JPS61148971A - Image processing system - Google Patents

Abstract

PURPOSE:To enable a new processing of the image information as using the identification information by recording the identification information in the image file along with the image signal. CONSTITUTION:An image processing part 44 performs the binarization pro cessing with a constant threshold and the half-tone processing simultaneously per each specified block of the image signal inputted, and further selects either one of the two binarization outputs in accordance with the result of the image acknowledgement of each block. The processing part 44 additionally stores in the image file 4 binarization image signal and the identification information that identifies the one of the two binarization processings that processed the binarization image signal. By virture of the above, when reproducing in the decoding part, the dither density correction processing can be achieved only on the part of the half-tone image.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an image processing system, and more particularly to an image information processing system capable of storing input image information.

[Prior Art] In view of the fact that the original input image of a car may have different types of images depending on the area, conventional devices of this type recognize the image tone of each area of the input image signal and perform image area separation. I was doing image processing. However, even if you try to store the image information after image area separation in another storage means and then read out the stored image information and perform new processing according to the image tone, it is difficult to recognize the image tone. Since the type information is lost, there is a drawback that new image processing cannot be performed without extracting some feature from the read image information.

1. The present invention was made in view of the above-mentioned drawbacks of the prior art, and its purpose is to provide an image processing system that can store both image information and identification information that identifies the image tone from the image information. It's there.

Embodiment L The present invention will be described in detail below with reference to one drawing. FIG. 1 is an external connection diagram of an image processing system according to an embodiment. l is
A microcomputer for system control and RAM,
A control unit (referred to as a work station) that is equipped with an internal memory consisting of ROM, etc., and an external memory consisting of a floppy disk or cartridge disk, etc.
2 is the image input section of a so-called digital copying machine,
A document reader converts the document information of the document placed on the document table into an electrical signal using an image sensor such as a COD, and 3 is an image output section of a digital copying machine, which converts the document information of the document placed on the document table into an electrical signal by a laser beam printer or the like. This is a high-speed printer that records images on recording materials based on information. 4 is an image file having a storage medium such as an optical disk or a magnetic disk, and in which a large amount of image information can be written and read. Reference numeral 5 denotes a microfilm file, which includes a microfilm search section and a microfilm league section that converts image information on the searched microfilm into an electrical signal using an image pickup device. Reference numeral 6 has a photosensitive belt in which a photoconductive layer is provided on a transparent conductive band-shaped substrate, and image light is generated by irradiating the photoconductive layer with a laser beam modulated according to an input image signal through the substrate. An electrostatic WI image corresponding to the brightness and darkness of the image is formed on the photoconductive layer, and the formed Wi image is developed with a toner (developer) having conductivity and magnetism held on a toner carrier to form one display image. It is a high-resolution soft display that forms.

Reference numeral 7 denotes a printer device such as a laser beam printer similar to the high-speed printer 3, but it is smaller and slower than the high-speed printer 3, and is installed as necessary. 8 is a CRT device that displays image information photoelectrically read by a digital copying machine and a microfilm file input scanner (LEAGUE), or system control information, and a display RAM 8bft@etet that stores image information to be displayed. There is. Reference numeral 9 denotes a switching device that switches the connection between each input/output device using a signal from the control section l. Cables 10 to 18 electrically connect each input/output device.

Further, 19 is a keyboard provided in the control unit l, and by operating this keyboard 19, system operation commands etc. are issued, 20 is an operation panel for issuing operation commands for the digital copying machine, It has keys for setting the number of copies, copy magnification, etc., a copy key 21 for instructing the start of copying, a numerical display, and the like. Reference numeral 22 denotes a mode changeover switch which will be described later, and 23 and 24 represent indicators made of light emitting diodes (LEDs) for displaying the mode selection state of the mode changeover switch 22.

FIG. 2 is a block diagram showing the circuit configuration of the image processing system shown in FIG. Each block corresponding to that in FIG. 1 is given the same number as in FIG. First, each block within the control section l will be explained. 31 is a keyboard, °first
Corresponding to the illustrated keyboard 19, the operator inputs system operation commands using this keyboard 31. 32 is a central processing unit (hereinafter abbreviated as CPU) consisting of a microcomputer; 33 is a read-only memory (hereinafter abbreviated as R0M) in which a system control program is written in advance; The controller operates according to the program written in the controller. 34 is a random access memory (RAM), which is mainly used as a working memory of the CPU 32 and a page memory for storing image signals exchanged with each input/output unit. 35
is an external memory consisting of a floppy disk, which stores a system control program, a database for image retrieval from image files 4, which will be described later, and the like. Reference numeral 36 denotes a communication interface that enables information to be exchanged with other similar systems or terminals using a communication line such as a local area network. 37 is an input/output interface that accomplishes information exchange between the control unit 1 and the switching device 9. 3B is a bit extraction circuit that thins out the image signal according to a predetermined rate; 39 is an optical disk interface for exchanging information with the image file 4; and 40 is a CRT interface for exchanging information with the CRT 8. Reference numeral 41 denotes a 16-bit bus through which signals are transferred between each block within the control section l.

A digital copy A42 consisting of a document reader 2 and a high-speed printer 3, a microfilm file 5, a soft display 6, a small printer 7, etc. are connected to a switching device 9 by cables 11, 12, 15, 16° 10, respectively, and are switched. The device 9 is connected to the input/output interface 37 of the control unit 1 via cables 13, 14a, and 14b. Reference numeral 44 denotes an image processing section, the details of which will be described later. Furthermore, 1
As mentioned above, 1-18 is a cable that electrically connects each input/output device, through which control signals and image signals are transmitted. Arrows on each cable 11-18 indicate the flow of image signals, and cables with two arrows allow control signals to flow in both directions. Further, the image file 4 and the CRT device 8 are connected to each interface 39. of the control unit 1 via cables 17.18, respectively.
40. Furthermore, since the image signal input/output by the switching device 9 is a serial signal, and the information on the path 41 of the control unit 1 is a parallel signal, the input/output interface 37 has a serial/parallel register for capturing the image signal. and a parallel/serial register for outputting image signals.

Next, the control operation will be explained.

Image signals outputted by the original reader 2 or the microfilm file 5 are input to the input/output interface 37 of the control unit 10 via the switching device 9 for every l line. The input/output interface 37 converts the serially inputted image signal into a parallel signal of every 16 hits and outputs it onto the path 41. The image signal output on path 41 is stored in RAM3.
One page of data is primary input to the image area No. 4. The ij [signal thus stored in the RAM 34 is outputted again to the path 41, outputted to the outside via the communication interface 36, written to the optical disk of the image file 4 via the optical disk interface 39, or , and is outputted to the switching device 9 via the input/output interface 37 and selectively transmitted to the high-speed printer 3, the soft display 6, or the small-sized printer 7, thereby forming an image. At this time, when performing image processing such as binarization processing/density adjustment of the 4-output image, the image signal written in the RAM 34 is sent to the image processing section 44 and after undergoing 1 image processing, the image signal is sent via the human output interface 37. The data is transmitted to and output from the high-speed printer 3 mentioned above.

Further, the image signal read from the optical disk of the image file 4 is written to the RAM 34, and then sent to the Fj'J conversion device 9, the high-speed printer 3. It is also possible to selectively transmit the information to the soft display 6 or small printer 7.

Incidentally, the image signal from the document reader 2 or the microfilm file 5 is selectively sent to the high-speed printer 3. It is also possible to directly transmit to either the soft display 6 or the small printer 7. In other words, if a simple copy operation is desired, image file 4 or CR
Since T8 is not required, image signals from, for example, the document reader 2 are directly supplied to the high-speed printer 3 without going through the control unit 1, and a real-time copying operation is executed.

The control related to the transmission of the image signal described above is performed by the CPU 32 according to the operation command input by the operator using the keyboard 31.
is executed.

Next, referring to FIG. 3, a 4x4 dither matrix (Ba7er) used in the image processing unit 44
The principle of image density correction processing will be explained below.

FIG. 3 shows a dither matrix pattern applied to the embodiment.

In this figure, 51 is a pixel of the input image read out from the image file 4, and each numerical value represents a threshold value of the image level.
is a 4x4 dide matrix, where each value represents vA, and 53 indicates that the pixel 51 is binarized by M[i of the dither matrix 52, and the display state of each display cell is determined in accordance with the 4x4 dither matrix 52. represents. Dithered image information is formed in response to turning on or off 53 and stored in the image file 4. In addition, A in the figure represents the main scanning direction, and B represents the sub-scanning direction. Note that the pixel 51 has 16 pixels from 0 to 15.
This shows a case in which a certain value can be taken.

The binarization method is performed as follows. The threshold value in the fourth row and first column of the teaser matrix 52 is 15, and the threshold value of the corresponding pixel 51 is 4, so the display cell of the display device 53 is turned off (black). In this way, sequentially input images are converted into dithered images for each pixel 51 by the 4x4 teaser matrix 52. In this case, when comparing the 4x4 dither matrix 52 and the pixel 51, 7 of the 4x4 display cells are turned on, and 7/16 of all display cells are turned on in the entire 0 display screen.

Therefore, the teaser image signal read out from one image file 4 can express halftone shading by using 4 bits in the main scanning direction A and 4 lines in the sub-scanning direction B as one pixel level. In other words, in order to adjust the shading of a dithered image, it is sufficient to regard 4x4 bits as one wheel for the main scanning and sub-scanning of the image signal, and invert the pixels in the order of the weights of the dither matrix. Specifically, when using the dither matrix to make the image signal (shown in FIG. 4 (L)) that is black up to the bit with weight 8 one level darker by one step, the bit with weight 9 is changed to black (shown in FIG. 4 (L)). It is only necessary to invert from white to black as shown in Figure 4 (l) of the bitl with weight 8 to make it thinner by one step. By performing the processing on all pixels in 4x4 units, the shading of the image can be adjusted.

The above is the binarization by dither processing performed on the halftone image by the image processing unit 44. Incidentally, in this embodiment, in addition to the above-mentioned halftone processing, 2-digit conversion using a defined tIl value, which will be described later, is also performed. The z-valued image signal may be transmitted via the communication interface 36 as described above, or may be output to the high-speed printer 3 or the like via the input/output interface 37.

However, halftone processing is not suitable for all input images; for example, if halftone processing is applied when the input image contains text, the resolution will deteriorate, and when halftone processing is performed using the dither method, Even if the black and white parts in the original image were separated, the black and white parts would be bivalentized separately as shown in Figure 4 (IiL) to (C), and the original image would have The degree of redundancy decreases, especially the pressure II rate in the case of run-length encoding, and a large amount of file area is required when storing in the image file 4 or the like.

Therefore, in the image processing section 44 of this embodiment, input processing is performed.

Bi-polarization processing using a fixed threshold value and binarization using halftone processing are simultaneously applied to each predetermined block of the image signal to be processed, and one of the two polarization outputs is determined according to the image tone recognition result for each block. Choose one. Then, the binarized image signal is stored in the image file 4 together with identification information identifying which binarization method was used to binarize the binarized image signal.

If the binary image signal is stored in the image file 4 as described above, it is possible to perform dither density correction processing only on the halftone image, for example, when the signal is reproduced by the decoding section described later. In addition, if two types of binary image signals are encoded using band compression, the run length can be reduced by applying the same binarization as halftone images even in areas with high redundancy such as character parts. Since no reduction occurs, redundancy can be eliminated at the maximum compression rate, and the capacity of the image file is also saved. Further, even if the compressed image signal is directly sent as a facsimile signal to the communication interface 36, there is an advantage that the line usage efficiency is improved.

FIG. 5 is a block diagram of the image processing section 44.

In the figure, image information input from the input/output interface 37 is divided into predetermined blocks by an image area separation processing section 51.
It is determined for each block whether the image signal should be subjected to halftone processing or the image information should be binarized using a fixed threshold value, and is sent to the fixed threshold binarization unit 52 and the halftone image processing unit 54. - The threshold binarization unit 52 and the halftone image processing unit 54 send the image signals binarized using the respective two-tilting methods to the image information selection processing unit 55. -・10,000, image area separation processing unit 5
The judgment result of 1 is the image processing CPU 59 (this image processing C
PU59 is rIS2 diagram (7) CPU32. ! :I'ff
1-1'))
and is used to select the two binarized signals.

Note that the determination in the image area separation processing unit 51 may be made based on, for example, whether the sum of density differences between adjacent pixels in a block has reached a predetermined value of 11 or not.

The binarized image signal selected by the image information selection processing section 55 is sent directly to the optical disk interface 39, or is compressed and encoded by the encoding processing section 56 and then sent to the optical disk interface 39. The information written to the image file 4 via the optical disk interface 39 includes, in addition to the unencoded binary image signal or the encoded binary image signal, the two types of binary image signals. The advantage of recording this identification information together with the binary image signal, including the identification information representing the type and the pixel address in the RAM 34 of block 71, is as described above.The reason for recording the 0 pixel address is explained below. by.

Fig. 7 shows the conceptual layout of a binary image when the image signal for one page in the RAM 34 is binarized. This is a set of pixels that have been converted into z-values.

In the Mokufu embodiment, the image area determination of the input image signal in the image area separation processing unit 51 is performed within a rectangular block, and the central pixel within the rectangular block is binarized according to the determination result. Therefore, such a set of center pixels is represented by a set of the start address of the shaded area and its length, as shown in Figure 7, and these can be written to the image file 4 line by line. Become.

FIG. 6 shows the operation of the image processing unit 44 by the image processing CPU.
59 control procedures.

This control procedure is carried out in the ROM 60 (RO) in the image processing section 44.
M60 may be the same as the ROM 33 in FIG. 2).

Next, the operation when such a configuration is adopted will be explained according to the flowchart shown in FIG. 6.

