JP2006140970A - Image processing apparatus, image input/output device and methods therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To effectively utilize a scan function of a copy machine having an image region separation processing function. <P>SOLUTION: If a scan controller 11 outputs a scan instruction to a color copy machine 12, the color copy machine 12 performs image region separation processing on scanned image data, applies image processing based on a resultant image region signal sen, outputs the image data to the scan controller 11 and outputs a refined image region separating signal as an attribute map. The scan controller 11 refers to the attribute map indicating character regions and photograph regions and performs image processing while switching parameters of binarizing processing or compressing processing. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus and system, for example, an image processing apparatus connected to an external apparatus via a network or another interface (I / F), and an image processing apparatus that transmits a read image to the external apparatus as digital image data. The present invention relates to an output device and a method thereof.

  In recent years, there is a digital color copying machine that performs color image copying by color-separating a color original and electrically reading it, and printing out the read color image data on a recording sheet. Instead of connecting an external device such as a PDL controller to such a digital color copying machine and reading a document image, a color image printing system for printing out color image data generated by the external device on a recording sheet has been developed. Yes.

  Conventionally, in this color image printing system, as proposed in, for example, Patent Document 1 and Patent Document 2, an image output when a feature for each pixel of an image is extracted and printed out using an image area separating unit. There is a means of using it for image processing for improving the quality of the image.

  FIG. 11 shows a basic form of a conventional color copying machine, in which a YMCK input image from an external device (1101) is converted into RGB by a density / luminance conversion unit 1102. The image area separation unit (1107) performs feature extraction such as edge detection (1108) and thickness determination (1109) on the basis of information from the surrounding pixels for each pixel. It is determined whether it is a graphic area or a photographic area. The determination result is input to the smoothing circuit 1104 by the signal sen. The smoothing circuit 1104 also receives a density signal converted by the luminance density converter 1103. The smoothing process (1104) performs smoothing processing only on the character / graphic area indicated by the signal sen, thereby smoothing the edges and obtaining a good image on the printer unit (1106). (Refer to patent document 1 etc.).

On the other hand, there is also an image reading system in which a part of a PDL controller as an example of an external device is connected as a scan controller to a digital color copying machine or the like and an image reading command can be issued from the PDL controller to the color copying machine. Some of such image reading systems also function as a scan controller that can receive an original image read by a scanner unit of a controller digital color copying machine as digital image data via an I / F. The read image is sent to a terminal such as a host computer connected to the scan controller, or stored in an HDD or the like built in the scan controller. In the case of a color image, the read image data is, for example, image data of 24 bits per pixel (8 bits for each color of RGB), and the user can freely use it from a host computer or the like.
JP 11-127353 A JP 2000-259819 A

  However, in this conventional image reading system, the compression method, compression rate, number of gradations, resolution, etc. when storing or using the image data sent by the scan controller are adapted to use. In order to change to the above, the image data must be analyzed again on the scan controller side, or a fixed value predetermined by the user or the like must be used. Therefore, in the scan controller, the processing time is prolonged by duplicating the image area separation processing performed in the color copying machine, or the final processing is performed by performing image processing such as compression using a fixed value. There is a problem that the quality of the output image is degraded.

  The present invention has been made in view of the above-described conventional example, and effectively uses the result of image analysis performed by an image input apparatus having a built-in image processing function such as a color copying machine, thereby shortening the processing time and improving the image quality. An object of the present invention is to provide an image processing apparatus, an image input / output apparatus, and a method thereof.

  This invention is made | formed in view of the said prior art example, and is provided with the following structures.

An original reading unit that reads an original image and generates image data composed of pixels, and an attribute map generation that generates an attribute map indicating an image attribute for each pixel or block of the image data based on the image data And an image input / output device connected to the image input / output device, the image data including the image data and an attribute map corresponding to the image data. An image processing apparatus for processing data based on the attribute map, wherein image compression processing corresponding to an attribute indicated by the attribute map corresponding to a target pixel or block is performed on the target pixel or block Image processing means to be applied;
And storing means for storing compressed image data of the pixel of interest or the block of interest compressed by the image processing means.

Alternatively, a document reading unit that reads a document image and generates image data including pixels;
An attribute map generating means for generating an attribute map indicating an attribute of an image for each pixel or block of the image data based on the image data;
Output means for outputting the image data and an attribute map corresponding to the image data.

  According to the present invention, it is possible to effectively use the result of image analysis performed by an image input apparatus having a built-in image processing function such as a color copying machine, and to realize reduction in processing time and improvement in image quality. it can.

<Configuration of color copier>
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. FIG. 2 is a block diagram of the digital color copying machine 12 that functions as an image reading apparatus constituting the image reading system of the present embodiment. The digital color copying machine 12 includes a CPU and a memory in addition to the image processing unit 220.

  The digital color copying machine 12 receives raster image data (image data composed of pixels) from an external device 1 (200) such as a PDL controller to the image input I / F unit (201) in addition to copying. Similarly to the image forming apparatus described with reference to FIG. 11, the printer unit (206) can form an image on a recording sheet. Image signals Y, M, C, and K input from the external device 1 (200) are sent to the density / luminance conversion unit (202) via the image input I / F (201), and the density / luminance conversion unit (202). Then, it is converted into RGB so-called additive mixed color three primary color signals by the lookup table ROM.

  In the case of a copy operation, the original image set on the platen is read by the CCD line sensor (207) for each pixel RGB signal and converted into a digital signal by the A / D converter (208), followed by shading and sensor colors. Correction (209) is performed to generate an RGB signal. The RGB signal is converted into Y, M, C, and K signals based on the so-called subtractive color mixing principle in the luminance density conversion unit (203), and is input to the smoothing circuit (204). The smoothing circuit (204) switches 400 lines / 800 lines for each pixel according to the image area separation result (SEN signal) of the image area separation unit (212) described later, and doubles the resolution of the input image. Generate data with In the γ conversion unit (205), the density data of each resolution is converted according to the gradation reproduction of the printer. The YMCK image signal processed in this way and the SEN signal, which is a 400/800 line switching signal, are sent to the laser driver, and density recording by PWM is performed in the printer unit (206).

  Here, the Y, M, C, K signals input from the image input I / F (201) and the Y, M, C, K signals output to the printer unit (206) In the case of a color electrophotographic system, the colors are sent one by one in the frame order. Further, when the printer is a four-drum color electrophotographic system, it may be determined by the printer and its control system, such as providing a delay according to the print timing of each color and sending the four colors in parallel. .

  The I / F between the external device 1 (200) and the image input I / F unit (201) is a video I / F that sends pixel signals in synchronization with the clock, and is externally synchronized with the printing timing of the printer (206). Although YMCK image data may be sent from the device 1, if the image input I / F unit (201) has a sufficient frame memory, the image data is received asynchronously with the printer (206). It can also be done. In the latter case, it is sufficient to select a physical I / F that has a sufficient data transfer speed, such as Ethernet (registered trademark) or DPRAM, and the degree of freedom increases, but the image input I / F unit (201) Costs will increase.

  FIG. 12 is a block diagram schematically showing an electrical configuration of a print controller (1220) which is an example of the external device 1 (200). In the case of a print operation, the host computer (1210) is a supply source of print data (job). Jobs sent from the host computer (1210) are usually described in a page description language (PDL) format such as postscript, and the printing mode such as the paper feed stage and the color mode is used for stapling, saddle stitching, etc. You can also specify the finishing action.

  The controller (1220) temporarily holds the print data supplied from the host computer (1210) via the cable (1240) and the external interface (1225) in the built-in HD (1224) via the HDD controller (1223). The print data held in the built-in HD (1224) is temporarily held in the PDL buffer (1226-2) on the built-in RAM (1226) via the CPU bus (1227). The controller (1220) expands the PDL data held in the PDL buffer (1226-2) to the frame memory (1226-1) and generates image data. At this time, full-color or single-color gray scale image data is generated according to the PDL data.

  The image data developed in the frame memory (1226-1) is sent to the color copying machine 12 functioning as the image forming apparatus (1230) via the cable (1250). Prior to that, information necessary for printing such as a paper feed stage and a print mode described in the PDL data is a color copier that also functions as an image forming apparatus (1230) via a cable (1250) as a command. 12 is sent. In the color copying machine 12 of this embodiment, a CPU (not shown) controls a job to be sent. Therefore, commands such as print information and a print start command are sent to and interpreted by a CPU (not shown) built in the color copier 12 functioning as the image forming apparatus (1230). Here, the cables 1240 and 1250 can be constituted by general-purpose cables such as a parallel cable, a SCSI cable, a serial cable, a network cable, or a combination thereof, or a predetermined dedicated cable.

  When the CPU (not shown) of the color copying machine 12 sends a print start command to the printer unit (206) and receives a signal indicating the printable timing from the printer, the image input I / F unit is synchronized with the timing. By (201), the developed image data is received from the print controller. The received image data is transmitted to the printer unit (206) as described above. At this time, the color copying machine 12 sends a synchronization signal for notifying the print controller (1220) of the printable timing. The print controller (1220) transmits image data for each page in synchronization with the synchronization signal, and can print by transmitting this data to the printer unit (206) via the image processing circuit shown in FIG. The image data is transmitted for each page at a proper timing.

  When the image input I / F unit (201) includes an HDD or memory for holding image data having a sufficient capacity, the developed image data sent from the controller (1220) is temporarily stored in the HDD or memory. It is also possible to send the image data by synchronizing the image input I / F unit (201) with the printable timing of the printer unit (206). As a result, the controller (1220) can sequentially transmit images from the developed pages without worrying about the timing of the printer (206) when transmitting image data.

  Here, at the time of copying, the RGB signal read by the CCD sensor (207) is scanned four times according to the print timing of each color of Y, M, C, and K in the case of a one-drum printer. It may be read or stored in a buffer memory (not shown) after the shading sensor color correction (209) and read in accordance with the timing of each color. Alternatively, Y, M, C, and K are simultaneously processed, converted by the γ table (205), immediately stored in a buffer memory (not shown) immediately before being sent to the printer unit (206), and output in accordance with the form of the printer. A flexible configuration can be taken.

  The image area separation unit (212) extracts features for each pixel or block based on the input RGB signals. In this embodiment, is it a partial pixel of an object composed of characters, lines, etc. (character pixel), or is it a partial pixel (photo pixel) of a bitmap image obtained by reading a picture or picture? For example, 1 is output as a 1-bit SEN signal if it is a character pixel, and 0 if it is a photographic pixel. Alternatively, the SEN signal is 2 bits, and each bit is assigned as a character bit and a photo bit. If both are 0, it is an area such as a background that is neither a character pixel nor a photo pixel. The detailed algorithm using the edge detection unit (212), the thickness determination unit (214), and the LUT (215) as an example is known as described in JP-A-11-127353 and the like. Let's briefly explain. When the edge of the image is detected by the edge detector 212, the direction of the edge is determined. Further, the thickness determining unit 214 determines the spread of the image such as the line width. If an edge in one direction is detected by the edge detection unit 212 and the edge has a certain width or less, it is determined that the pixels inside the edge belong to the character area (for example, sen signal = 1). The The other pixels are determined to be non-character areas, that is, photographic image areas (for example, sen signal = 0).

  The color copying machine 12 of this embodiment operates not only as an image forming apparatus but also as an image reading apparatus. In this case, based on a command from the external device 2 (211) such as a scan controller, the RGB signal of the image data of the document read by the CCD sensor (207) is sent to the printer unit (206) as in the case of copying. Instead, the image data is sent to the external input device 2 (211) such as a scan controller via the image output I / F (210). Here, the external device 1 (200) and the external device 2 (211) are shown and described as separate devices in FIG. 2 for ease of explanation, but both have an input / output I / F and are simple. One controller can have both functions and can be configured as one external device. In the present embodiment, the print controller and the scan controller are used for the purpose of ensuring compatibility by converting the data format between the color copying machine 12 and a host computer or the like.

  The color copying machine 12 starts a reading (image scanning) operation using the CCD sensor (207) when a built-in CPU (not shown) receives a scan start command from the external device 2 (211). Data can be transmitted (so-called pull scan), a reading operation is started in response to a user instruction from an operation unit (not shown) provided in the color copying machine, and a request is made to the connected external device 2 (211). It is also possible to adopt a configuration (so-called push scan) in which image data is forcibly sent.

  In the present embodiment, the scan controller 11 that is the external device 2 (211) has the configuration shown in FIG. As shown in FIG. 1, the scan controller (11) has exactly the same hardware configuration as the print controller (1120) shown in FIG. The same I / F cables 13 and 1240 with the host computer (10, 1210) can be used. That is, the program operated by the CPU (111, 1221) and the driver and application operated by the host computer (10, 1210) are switched to scan and print, and the copying machine is connected to the external interface (115, 1225). If the functions for both the image input I / F (201) and the image output I / F (210) are provided, the print controller (1220) and the scan controller (11) have the same hardware as described above. It can also be a controller.

  Further, the part of the external interface (115, 1225) to the copier 12 that is particularly used for image transmission is capable of bidirectional data communication, and the image input I / F (201 on the copier side in FIG. 2 is used. ) And the image output I / F (210) unit can be integrated.

  In the present embodiment, when image data is sent to the scan controller 11 during scanning of an original image, the image area separation result by the image area separation section (212) is also displayed together with the image output I / F section (210). Send to. An image in which the sen signal value, which is the result of image area separation, is two-dimensionally arranged in accordance with the arrangement of the corresponding pixel or block of image data is called an attribute map.

  In the present embodiment, the attribute map receives the output of the sen signal from the image area separation unit (212) in FIG. 2, and stores it in a buffer memory (not shown) in the image output I / F (210) or image output. When the I / F unit (210) does not have sufficient memory, it is transmitted to the scan controller (11) as the external device 2 (211) as a real-time signal in synchronization with the image reading timing of the CCD sensor (207). The Then, the attribute map memory unit (116-2) in the RAM (116) in the scan controller (11) is finally constructed.

  The image output I / F unit (210) delays the RGB image signal by an image area separation process if the I / F with the external device (211) is a clock-synchronized video I / F, for example. The attribute map signal and the pixel unit are synchronized for transmission. If this I / F is a network I / F such as Ethernet (registered trademark), a packet combining RGB image data and attribute map data is created in accordance with the protocol to be used and transmitted to the external device 2 (211). To do. Thus, the image output I / F unit (210) takes a necessary form in accordance with the I / F with the external device 2 (211) actually used.

<Example of image area separation processing and attribute map>
FIG. 3 is a diagram illustrating an example of an attribute map associated with each pixel according to the present embodiment. In FIG. 3, an image in which the number 1 that is a character pixel is included in a background image that is a bitmap, that is, a photographic pixel is shown. If the sen signal generated by the image area separation unit represents character pixels as 1 and photographic pixels as 0, the attribute map having a two-dimensional arrangement of 0 and 1 shown in FIG. 3 is completed. Of course, the photographic pixel may be represented by 1 and the character pixel may be represented by 0. In this case, the attribute map is obtained by reversing the positions of 0 and 1 in FIG. The attribute map may be configured in any manner as long as the attribute map is stored so as to be associated with each pixel. For example, as an attribute map plane as shown in FIG. 3, the image data may be stored in the image memory 116-1 as shown in FIG. 2, and the attribute map may be stored in the attribute map memory 116-2. good. If they are stored in the same area of the memory (116), as shown in FIG. 4, four planes obtained by adding an attribute map plane to each of the RGB planes may be used as one page image (Planner). When RGB data is configured in one pixel as shown in FIG. 5, it may be embedded in a form added to the RGB information of each pixel (Chunky). Also, in order not to increase the amount of data, it may be necessary to embed an attribute map in the lower bits for each pixel of one or more of the RGB planes. You can decide the format according to your needs.

  Hereinafter, a specific example of the result of the image area separation processing in the image area separation unit (212) of FIG. 2 will be described with reference to FIGS. FIG. 6 is a view showing an original image of a document read by a scanner, that is, a CCD sensor (207). In FIG. 6, 61 is a character written as “ABC” and “Iroha”, and 62 is an object that is a photograph. As shown in the figure, the photo 62 is a manuscript in which a letter 63 written “H” is embedded.

  FIG. 7 is an example of an attribute map generated by inputting the image data of FIG. 6 to the image area separation unit 212. A portion drawn in black is a portion where the value of the attribute map is 1 (this attribute map value is referred to as an attribute value). In this case, the portion determined as the character pixel in the image of FIG. 6 becomes the black region of the attribute map of FIG. Note that the background pixels are determined to be 0 together with the photographic pixel area.

  FIG. 8 is another example of the attribute map generated from FIG. As in FIG. 7, the part that is black is the part that is 1 in the attribute map. In this case, the image area separation unit sets the sen signal to 1 corresponding to the pixels of the photographic image, and corresponds to the other pixels. The sen decimal is output as 0. As a result, in FIG. 8, the attribute value of the portion corresponding to the background, that is, the background on the original is 0 as in the case of the character pixel, and the attribute value is set to 1 only for the portion where the photographic image is drawn.

  In this way, it is possible to extract only character pixels or, conversely, extract only photographic pixels depending on the method of image area separation. Furthermore, it is also possible to send a 2-bit attribute map for each pixel having the two types of information shown in FIGS. 7 and 8 to the controller 11. In this way, the controller can obtain an attribute map consisting of 2 bits for each pixel. As a result, it is possible to determine whether each pixel of the sent scan image is one of a character pixel, a photographic pixel, and the other (background pixel).

  In general, the image area separation processing has a large calculation scale. In the scan controller 11, the CPU may be responsible for a lot of image processing in software, so that processing with a large amount of calculation such as image area separation processing is heavy. In addition, when image area separation is performed by the color copying machine 12, the processing is duplicated. On the other hand, in a digital color copying machine, image area separation processing is often performed in hardware by an electronic circuit, and image area separation processing can be performed in real time for a reading operation. Therefore, in the present embodiment, when image area separation is required for the scanned image by the scan controller 11, the image received on the scan controller 11 side is not processed and image area separation is performed, but digital color copying is performed. By using the image area separation circuit used at the time of printing in the machine 12 also at the time of scanning, the processing load on the controller side is greatly reduced.

  Next, a usage example of the received image area map in the scan controller (11) will be described. In the present embodiment, the image data received from the color copying machine (12) functioning as the image reading apparatus of FIG. 1 is a multi-value image of 24 bits in total for each RGB color of 8 bits. This image data is temporarily stored in the image memory (116-1). In addition, the attribute map shown in FIGS. 7 and 8 for each pixel of 2 bits is stored in the attribute map memory (116-2).

  In this scan controller 11, since the multi-valued color image data has a large capacity, it is binarized and stored as a binary color image in the built-in HDD (114) or sent to the host computer (10). To do. In such a case, the CPU (111) performs a well-known binarization process such as an error diffusion method or a dither process by a program stored in the ROM (112) in advance, and stores it on the image memory (116-1). It is necessary to convert the multi-value image data into binary image data and send it to the built-in HDD (114) or the host computer (10). At this time, the optimum binarization method varies depending on the attribute of the image. In other words, graphics such as characters and figures, ie, character pixels, are preferably binarized by reducing the dither matrix size and emphasizing resolution, and natural images such as photographs, ie, photographic pixels, are increased in matrix size. Therefore, it is preferable to emphasize the gradation reproducibility.

  Therefore, in the present embodiment, the 3 × 3 dither matrix shown in FIG. 9 is used for the character pixel area, and the 8 × 8 dither matrix shown in FIG. 10 is used for the photographic pixel area. Either can be used for other areas (background, etc.), but in general, the background is often white or solid in one color in a wide range. The dither matrix shown in FIG. 10 is used.

  In other words, in the present embodiment, in the attribute map shown in FIG. 7, the pixel having the attribute of 1 is binarized using the threshold matrix shown in FIG. 9, and the threshold matrix shown in FIG. Use to binarize. In this case, the attribute map of FIG. 8 is not used. The correspondence between the threshold matrix elements (thresholds) and the pixels uses the correspondence when each threshold matrix is arranged on the image to be processed without any overlap or gap. In this way, each pixel is associated with one element of the threshold matrix for each of the character image and the photographic image. Therefore, referring to the attribute value of the pixel, if the value indicates a character image (for example, 1), the corresponding element in the character threshold matrix is used as the threshold value, and an image with a value other than the character (for example, 0) is used. If indicated, the binarization process is executed using the corresponding element in the threshold matrix for photographic images as a threshold. The association is dynamically performed according to the progress of the process, but since the number of pixels in the vertical and horizontal directions of the original image data is known in advance, the position of the target pixel (the line number in the vertical direction and the horizontal direction) Is represented by the number of the pixel in the line.) For the components in each direction indicating (), the remainders modulo the vertical and horizontal sizes of the threshold matrix are obtained, and the position of the threshold matrix component to be used is represented in the matrix.

  As shown in FIGS. 9 and 10, the matrix in FIG. 10 has a larger calculation load because it is 8 × 8. When the processing speed is important or when it is known that the background portion is not halftone, the attribute map obtained by extracting only the photographic pixels of FIG. 8 is used, and the matrix of FIG. 9 may be applied, and the remaining region shown in FIG. 9 may be applied.

  Here, the multi-value data used in this embodiment is 8 bits for each color, that is, 0 to 255 levels. The pixel value is compared with the threshold value of the dither matrix component (cell) in FIGS. 9 and 10 corresponding to the pixel. If the value of the multi-value data of the pixel is larger than the threshold value, the dot is turned ON ( 1), and if it is below the threshold value of each cell, the dot is turned OFF (0).

  FIG. 9 shows a 3 × 3 size dither matrix. If the resolution of the image data to be processed is 600 dpi, the image can be reproduced with a resolution of one third of 200 dpi, and the number of gradations is 10 gradations. It is. Therefore, although the number of gradations is small, the image can be reproduced with high resolution. On the other hand, FIG. 10 shows an 8 × 8 dither matrix. If the resolution of the image data to be processed is 600 dpi, the resolution is 1/8 of 75 dpi, and the number of gradations is 65 gradations. Although the resolution is low, the number of gradations is large, and natural images can be reproduced with higher quality than the dither matrix shown in FIG.

  Thus, by switching the threshold matrix using the attribute map received together with the image data from the color copying machine 12, it is possible to form a good binarized image at high speed without imposing a load on the CPU (11). Thus, it is possible to output the character and line portions of the image with high resolution and the photo portion with high gradation while reducing the data capacity.

  Of course, the scan controller (11) does not binarize as in the present embodiment, but as a multi-valued image, for example, the character area is reduced to 32 gradations and the photographic area remains 256 gradations. As a matter of course, it is also possible to reduce the capacity of the color image data by reducing the photo area to 200 dpi while maintaining the high resolution 600 dpi equivalent in the character area.

  Furthermore, using a known data compression algorithm such as JPEG or JBIG, the character area is made reversible, and the photo area is made irreversible compression, for example, MAX 1/8 times. It is also possible to reduce the compression capacity of the image data while maintaining the image quality.

  Of course, it is also possible to determine a predetermined communication format, send the image data as it is to the host computer (10), and send an attribute map at the same time, and perform these processes by the application or driver software on the host computer. Is self-explanatory.

<System flow>
Next, a processing flow of processing executed in the system of this embodiment (comprising the color copying machine 12 and the scan controller 11) will be described with reference to FIG. FIG. 13 is a flowchart showing a processing flow executed in the present embodiment.

  First, in step S131, the application software on the host computer receives an instruction from the user and issues an image reading instruction to the scan driver (PlugIn, etc.). Of course, at this time, the user needs to set the document to be read on the document table of the color copying machine (12) in advance.

  In step S132, the controller (11) having received the scan start command to the external interface unit (115) via the interface (13) such as a network is stored in the ROM (112) operating on the CPU (111). In accordance with the instructions of the program, a scan command is issued to the color copying machine (12) connected to the external interface unit (115) via the cable (14).

  In step S133, the color copying machine (12) reads the image of the document set on the document table in accordance with the scan command received via the I / F cable (14).

  In step S134, the color copying machine 12 sends RGB image data, which is image data of the read original, to the image area separation unit (212) and the image output I / F unit (210).

  In step S135, the color copier 12 performs image area separation on the image sent to the image area separation unit (212), extracts attributes for each pixel, forms an attribute map, and outputs an image output I / F. Part (210).

  In step S136, the image output I / F unit (210) of the color copying machine 12 performs processing such as delay processing and packetization in accordance with the transfer method of the image portion of the I / F (14) to the scan controller 11. The attribute map data is sent to the controller simultaneously with the RGB image data.

  In step S137, the scan controller (11) temporarily receives the image data and attribute map received via the external interface unit (115) via the CPU bus (117) and the image memory (116-1) and attribute map memory. (116-2). Based on the information for each pixel in the attribute map memory (116-2), predetermined image processing such as compression and resolution / gradation conversion is performed on the image data in the image memory (116-1).

  In step 138, the image subjected to the image processing is sent to the host computer via the I / F (13). In step 139, the scanner driver passes the received image data subjected to predetermined image processing to application software.

  FIG. 14 shows a processing procedure for performing dither processing as an example of step S137. First, attention is focused on the first pixel of image data to be processed (S1401). In the scan controller 12, the target pixel can be specified by a pointer variable or the like secured in the RAM. Next, an attribute value corresponding to the pixel of interest is determined in the attribute map (S1402). If the attribute value indicates a character image, the element of the character image threshold matrix at the position corresponding to the target pixel is stored as the target threshold (S1404). On the other hand, if the attribute value indicates that it is not a character image, the element of the threshold matrix for photographic images at the position corresponding to the pixel of interest is stored as the threshold of interest (S1405). The correspondence between the pixel of interest and the elements of the threshold matrix is as described above. Then, the target pixel value is compared with the target threshold value (S1406). If the pixel value is larger, the binarized value corresponding to the target pixel is set to 1 (S1408). Otherwise, the second value corresponding to the target pixel is set. The value after digitization is set to 0 (S1409). Then, the binarized value is stored as a pixel constituting the binarized image data (S1410). It is determined whether binarization has been completed for all pixels (S1411). If not, the next pixel (raster scanning order) is focused (S1412) and the process is repeated from step S1402. By this procedure, binarization can be performed by switching the threshold matrix according to the attribute map, so that a high-quality binary image can be obtained with a low load.

  FIG. 15 shows a processing procedure for performing a simple compression process as another example of step S137. First, attention is focused on the first pixel of the image data to be processed (S1501). Next, an attribute value corresponding to the pixel of interest is determined in the attribute map (S1502). If the attribute value indicates that it is a character image, for example, the last 3 bits are discarded for the pixel of interest and quantized from 8 bits to 5 bits (S1504). Then, the quantized pixel data is stored (S1505). The pixel data is preferably packed for data compression. On the other hand, if the attribute value indicates that it is not a character image, quantization is not performed, and, for example, thinning from 3 pixels to 1 pixel is performed. Therefore, it is determined whether or not the pixel of interest is to be saved (S1506). This thinning can be realized simply by dividing the image data into 3 × 3 pixel regions, storing only one pixel, for example, the upper left pixel for each region, and discarding the other pixels. Therefore, for the coordinate values (for example, line number and pixel number) in the vertical and horizontal directions indicating the position of the pixel of interest, a remainder is obtained by using 3 as the modulus, and the pixel that becomes 0 is determined as the pixel to be stored, and information indicating that is provided. Discard pixels that have a remainder of 1 or 2 that are stored in memory. The pixel data to be saved is saved as pixel data constituting the compressed image data (S1508). In steps S1505 and S1508, compressed data is generated while packing, for example, by storing the compressed data at an address held by a common pointer and advancing the common pointer by the stored data length.

  Then, it is determined whether or not the binarization has been completed for all the pixels (S1509), and if not, paying attention to the next pixel (raster scanning order) (S1510), the process is repeated from step S1502. According to this procedure, compression processing can be performed by switching the compression method according to the attribute map, so that high-quality compressed image data can be obtained with low load.

