversion table for.

 Next, the color identification method will be described in more detail.

In this embodiment, the input color image data is given as 6-bit data for each of R, G and B. This is R,
If G and B are respectively assigned to the orthogonal coordinate axes, they can be considered as one vector in the three-dimensional space having the origin as the starting point. Considering the two color information, it can be considered that the two vectors have the same color when they match.

However, there is always some fluctuation in the general input system. Therefore, in this embodiment, two colors are provided to identify the colors.

First, the first criterion is to use the above two color vectors Then When the above condition is satisfied, different colors are determined. |
Since | is considered to be the brightness of that color, a different code is assigned when the brightness differs to some extent.

Another criterion Is. The numerator in the above formula represents the inner product of two vectors. That is, the left side of the above equation represents the angle formed by the two vectors. 2 here
If the directions of the vectors differ by a certain degree or more, they are regarded as different colors.

FIG. 4 shows these vectors in a three-dimensional space, and each coordinate axis represents R, G, B. In the figure Are regarded as different colors because the difference δ in their lengths is δ ₀ or more. Also The angle formed by and is greater than θ _{0, and} is therefore considered to be a different color.

FIG. 5 is a flowchart showing details of this color identification.

Here, [i] represents a conversion table, N [i] represents how many times the i-th color is input, i is a variable representing the sign of the latest registered color vector, and n is the registered color. Represents a number.

First, [i], N [i], i, and n are cleared and initialized (step 501). When the color vector v of one pixel of the color document image is input from the scanner (step 50
3), the calculation of Δ (, [i]) = ||||| (i) || is performed based on the equation (1) (step 505), and the comparison result δ and the threshold δ ₀ are compared. Is performed (step 507), and the calculation result δ is the threshold value δ _0.
In the following cases, the calculation of θ = θ (, [i]) = cos ^-1 [(, [i]) / ||| [i] |] is performed based on the equation (2) (step 509), calculation result θ and threshold θ
A magnitude comparison with ₀ is performed (step 511). Step 5
When it is determined that the calculation result δ is larger than the threshold value δ _{0 in} the determination of 07, and when it is determined that the calculation result θ is larger than the threshold value θ _{0 in} the determination of step 511, and [i] are After being searched for a different color and searching for a registered color by a loop using a for statement (step 513),
A code is output (step 515).

If the calculation result θ is equal to or smaller than the threshold value θ _{0 in} step 511, it is regarded as the same color as [i] and is output as a code (step 515).

Next, [i] and N [i] are [i] = ([i] .N [i] +) / (N [i] +
1) The registered color vector is corrected by converting N [i] = N [i] +1 (step 51
7) The above operation is repeated until it is determined that the processing is completed (step 519).

In this way, a code is assigned to the color used in the target color document image.

In this embodiment, although the color is identified by the above algorithm, it is also possible to identify the color from the frequency distribution of each RGB color component.

Next, another embodiment in which the first conversion unit 2 and the second conversion units 4a and 4b are fixed tables will be described.

FIG. 6 is a block diagram of the first conversion unit 2 of this embodiment. As shown in FIG. 6, this conversion unit is composed of four random access memories (RAM) 22a to 22d, and the scanners 11 to R are provided. , G, B are input with 6 bits each
The 18-bit color document image data is input as an address to each RAM 22a-22d. A conversion table is stored in advance in each of the RAMs 22a to 22d, and the input 18-bit color document image data is converted by this conversion table, and 1-bit data is output from each of the RAMs 22a to 22d. It is output as encoded image data.

The above conversion table is supposed to be input in advance.
Colors expressed in 18 bits, which correspond to one piece of coded image data expressed in 4 bits, with multiple pieces of color document image data expressed in 18 bits with similar hue and gradation. Document image data is converted to 4 by this conversion table.
Converted to bits.

FIG. 7 is a block diagram of the second conversion unit 4a (or 4b) of this embodiment, and this second conversion unit 3a is composed of three RAMs 42a-42.
The conversion table is stored in advance in each of the RAMs 42a to 42c, and the 4-bit encoded image data is stored in each of the RAMs 42a to 42c.
When input to the RAM 42c, it is converted by the conversion table in each of the RAMs 42a to 42c, and 8-bit color document image data of R, G, B is output from each of the RAMs 42a to 42c.

The above conversion table maps color image data corresponding to 4-bit encoded image data to color document image data represented by 24 bits. The encoded image data represented by 4 bits is By conversion table
Converted to 24 bits. The number of bits of the color document image data can be appropriately changed depending on the color configuration of the output device.

Further, in the above embodiment, since the color image data is encoded by 4 bits, the colors that can be identified are 16 colors. However, if the number of encoding bits is increased, the number of colors that can be identified can be further increased.

[Effects of the Invention] As described above, according to the present invention, since color image information is coded, the amount of hardware required for advanced and extremely efficient color document image processing is small, and therefore at a very low price. realizable.

[Brief description of drawings]

FIG. 1 is a block diagram showing the basic configuration of an embodiment of the present invention, FIG. 2 is a configuration diagram of the same embodiment, FIG. 3 is a schematic flowchart of color identification in the same embodiment, and FIG. 4 is a three-dimensional space. FIG. 5 is an explanatory view of a color vector and its identification method in FIG. 5, and FIG. 5 is a detailed flowchart of color identification code allocation of the same embodiment,
FIG. 6 is a configuration diagram of a first conversion unit according to another embodiment of the present invention,
FIG. 7 is a block diagram of the second conversion unit of the embodiment. 2 ... 1st conversion part 4a, 4b ...... 2nd conversion part 6 ... Control means

Claims (1)

[Claims] 1. An input means for inputting n-bit color image data, and a plurality of memories in which a conversion table is stored in advance, wherein the n-bit color image data input by the input means is stored. By inputting to each of the plurality of memories and outputting 1-bit encoded image data from each memory based on the conversion table stored in the plurality of memories, m bits (m <n) are formed. First conversion means for converting into encoded image data, and enlargement / reduction means for performing enlargement / reduction processing and compression / decompression processing for the m-bit encoded image data encoded by the first conversion means. An image processing unit having a compression / decompression unit and a plurality of memories in which a conversion table is stored in advance, and an m-bit image processed by the image processing unit. Second conversion in which encoded image data is input to each of the plurality of memories and converted into color image data of 1 bit (where l> m) based on the conversion tables stored in the plurality of memories Means, output means for outputting the 1-bit color image data converted by the second converting means, and the image processing of the m-bit encoded image data converted by the first converting means. A data transfer bus having an m-bit width for transferring the m-bit encoded image data image-processed by the image processing means to the second converting means. A color document image processing device characterized by: