JP2008017032A - Image processor and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor and an image processing method that comply with a variety of user's demands. <P>SOLUTION: When a recognizing function is for top-bottom discrimination, a second image data processor 104 uses an input image 201 to perform recognition processing 203. Image data are suitably processed according to incidental information on a document mode, magnification rate, etc. Correction processing 206 is performed according to "some condition" between a recognition result 204 obtained by the recognition processing 203 and information included in the incidental information. When the recognizing function is for skew correction, the second image data processor 104 uses the input image 201 to perform the recognition processing 203. The skew angle of the image data is output based upon an execution result. Further, contents of the recognition processing and correction processing 206 are switched according to the condition included in the incidental information 202. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and an image processing method used in an application such as a copy / fax / scanner / printer / HDD storage application, whether priority is given to performance such as recognition accuracy, processing speed, or function. The present invention relates to an image processing apparatus and an image processing method that can select the above conditions according to the user's own environment, application, and preferences.

  Recent MFP machines support various applications such as copy, scanner, FAX, and HDD storage applications, but the MFP also handles so-called image recognition functions such as “recognize and correct images” for top and bottom identification and skew correction. It is desired to support various applications. Under such circumstances, various processing methods have been proposed so far in order to improve recognition accuracy, increase processing speed, and the like in image recognition functions such as top and bottom identification and skew correction.

  For example, in Patent Document 1, as a binarization process used in character recognition (OCR), a defect (decrease in recognition accuracy) in a so-called simple binarization method using a single threshold value for the entire image is improved. A method has been proposed.

  Patent Document 2 proposes a method for determining the top-to-bottom direction using character recognition (OCR).

Further, in Patent Document 3, the arrangement information of the in-line rectangle of the character line is quantized and a symbol is added to create a symbol series of the in-line rectangle, and an engram model of the character line is generated from the symbol series, and the character string is generated from the reference document image. A method has been proposed in which a character direction is identified by comparing with an engram model created for each direction and rotation direction.
JP 2003-058879 A Japanese Utility Model Publication No. 5-012960 JP 2006-031546 A

  However, in Patent Document 1, the binarization processing used in character recognition (OCR) inevitably increases the amount of processing computation compared with the simple binarization method, and the processing speed is reduced.

  Further, Patent Document 2 uses a character recognition (OCR) method to determine the top / bottom direction. While high recognition accuracy can be achieved, the dictionary data is large and has a drawback that it takes a lot of time to collate with the dictionary. , Resulting in reduced processing speed.

  On the other hand, Patent Document 3 is a new method for quickly determining the top-and-bottom direction with a small amount of information storage and a small amount of calculation, but it is disadvantageous in terms of recognition accuracy.

  As described above, some of the conventional technologies represented by the above individually propose improvement of processing speed, improvement of recognition accuracy, etc., but do you want to give priority to processing speed or recognition accuracy? No system provides a system-like mechanism that allows the user to select the conditions such as, etc. according to the environment, application, and preferences.

  Therefore, the present invention provides an image processing apparatus and an image processing method including selection means for selecting any condition that affects performance or function, and processing means for switching processing according to the selected condition.

  In order to solve the above problems, in the invention described in claim 1, an image forming apparatus that obtains image data obtained by electronically reading a document, an image writing apparatus that prints image data on transfer paper, the image data, A memory device for accumulating auxiliary information of the image data; an external interface device for transmitting and receiving the image data; a first image data processing device for processing the image data from the image reading device; In the image processing apparatus having a second image data processing apparatus that processes the image data from the reading apparatus and a bus control apparatus that connects the respective apparatuses, the second image data processing apparatus is configured to perform the image processing. A selection unit that selects a condition that affects either the performance or the function, and a switching unit that switches processing according to the selected condition. Characterized in that it is a that image processing apparatus.

  According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, the second image data processing apparatus includes a top / bottom identification unit that identifies the image data and a skew correction processing unit that performs a skew correction process. It is characterized by having.

  According to a third aspect of the present invention, in the image processing apparatus according to the second aspect, the conditions selected by the selection means are two, which prioritize recognition accuracy of the image processing and prioritize processing speed. It is characterized by that.

  According to a fourth aspect of the present invention, in the image processing apparatus according to the third aspect, the switching means has a binarization process in the recognition process as a process to be switched.

  According to a fifth aspect of the present invention, in the image processing apparatus according to the third aspect, the switching unit includes a pixel density conversion process in a recognition process as a process to be switched.

  According to a sixth aspect of the present invention, in the image processing apparatus according to the third aspect, the switching means is a part for determining the direction from the feature of the image data in the main process in the recognition process as a process to be switched. And a portion for detecting an inclination angle.

  According to a seventh aspect of the present invention, in the image processing apparatus according to the third aspect, the processing target switched by the switching unit is when the electronic data is distributed to the client PC and the file output format is the PDF format. In addition, the correction processing is to actually perform rotation processing, or to perform rotation by the viewer on the client PC side by describing the rotation information in the header of the PDF file without performing the rotation processing.

  According to an eighth aspect of the present invention, in the image processing apparatus according to the third aspect, the processing target to be switched by the switching unit is a work memory amount used during the rotation process of the correction process.

  According to a ninth aspect of the present invention, there is provided a step of obtaining an image data obtained by reading an original document by an image forming unit, a step of printing image data on a transfer paper by an image writing unit, and a step of printing the image data by a memory unit. A step of accumulating data and incidental information of the image data, a step of transmitting and receiving the image data to and from an external device by an external interface means, and a step of sending the image data from the image reading device by a first image data processing means. An image processing method comprising: a step of processing; a step of processing the image data from the image reading device by a second image data processing unit; and a step of connecting the devices by a bus control unit. The second image data processing means selects a condition that affects either the performance or function of the image processing, and performs processing according to the selected condition. Characterized in that an image processing method and a step of switching a.

  According to a tenth aspect of the present invention, in the image processing method according to the ninth aspect, the second image data processing means includes a step of identifying the image data upside down and a step of performing skew correction processing. Features.

  According to an eleventh aspect of the present invention, in the image processing method according to the tenth aspect, the condition to be selected by the selection step is two priority is given to the recognition speed of the image processing, and priority is given to the processing speed. It is characterized by that.

  According to a twelfth aspect of the present invention, in the image processing method according to the eleventh aspect, the switching step includes binarization processing means in a recognition process as a switching target.

  A thirteenth aspect of the present invention is the image processing method according to the eleventh aspect, wherein the switching step includes pixel density conversion processing means in a recognition process as a switching target.

  According to a fourteenth aspect of the present invention, in the image processing method according to the eleventh aspect, the processing target to be switched in the switching step is a part for determining the direction from the characteristics of the image data in the main processing in the recognition processing. And a portion for detecting an inclination angle.

  According to a fifteenth aspect of the present invention, in the image processing method according to the eleventh aspect, the processing object to be switched by the switching step is when the electronic data is distributed to the client PC and the file output format is the PDF format. In addition, the correction processing is to actually perform rotation processing, or to perform rotation by the viewer on the client PC side by describing the rotation information in the header of the PDF file without performing the rotation processing.

  According to a sixteenth aspect of the present invention, in the image processing method according to the eleventh aspect, the processing target to be switched by the switching step is a work memory amount used during the rotation process of the correction process.

  As described above, according to the present invention, it is possible to select according to the user's own environment, application, preference, etc., and to satisfy various user requirements by enabling the process to be switched according to the selected condition. it can.

  In the present embodiment, it is possible to select whether conditions such as recognition accuracy and processing speed are prioritized or whether functions are prioritized according to the user's own environment, application, and preferences, and processing is performed according to the selected conditions. An image processing apparatus and an image processing method are provided.

  Hereinafter, the image processing apparatus and the image processing method of this embodiment will be described in detail.

  FIG. 1 shows an outline of the MFP system configuration of this embodiment, and the components will be described. The reading device 101 includes a line sensor composed of a CCD photoelectric conversion element, an A / D converter, and a driving circuit thereof, and converts digital image data of 8 bits for each of RGB from grayscale information obtained by scanning a set original. Generate and output.

  The first image data processing apparatus 102 performs a process for unifying the digital image data from the reading apparatus 101 to a predetermined characteristic and outputs the digital image data. In this apparatus, since the characteristics of the reading apparatus 101 are not changed and the output is predetermined and unified image data, predetermined image processing is realized by an ASIC which is a dedicated image processing circuit.

  The bus control device 103 is a data bus control device that exchanges various data such as image data and control commands necessary in the digital image processing device, and also has a bridge function between a plurality of types of bus standards. In the present embodiment, the first image data processing device 102, the second image data processing device 104, and the CPU 106 are a PCI or PCI-Express bus that is a bus connecting between parts in the personal computer, and the HDD 105 is an ATA bus. Connected by ASIC.

  The second image data processing device 104 performs image processing suitable for the output destination designated by the user on the digital image data whose characteristics predetermined by the first image data processing device 102 are unified, and outputs the result. Since the input image data of the second image data processing device 104 is a unified characteristic in the recognition processing by the image processing of this device, the recognition processing is performed before the image processing suitable for this output destination, and this device Correction processing based on the result of recognition processing by the image processing is performed in the subsequent stage. Note that the recognition process and the correction process are performed according to preselected conditions.

  The HDD 105 is a large-sized storage device for storing electronic data that is also used in a desktop personal computer, and mainly stores digital image data and accompanying information of the digital image data in the digital image processing apparatus. In this embodiment, an ATA bus connection hard disk standardized by expanding IDE is used.

  The CPU 106 is a microprocessor that controls the entire control of the digital image processing apparatus. In this embodiment, sequence management for overall control, request management for each device, hardware setting management, timing management, task management for a plurality of user requests, and image processing by software of the second image data processing device 104 Is also executed using the CPU 106.

  The memory 107 is a volatile memory that temporarily stores programs, intermediate processing data, and image data when the CPU 106 executes the contents programmed for control. An object-oriented program stores object data that is objectized in each control unit, and performs control processing according to the responsibility of each object according to a user request.

  The plotter I / F device 108, in accordance with the print request of the CPU 106, digital image data having a color expression method C (cyan) M (magenta) Y (yellow) K (black) sent via the general-purpose standard I / F. Is received, a bus bridge process for outputting to the dedicated I / F of the plotter device 109 is performed. The general-purpose standard I / F used in the present embodiment is a PCI or PCI-Express bus.

When the plotter device 109 receives the digital image data composed of CMYK, the plotter device 109 outputs the received image data on the transfer paper using an electrophotographic process using a laser beam.
S.B.113 is one of chip sets used in personal computers, and is a general-purpose electronic device called a south bridge.

  The ROM 114 is a memory that stores a control program used when the CPU 106 controls the digital image processing apparatus.

  In addition, image processing assembler code for middleware (digital signal processor or DSP) in the second image data processing device 104 and image processing parameters such as filter processing and γ processing, which are image processing, are stored.

  In addition, when image processing is performed by a microprocessor DSP brought to the processing of voice, images, etc., assembler codes and image processing parameters for programming in a lower language corresponding to the machine language stored in the ROM 114 in one-to-one correspondence. Used when downloading and performing image processing with a DSP.

  The operation display device 110 is a part that performs a user interface with the digital image processing device, and is composed of a liquid crystal display device (hereinafter referred to as LCD) and a key switch, and displays various states and operation methods of the device on the LCD. Detects key switch input.

  In the present embodiment, the operation display device 110 is connected to the CPU 106 via a PCI or PCI-Express bus. It supports copy, scanner, HDD storage, and FAX transmission.

  The line I / F device 111 is a device that connects a PCI or PCI-Express bus to a telephone line. This apparatus enables the digital image processing apparatus to exchange various data via the telephone line. A FAX is a normal facsimile, and receives image data from the digital image processing apparatus via a telephone line.

  The external I / F device 112 is a device that connects a PCI or PCI-Express bus to an external device, and this device enables the digital image processing apparatus to exchange various data with the external device. In this embodiment, an IP address is assigned to the device side, and the connection I / F is connected by TCP / IP using an Ethernet (registered trademark) work.

  The client PC is a so-called personal computer, and through the application software and driver installed in the client PC 116, in this embodiment, the user scans and prints with the application software for connecting the image input device called a Twin driver to the personal computer. Do out.

  FIG. 2 shows a detailed configuration of the second image data processing apparatus 104. First, as a component, normal image processing 205 that performs processing suitable for an output destination designated by the user, recognition processing 203 and recognition result 204 required in the present embodiment, and correction processing 206 based on the result of recognition processing There is.

  For example, in the case of a scanner distribution operation, an image digitized by the reading device 101 is input to the system. Based on the data, the image data device 102 converts the image into a unified image with a certain characteristic (color space or the like). The input image 201 of the second image data processing device 104 is digital image data unified by the image data device 102.

  The other incidental information 202 becomes incidental information of the input image 201. The supplementary information includes input information (manuscript mode, magnification, image size, etc.) on the operation display device 110, separation data (character / photo / halftone dot, etc.) corresponding to each pixel of the image data, and this implementation. Some conditions that affect the performance and functionality required by the form are also included. The operation display device 110 is the most common selection means for any condition, and is selected in advance before the scanner distribution operation.

  The selection means may be remote setting via a network or the like. In the second image data processing device 104, first, recognition processing 203 is performed using the input image 201. At that time, according to “some condition” in the information included in the incidental information 201, the processing suitable for the “processing of condition 1”, “processing of condition 2”,... Switch to, and execute the process.

  Next, the image data is subjected to suitable image processing in accordance with incidental information 202 such as a document mode and a variable magnification. After that, “some condition” of the recognition result 204 obtained by the recognition processing 203 and the information included in the incidental information 202 for image data that has been subjected to suitable image processing 205 (eg, filter processing or gamma processing). Accordingly, the correction process 206 is performed.

  The operation for switching to a suitable process is the same as the recognition process 203 described above. Then, the image data subjected to the correction process 206 is output as the output image 207 of the second image data processing apparatus 104. Thereafter, the image data is distributed to the client PC via the external I / F device 112.

  In this embodiment, the correction process 206 is arranged after the image process 205 suitable for the output destination. However, the correction process 206 may be arranged in reverse. What is important is that the contents of the recognition process 203 and the correction process 206 are switched according to some condition included in the incidental information 202.

  As mentioned above, whether to prioritize performance such as recognition accuracy and processing speed, or prioritize functions when using image recognition processing in applications such as copying, faxing, scanners, printers, and HDD storage apps By making it possible to select conditions according to the user's own environment, application, preferences, and the like, and enabling switching of processing according to the selected conditions, it is possible to satisfy various user requirements.

  Next, FIG. 3A is a diagram showing a detailed configuration of the recognition processing 203 of the present embodiment. This is a case where top-to-bottom identification or skew correction is applied as the recognition function of the second image data processing device 104. The component includes a pre-process 301 and a main process 304.

  The pre-process 301 is a process as a previous step with respect to the main process as a recognition function, and includes a binarization process 302 and a pixel density conversion process 303. The binarization process 302 is a process for making a binary image suitable for the input of this process, and the pixel density conversion process 303 is a process of converting the pixel density (image size / resolution) input to this process.

  The main processing 304 is a main processing as a recognition function. In the case of top-and-bottom identification, the processing 304 refers to processing for discriminating the east, west, north, and south directions from the characteristics of the image data. In the case of skew correction, the skew angle ( (Tilt angle) is detected.

  FIG. 3B is a diagram showing the configuration of the correction process 206 of the present embodiment. Since the correction process 206 is not subdivided like the recognition process 203, the correction process 206 has the same configuration as that in FIG. The correction process 206 refers to a process of correcting (rotating) in the north direction from the result of the recognition process 203 in the case of top-and-bottom identification, and refers to a process of correcting (rotating) in a non-inclined direction in the case of skew correction.

  Hereinafter, the operation of the correction process of this embodiment will be described in detail.

  First, when the recognition function is top-and-bottom identification, the second image data processing apparatus 104 first performs recognition processing 203 using the input image 201. At that time, pre-processing 301 (binarization processing 302 and pixel density conversion processing 303) which is a component of the recognition processing 203 and main processing 304 (image data of the image data) according to [some conditions] of the information included in the incidental information 202. The process is switched from the process of determining the direction of east, west, north, and south to a suitable process, and each process is executed.