  In order to decompress the image data compressed in this way, it is necessary to select the decompression method with reference to the attribute map and perform decompression. That is, attention is sequentially paid from the first attribute value in the attribute map, and the position of interest for the compressed data is held by a pointer or the like. If the attribute value indicates a character, for example, 0 is added to the lower 3 bits obtained by extracting 5 bits from the target position of the compressed data, and an 8-bit pixel is generated and stored. On the other hand, if the attribute value does not indicate a character, it is determined whether or not the position of the pixel to be decoded is a position to be stored. Since the pixel position to be decoded is known by sequentially decoding the pixels, if the size of the original image is given, this determination can be performed as in step S1506. If it is a storage target, data for one pixel is called from the target position of the compressed data and stored as decoded pixel data. If it is not to be saved, it is classified. If the line is the same as the pixel to be saved, the pixel data immediately before the decoded pixel data is duplicated and saved as decoded pixel data. If it is not the same line as the pixel to be saved, the pixel data immediately above the decoded pixel data is duplicated and saved as decoded pixel data. If it is not a character, the pointer indicating the compressed image is advanced by one pixel for every three pixels decoded. In this way, decoding can be performed using the attribute map. Note that the quantization and decimation rates in FIG. 15 are merely examples, and other numerical values may be used.

  FIG. 16 shows a procedure for performing two-dimensional encoding such as JPEG as an example of step S137. First, the image data is divided into regions according to the attribute map (S1601). For simplification of processing, it is preferable to divide so that each divided area is a rectangle without being strictly divided. Each area is given a unique identifier such as an order. Next, paying attention to the first area (S1602), referring to the corresponding attribute map (S1603), if it is a character, the area is reversibly compressed (S1604), and if it is not a character, it is irreversibly compressed (S1605). Then, it is recorded in a newly created area map and stored in a memory or the like (S1606). The area map shows the identifier and the position and range in the original image for each area. A code indicating a compression method may be added thereto. Then, the compressed data is stored so that it can be identified which area it is (S1607). This is repeated for all areas (S1608, S1609). FIG. 17 shows an example of compressed data. The compressed data includes an attribute map 1701, an area map 1702, and compressed data 1703 in which a size, an area identifier, and compressed data are grouped for each area. By referring to these pieces of information, it is possible to identify which region is located at which position in the original image and by which method. As described above, in the image processing system according to the present embodiment, processing such as compression can be performed efficiently by performing image processing by switching binarization processing and compression processing parameters using an attribute map.

  As described above, according to this embodiment, the application software on the host computer can receive a document image after predetermined image processing has been performed. For example, even if the I / F (13) between the controller (11) and the host computer (10) is not fast enough to send 8-bit color multi-value data such as serial communication, predetermined image processing is performed. Sufficient performance can be obtained by using binarization or an efficient compression method.

[Modification]
In this embodiment, the pull scan from the host computer that performs the document reading operation in accordance with the command from the host computer has been described. However, the user uses the user I / F provided in the controller (11). It is also possible to adopt a method in which image data is stored in the HDD (14) built in the controller (11). In this case, the user can freely select an image stored in the built-in HD (14) of the controller (11) by using another type of driver / application from the host computer, and the desired image can be selected from the I / F (13). You may make it take it through. When the I / F (14) is Ethernet (registered trademark), from a user I / F of the controller, for example, from a so-called controller that sends the image file to the host computer (10) in the form of attaching an image file to an e-mail. Of course, it is possible to realize the form of Push Scan.

  Alternatively, the user I / F provided in the image reading device (12) is used to read a document image locally on the image reading device side, and the read image is subjected to the controller (11) via the I / F (14). It is of course possible to apply the present invention when a transmission request is made and a Push Scan is performed from the image reading apparatus (12) side.

  In this embodiment, a color digital copying machine has been used as an example of the image reading device (12). However, the image processing unit as shown in FIG. It goes without saying that the present invention can be applied by providing. Further, it goes without saying that the present invention can be applied to network scanners that are becoming popular in recent years, and it is of course possible to adopt a configuration in which the controller (11) and the image reading device (12) are integrated.

  As described above, according to the present invention, at the time of reading a document image, image processing suitable for each attribute is performed on each area of the read image without increasing the load on the scan controller or the host computer. It is possible to provide an image reading apparatus and method that can be performed, and an image reading system.