First, in step 100, the gradation image signal is input line by line from the RAM 34 to the image area separation processing unit 51 via the input/output interface 37.Next, in step 102, the image area separation processing unit 51 performs image area determination. , constant threshold value 21 straightening unit 5
2 and halftone image processing section 54 are made to binarize their respective image signals. As mentioned above, the image area separation processing section 51
Then, using a line memory or the like, divide the input image into rectangular blocks, determine the image tone within each block, and sequentially convert the central pixel within each block to the constant partial threshold E-binarization unit 52 and the halftone block. The image is sent to the image processing section 54. The sent image signal of the center pixel is applied to a fixed threshold value 2 polarization unit 5
In No. 2, the image is binarized using a constant threshold value, and in the halftone image processing unit 54, the image is binarized using a dither matrix.

Next, based on the determination sent from the image area separation processing unit 51 in step 104, the image processing CPU 59 causes the image information selection processing unit 55 to select one (steps 106 and 108).
, Next, in step 110, it is checked whether encoding for compression is to be performed. Whether or not this encoding is to be performed may be instructed by the operator in advance from the keyboard 31, etc. The 2-bond image information that is not encoded may be stored within the same image processing system (i.e., not through a communication line). This is effective because there is no need to decode the data when it is transmitted and output to, for example, a high-speed printer 3 (the time required for transmission is relatively less of a problem even if the data is not encoded). If encoding is to be performed, the encoding is performed in step 112.
0.14 encoded 211 image image values or uncoded binary image signals are written into the image file 4 via the optical disk interface 39. Furthermore, in step 118, identification information indicating the type of binary conversion (in this embodiment, two types of binary conversion are illustrated, so 1 bit is sufficient), and the start address and address of the area as explained in FIG. The length and, if necessary, a flag indicating whether or not encoding has been performed are created as index information, and in step 118, this index information is written one after another into the RAM 61 (or RAM 34).

The f# control from steps 100 to 118 described above is performed every time one line is input, and continues until one page's worth of image information is processed (step 120). When one page's worth of image information is processed, the RAM 61 ( Or the Intex information stored in the RAM 34) can be transferred to the optical disk interface 3.
9 to the image file 4, and the process ends. Of course, if you input the image of the next page, step l
Repeat from OO.

As explained above, according to this embodiment, if the image tone identification information of one image value number is stored in the image file, which is the storage means, together with the input image signal, the identification information can also be obtained when the image signal is read out later. Image processing can be performed efficiently again according to the identification information. - The same effect can be obtained even when the identification information in 1fiJ is stored together with the image signal converted into 24t1 by a different 24t1 conversion means.

Furthermore, if the image signal that has been binarized using a different binarization method is further encoded, the run length will not be shortened, so efficient information compression becomes possible, and it can be transmitted over a single communication line or stored in an image file. The system can be operated efficiently even when

In this embodiment, the dither matrix used in the halftone image processing section 54 is of the Bayer type, but the same applies to the 2-dot medium type and the spiral type. In addition, although the binary A5 method is described as two, multiple threshold matrices may be used for halftone processing to use multiple binarization methods. A plurality of division methods may be used to set .

[Effects] As explained above, according to the image processing system of the present invention, by recording the identification information of the binarization method suitable for the image tone of the input image signal in the image file together with the image signal, it is possible to extract information from the image file. Since the identification information can also be obtained when reading the image information, the image information can be newly processed using the identification information.

In addition, if binarization is performed according to image tone recognition, it is possible to binarize the input image without reducing its redundancy, so even if encoding is performed, the effect of redundancy compression will be minimal. is obtained.

[Brief explanation of drawings]

Fig. 1 is a system configuration diagram of the embodiment, Fig. 2 is a block circuit diagram of the embodiment, Figs. 3 and 4 (a) to (C) are diagrams explaining image density correction processing using a dither matrix, @ FIG. 5 is a block circuit diagram of the image processing section 44, FIG. W46 is a flowchart showing the control procedure of the image processing section 44, and FIG. 7 is a diagram showing image information separated and displayed according to the type of binarization. In the figure, 4...Image file, 32...CPU, 34
．． 61... RAM, 36... Communication interface, 37... Input/output interface, 39... Optical disk interface, 44... Image processing unit, 5... Image area separation processing unit, 52...・・Constant threshold value binarization unit,
54... Halftone image processing section, 56... Encoding impossible section, 55... Image information selection processing section, 59. . . . image processing CPU, 60 . . . ROM. Figure 4 (a) (b) (C) Figure 7

Claims (3)

[Claims] (1) It has an image tone identification means for identifying an image tone within a predetermined area from a gradation image signal, and a storage means for storing image information,
An image processing system characterized in that the storage means stores identification information from an image tone identification means together with image information. (2) Furthermore, a plurality of binarization means for binarizing the gradation image signal, and any one of the binarized images binarized by the binarization means according to the identification result of the tone identification means. 2. The image processing system according to claim 1, further comprising a selection means for selecting a signal, and the image information stored in the storage means is the selected binary image signal. (3) A patent claim further comprising encoding means for compression encoding the binary image signal, and the image information stored in the storage means being the compression encoded binary image signal. The image processing system according to scope 2.