  According to this execution result, the east, west, north, and south directions of the image data are output. Next, the image data is subjected to suitable image processing in accordance with incidental information 202 such as a document mode and a variable magnification. After that, with respect to the image data that has been subjected to suitable image processing 205 (for example, filter processing or gamma processing), the recognition result 204 (direction of east, west, north, and south) obtained by the recognition processing 203 and the information included in the incidental information 202 are displayed. A correction process 206 (a process of rotating in the north direction) is performed in accordance with [some conditions].

  The operation for switching to a suitable process is the same as the recognition process 203 described above. Then, the image data that has been subjected to the correction process 206 is output as the output image 207 of the second image data processing apparatus 104.

  Thereafter, the image data is distributed to the client PC via the external I / F device 112. In the present embodiment, the correction process 206 is arranged after the image process 205 suitable for the output destination. However, the correction process 206 may be arranged in reverse. What is important is that the contents of the recognition process 203 and the correction process 206 are switched according to some condition included in the incidental information 202.

  Next, when the recognition function is skew correction, the second image data processing apparatus 104 first performs recognition processing 203 using the input image 201. At that time, pre-processing 301 (binarization processing 302 and resolution conversion processing 303), which are components of the recognition processing 203, and main processing 304 (characteristics of image data) according to [some conditions] of the information included in the supplementary information 202 The processing is switched from the processing for detecting the skew angle to the appropriate processing, and each processing is executed. Based on the execution result, the skew angle (tilt angle) of the image data is output.

  Next, the image data is subjected to suitable image processing in accordance with incidental information 202 such as a document mode and a variable magnification. After that, with respect to image data that has been subjected to suitable image processing 205 (eg, filter processing or gamma processing), the recognition result 204 (skew angle) obtained by the recognition processing 203 and information included in the accompanying information 202 A correction process 206 (a process of rotating in a non-tilt direction) is performed according to [some conditions]. The operation for switching to a suitable process is the same as the recognition process 203 described above.

  Then, the image data subjected to the correction process 206 is output as the output image 207 of the second image data processing apparatus 104. Thereafter, the image data is distributed to the client PC via the external I / F device 112. In the present embodiment, the correction process 206 is arranged after the image process 205 suitable for the output destination. However, the correction process 206 may be arranged in reverse. What is important is that the contents of the recognition process 203 and the correction process 206 are switched according to some condition included in the incidental information 202.

  In addition, when using image recognition processing in applications such as copy, fax, scanner, printer, HDD storage app as described above, whether priority is given to performance such as recognition accuracy, processing speed, etc. Can be selected according to the user's own environment, application, and preferences, and processing can be switched according to the selected condition, thereby satisfying various user requirements.

  Hereinafter, a case in which “some condition” and “switching process” are embodied will be described.

  First, some conditions to be selected are “prioritize recognition accuracy” and “prioritize processing speed”, and the target of the switching process is the binarization process 302 in FIG. FIG. 4 shows a simple binarization process 401 and an adaptive binarization process 402 as an embodiment of the binarization process 302 to be switched.

  The simple binarization process 401 is a simple method using a single threshold value for the entire image. Since this process performs the same process on the entire image surface, the recognition accuracy (image quality) in the present process 304 tends to be a factor that decreases. However, the amount of calculation is small, which is advantageous in terms of processing speed. Therefore, this is an example of processing applied when “prioritize processing speed” is selected.

  The adaptive binarization process 402 is a general term for a method of binarization suitable by changing the process according to the image content. This process makes it possible to obtain a high-quality binary image for improving the recognition accuracy in the present process 304, but it tends to increase the amount of calculation and is disadvantageous in terms of processing speed. Therefore, this is an example of processing applied when “prioritize recognition accuracy” is selected. Regarding the operation, it is possible to apply a condition and a switching process target.

  Further, when the recognition function is the top / bottom identification process or the skew correction process as described above, the conditions “prioritize recognition accuracy” and “prioritize the processing speed” can be selected, and the recognition process is performed according to the selected condition. By making it possible to switch the binarization process, it is possible to satisfy various user requirements.