  Note that the present invention can be applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, and a printer), and a device (for example, a copying machine and a facsimile device) including a single device. You may apply to. Another object of the present invention is to supply a storage medium (or recording medium) in which a program code of software that realizes the functions of the above-described embodiments is recorded to a system or apparatus, and the computer (or CPU or CPU) of the system or apparatus. (MPU) can also be achieved by reading and executing the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code itself and the storage medium storing the program code constitute the present invention.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an operating system (OS) running on the computer based on the instruction of the program code. A case where part or all of the actual processing is performed and the functions of the above-described embodiments are realized by the processing is also included. Furthermore, after the program code read from the storage medium is written into a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer, the function is determined based on the instruction of the program code. This includes a case where the CPU or the like provided in the expansion card or the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

It is a block diagram of the image processing system of an embodiment. 1 is a block diagram of a color copying machine. It is a figure which shows an example of an attribute map. It is a figure which shows another example of an attribute map. It is a figure which shows another example of an attribute map. It is a figure which shows an example of image data. It is a figure which shows an example of the attribute map corresponding to the image data of FIG. It is a figure which shows another example of the attribute map corresponding to the image data of FIG. It is a figure which shows an example of the dither matrix for character areas. It is a figure which shows an example of the dither matrix for photograph areas. It is a block diagram of a conventional printer. It is a block diagram of the image processing system of an embodiment. It is a flowchart which shows the procedure of the pull scan in the image processing system of embodiment. It is a flowchart of an example of the image processing in step S137 of FIG. It is a flowchart of another example of the image processing in step S137 of FIG. It is a flowchart of another example of the image processing in step S137 of FIG. It is a figure which shows an example of the compression data produced | generated by the procedure of FIG.

Claims (9)

An original reading unit that reads an original image and generates image data composed of pixels, and an attribute map generation that generates an attribute map indicating an image attribute for each pixel or block of the image data based on the image data And an image input / output device connected to the image input / output device, the image data including the image data and an attribute map corresponding to the image data. An image processing apparatus for processing data based on the attribute map, wherein image compression processing corresponding to an attribute indicated by the attribute map corresponding to a target pixel or block is performed on the target pixel or block Image processing means to be applied;
An image processing apparatus comprising: a storage unit that stores compressed image data of a pixel of interest or a block of interest compressed by the image processing unit.
  In the case where the attribute indicated by the attribute map corresponding to the target pixel is a value indicating that it is a character image, the image processing means is smaller than the case where the attribute is not a value indicating that it is a character image. The image processing apparatus according to claim 1, wherein a quantization process using a size threshold matrix is performed on the target pixel.   The image processing means performs a quantization process on the target pixel when the attribute indicated by the attribute map corresponding to the target pixel is a character image, and is a character image. 2. The image processing apparatus according to claim 1, wherein when the value is not a value indicating, pixel thinning processing is performed every predetermined pixel.   The image processing means performs a reversible compression process on the target block when the attribute indicated by the attribute map corresponding to the target block is a character image, and is a character image. The image processing apparatus according to claim 1, wherein when the value is not a value indicating irreversible compression, irreversible compression is performed. An original reading means for reading an original image and generating image data composed of pixels;
An attribute map generating means for generating an attribute map indicating an attribute of an image for each pixel or block of the image data based on the image data;
An image input / output apparatus comprising: output means for outputting the image data and an attribute map corresponding to the image data. An original reading unit that reads an original image and generates image data composed of pixels, and an attribute map generation that generates an attribute map indicating an image attribute for each pixel or block of the image data based on the image data Image data composed of pixels input from an image input / output device, comprising: means; and output means for outputting the image data and an attribute map corresponding to the image data, based on the attribute map An image processing method for performing image compression processing on the pixel of interest or the block of interest according to an attribute indicated by the attribute map corresponding to the pixel of interest or block of interest;
An image processing method comprising: a storage step of storing compressed image data of a pixel of interest or a block of interest compressed by the image processing step.
A document reading step of reading a document image and generating image data composed of pixels;
An attribute map generating step for generating an attribute map indicating an attribute of an image for each pixel or block of the image data based on the image data;
An image input / output method comprising: an output step of outputting the image data and an attribute map corresponding to the image data.
An original reading unit that reads an original image and generates image data composed of pixels, and an attribute map generation that generates an attribute map indicating an image attribute for each pixel or block of the image data based on the image data Image data composed of pixels input from an image input / output device, comprising: means; and output means for outputting the image data and an attribute map corresponding to the image data, based on the attribute map An image processing method to be executed by a computer, wherein image compression processing corresponding to an attribute indicated by the attribute map corresponding to a target pixel or block is performed on the target pixel or block Image processing steps to be applied;
A program for causing a computer to execute a storage step of storing compressed image data of a pixel of interest or a block of interest compressed by the image processing step. A document reading step of reading a document image and generating image data composed of pixels;
An attribute map generating step for generating an attribute map indicating an attribute of an image for each pixel or block of the image data based on the image data;
A program causing a computer to execute the image data and an output step of outputting an attribute map corresponding to the image data.

Images