  Next, the selection condition is “prioritize recognition accuracy” or “prioritize processing speed”, and the target of the switching process is the pixel density conversion process 303 in FIG. As an embodiment of the pixel density conversion process 303 to be switched to FIG. 5, a pixel density 1 / 2-fold conversion process 501, a pixel density 1 / 4-fold conversion process 502, and a pixel density 1-fold conversion process (process that does not convert) 503 are performed. Show.

  The pixel density ½ conversion process 501 is a process of halving the pixel density, that is, a process of converting two pixels at a ratio of one pixel, and simple thinning processing, OR thinning processing, and the like are common. As the image density is reduced (the number of pixels is reduced), the amount of data to be processed is simply reduced, so the processing time in this processing 304 is shortened. However, the image quality is deteriorated and the recognition accuracy is lowered accordingly. It tends to be a factor. Therefore, this is an example of processing when “prioritize processing speed” is selected.

  Since the pixel density ¼ conversion processing 503 is an example in which the density conversion magnification is further increased, it is advantageous in terms of processing speed compared to ½ times. The pixel density 1-time conversion process 504 is a process for transferring the input image as it is to the main process 304 without performing density conversion. This can be said to be the process that most reduces the recognition accuracy for the present process 304, and is an example of the process when “prioritize recognition accuracy” is selected.

  Considering this item comprehensively from the contents described so far, whether density conversion is performed or not is performed according to the pixel density of the image input to the second image data processing device 104, and which one Providing a selection means that adjusts whether density conversion should be performed at a certain magnification would be more effective. Regarding the operation, a condition and a switching process target can be applied.

  Further, when the recognition function is the top / bottom identification process or the skew correction process as described above, the conditions “prioritize recognition accuracy” and “prioritize the processing speed” can be selected, and the recognition process is performed according to the selected condition. By making it possible to switch the pixel density conversion process, various user requirements can be satisfied.

  Next, an embodiment of the present processing 304 that is a processing target to be switched to FIG. 6 when some condition to be selected is “prioritize recognition accuracy” and “prioritize processing speed” and the target of the switching processing is the case of the present processing 304. The OCR top and bottom identification processing 601 and the high-speed and lightweight top and bottom identification processing 602 when the recognition function is top and bottom identification are shown.

  The OCR top / bottom identification processing 601 is a method of determining the top / bottom direction using character recognition (OCR), which is a known technique. In the case of this method, although stable recognition accuracy can be obtained, the dictionary data is large and has a drawback that it takes a lot of time to collate with the dictionary. Therefore, this is an example of processing when “prioritize recognition accuracy” is selected.

  The high-speed lightweight top-and-bottom identification process 602 is a new method for quickly determining the top-and-bottom direction with a small amount of information storage and a small amount of calculation. In this case, since the aspect of recognition accuracy is sacrificed to some extent, this is an example of processing when “prioritize processing speed” is selected.

  This time, the top and bottom identification is given as an example of the recognition function, but the same idea can be applied to the case of skew correction. Regarding the operation, it is possible to apply a condition and a switching process target.

  Further, when the recognition function is the top / bottom identification process as described above, the conditions “prioritize the recognition accuracy” and “prioritize the processing speed” can be selected, and the main process (image in the recognition process) is selected according to the selected condition. It is possible to satisfy various user requirements by enabling switching of the direction determining part).

  Next, in the case of the correction process 206 based on the result of the recognition process, the recognition process which is the process target to be switched to FIG. As an embodiment of the correction process 206 based on the result of the above, a rotation process 701 and a correction process 702 using a PDF header are shown.

  The rotation process 701 is a process of rotating the actual image data in the north direction according to the recognition result 204 (the direction of east, west, north, and south in the case of top and bottom identification). When it is realized by hardware rotation, it tends to be disadvantageous in terms of cost, and when it is realized by software rotation, it tends to be disadvantageous in terms of processing speed.

  However, in the case of the scanner distribution function, the final file format can be distributed as an arbitrary output file format regardless of the type of file format, which is functional in comparison with a correction process 702 using a PDF header described later. It is advantageous. Therefore, this is an example of processing that is applied when [prioritize function] is selected.

  The correction processing 702 using the PDF header is based on the recognition result 204 (or the direction of east, west, north, and south in the case of top and bottom identification), and does not actually rotate the image data in the north direction, but includes rotation information for correction in the PDF header. This process uses a function specific to the PDF file format that is described and displayed by the viewer on the client PC side according to the rotation information.

  This processing is advantageous in terms of processing speed because it does not perform actual rotation processing, but is limited in function because it cannot be output in other than PDF file formats such as TIFF and JPEG.

  Therefore, this is an example of processing applied when “prioritize processing speed” is selected. In this example, top-and-bottom identification is given as an example of the recognition function, but the same idea can be applied to skew correction.

  In this example, top-and-bottom identification is given as an example of the recognition function. However, even in the case of skew correction, only the content to be corrected is replaced by “rotating in a non-tilt direction”, and the same concept can be applied. Regarding the operation, a condition and a switching process target can be applied.

  In addition, when the recognition function is the top / bottom identification process or the skew correction process as described above, the conditions “prioritize the processing speed” and “prioritize the function” can be selected, and the rotation process is performed as the correction process according to the selected condition. By making it possible to switch between using the PDF function without performing the rotation process, it becomes possible to satisfy various user requirements.

  Further, in the case where some conditions to be selected are “prioritize the processing speed of the recognition function” and “prioritize the processing speed other than the recognition function” and the target of the switching process is the correction process 206 based on the result of the recognition process, FIG. As an embodiment of the correction process 206 based on the result of the recognition process to be switched, a general rotation process (work memory amount = one image) 801 and a work memory rotation process (work memory amount <less than one image) 802 is shown.

  The general rotation processing (work memory amount = one image) 801 is the work memory amount when rotating the actual image data in the north direction according to the recognition result 204 (or the direction of east, west, south, and north in the case of top and bottom identification). Is a process when one image is used.

  This processing is advantageous in terms of processing speed because one surface can be rotated at a time, but the amount of work memory used increases, and in the case where limited memory is shared by various functions, other functions Except for the recognition function by squeezing resources, the processing speed tends to decrease. Therefore, this is an example of processing applied when “prioritize processing speed of recognition function” is selected.

  The work memory rotation processing (work memory amount <less than one image) 802 is the amount of work memory when rotating the actual image data in the north direction according to the recognition result 204 (or the direction of east, west, north, south) This is a process in the case of realizing with a small memory without using one image. In the case of this processing, it is disadvantageous in terms of processing speed because it cannot be rotated at once but needs to be divided and sequentially rotated.

  However, since the amount of work memory used can be saved, in the case where limited memory is shared by various functions, it is possible to secure many resources that other functions can use, and as a result, the processing speed of other functions can be improved. Connected. Therefore, this is an example of processing applied when “prioritize processing speed other than the recognition function” is selected.

  In this example, top-and-bottom identification has been taken as an example of the recognition function. However, even in the case of skew correction, the content to be corrected is simply replaced by “rotating in a non-tilt direction”, and the same concept can be applied. Regarding the operation, it is possible to apply a condition and a switching process target.

  As described above, in the case of the top-and-bottom identification processing or the skew correction processing, the recognition function can select the conditions “prioritize the processing speed of the recognition function” and “prioritize the processing speed other than the recognition function”. Therefore, by making it possible to switch the amount of work memory used during the rotation process as the correction process, it is possible to satisfy various user requirements.

1 shows an outline of a configuration of an MFP system according to the present embodiment. 2 shows a detailed configuration of a second image data processing apparatus 104 according to the present embodiment. (A) shows the detailed structure of the recognition process 203 of this embodiment. (B) shows the configuration of the correction processing 206 of the present embodiment. A simple binarization process 401 and an adaptive binarization process 402 are shown as embodiments of the binarization process 302 to be switched in this embodiment. A pixel density conversion process 303 that is a process target to be switched in the present embodiment is shown. The present processing 304 that is a processing target to be switched in the present embodiment is shown. A rotation process 701 and a correction process 702 using a PDF header are shown as an example of the correction process 206 based on the result of the recognition process to be switched in this embodiment. A general rotation process 801 and a work-saving memory rotation process 802 are shown as an example of the correction process 206 based on the result of the recognition process to be switched in this embodiment.

Explanation of symbols

101 Reading device 102 First image data processing device 102
103 Bus control device 104 Second image data processing device 104
105 HDD
106 CPU
107 Memory 108 Plotter I / F Device 109 Plotter Device 110 Operation Display Device 111 Line I / F Device 112 External I / F Device 113 B.
114 ROM
201 Input Image 202 Auxiliary Information 203 Recognition Processing 204 Recognition Result 205 Image Processing Suitable for Output Destination 206 Correction Processing Based on Recognition Processing Result 207 Output Image 301 Preprocessing 302 Binary Down-Processing 303 Pixel Density Conversion Processing 304 Main Processing 401 Simple 2 Value conversion processing 402 Adaptive binarization processing 501 Pixel density ½ times conversion processing 502 Pixel density ¼ time conversion processing 503 Pixel density 1 time conversion processing 601 OCR top and bottom identification 602 High-speed and lightweight top and bottom identification 701 Rotation processing 702 PDF header Correction processing used 801 General rotation processing 802 Work memory rotation processing

Claims (16)

An image reading apparatus for obtaining image data obtained by reading a document and digitizing it;
An image writing device for printing the image data on transfer paper;
A memory device for storing the image data and incidental information of the image data;
An external interface device that transmits and receives the image data to and from an external device;
A first image data processing device for processing the image data from the image reading device;
A second image data processing device for processing the image data from the memory device;
In an image processing apparatus having a bus control device for connecting the above devices,
The second image data processing device includes a selection unit that selects a condition that affects either one of the performance or function of the image processing;
An image processing apparatus comprising switching means for switching processing according to the selected condition.   The image processing apparatus according to claim 1, wherein the second image data processing apparatus includes a top / bottom identification unit that identifies the image data, and a skew correction processing unit that performs a skew correction process.   The image processing apparatus according to claim 2, wherein the condition selected by the selection unit includes two conditions: priority is given to recognition accuracy of the image processing, and priority is given to processing speed.   The image processing apparatus according to claim 3, wherein the processing unit to be switched is a binarization process in a recognition process.   The image processing apparatus according to claim 3, wherein the switching unit is a pixel density conversion process in a recognition process.   4. The switching unit according to claim 3, wherein the switching target is a part for determining a direction from a feature of image data and a part for detecting an inclination angle in the main process in the recognition process. Image processing device. The processing object to be switched by the switching means is when electronic data is distributed to the client PC and when the file output format is PDF format.
4. The method according to claim 3, wherein the correction processing is actually rotation processing, or rotation is described by a viewer on the client PC side by describing rotation information in a header of the PDF file without performing rotation processing. Image processing device   The image processing apparatus according to claim 3, wherein the processing target to be switched by the switching unit is a work memory amount used during a rotation process of the correction process. A step of obtaining electronic image data by reading a document;
Printing the image data on transfer paper;
Storing the image data and incidental information of the image data;
Processing the image data from the memory device;
The step of processing the image data includes a selection step of selecting a condition that affects either one of the performance or function of the image processing;
A switching step of switching processing according to the selected condition.   The image processing method according to claim 9, wherein the step of processing the image data includes a step of identifying the image data upside down and a step of performing a skew correction process.   11. The image processing method according to claim 10, wherein the conditions selected by the selection step are two: priority is given to recognition accuracy of the image processing, and priority is given to processing speed.   12. The image processing method according to claim 11, wherein in the switching step, a processing target to be switched is a binarization processing unit in a recognition process.   The image processing method according to claim 11, wherein in the switching step, a processing target to be switched is a pixel density conversion process in a recognition process.   The processing target to be switched in the switching step includes a part for determining a direction from a feature of image data and a part for detecting an inclination angle in the main process in the recognition process. Image processing method. The processing target to be switched by the switching step is when the electronic data is distributed to the client PC and the file output format is the PDF format.
12. The method according to claim 11, wherein the correction processing is to actually perform rotation processing, or to perform rotation by a viewer on the client PC side by describing rotation information in a header of the PDF file without performing rotation processing. Image processing method.   The image processing method according to claim 11, wherein the processing target to be switched by the switching step is a work memory amount used during a rotation process of the correction process.

Images