JP2016051914A - Image processing apparatus, image processing system, computer program and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of outputting a desired output data, by automatically setting an optimum output resolution, without requiring complicated settings by a user, and to provide an image processing system, a computer program and a recording medium.SOLUTION: An image processing apparatus includes a character generation unit generating character identification information for identifying character candidates included in an image, based on the pixel values of a plurality of pixels constituting the image, a character string generation unit generating character string identification information for identifying character string candidates constituted of a plurality of characters based on the character identification information thus generated, a resolution determination unit for determining the resolution of an image based on the character string identification information thus generated, and a conversion unit for converting the resolution of an image by using the resolution thus determined.SELECTED DRAWING: Figure 1

Description

  The present invention automatically sets parameters to obtain optimum output data according to the characteristics of the input image and the purpose of use of the output data without performing complicated settings when acquiring scan data, The present invention relates to an image processing apparatus capable of outputting desired output data, an image processing system including the image processing apparatus, a computer program for realizing the image processing apparatus, and a recording medium on which the computer program is recorded.

  In recent years, digital image processing systems have made remarkable progress, and the construction of digital image processing technology is progressing. For example, in the field of a copying machine using an electrophotographic method or an inkjet method, a multi-function printer (MFP), a document original is read by a scanner and stored as a document file which is electronic data. Stored document files are managed. Furthermore, the document file is compressed and transmitted by e-mail.

  In general, since an image read by a scanner (hereinafter referred to as a scanned image) has a large file size, the user manually sets a resolution or a compression rate in accordance with the purpose of storing or transmitting the scanned image. It is necessary. However, in setting the resolution or compression ratio, the optimum values differ depending on the type of document, the size of characters, the resolution of the printed matter, etc., and these values are set correctly by general users and the optimum output data. Is difficult to get.

  Therefore, for example, a histogram is created from the image data, the created histogram is analyzed, and the document type is classified into characters, photos, and characters / photos. For a document that has been determined to be text, the maximum black run length (maximum length (number of pixels) of the connection (run) of black pixels on the scan line of the document) is determined, and the width is divided into three parts before and after the character peak density. When all the histograms are larger than the fine character determination threshold and the maximum black run length in the main scanning direction is smaller than the black run length threshold, it is determined as a fine character document, and in other cases, it is determined as a standard character document. In the case of a standard character document, the resolution is set to the standard mode. In the case of a thin character document, the high-definition mode is set. For the photo document and the character / photo document, the standard mode is set. An image forming apparatus is disclosed (see Patent Document 1).

JP 9-46512 A

  However, in the apparatus described in Patent Document 1, when a large original character such as a title and a small fine character such as a footnote are mixed in a scanned document, if there is a ruled line, the ruled line is a black black character. If the run length is processed and the black run length is greater than or equal to the threshold, even if there are small and fine characters, the standard mode is set, and the scanning resolution is lowered, resulting in small and fine characters. There is a possibility of being crushed.

  The present invention has been made in view of such circumstances, and an image processing apparatus that automatically sets an optimum output resolution and that can output desired output data without complicated settings by the user, and the image processing An object of the present invention is to provide an image processing system including the apparatus, a computer program for realizing the image processing apparatus, and a recording medium on which the computer program is recorded.

  An image processing apparatus according to the present invention generates a character generation unit that generates character specifying information for specifying a character candidate included in an image based on pixel values of a plurality of pixels constituting the image, and the character generation unit A character string generating unit that generates character string specifying information for specifying a character string candidate composed of a plurality of characters based on the character specifying information, and the image based on the character string specifying information generated by the character string generating unit A resolution determining unit that determines a resolution, and a conversion unit that converts the resolution of the image using the resolution determined by the resolution determining unit.

  The image processing apparatus according to the present invention includes a determination unit that determines whether or not the character string generation information can be generated by the character string generation unit, and the resolution determination unit determines whether or not the character string specification information is generated by the determination unit. If it is determined that the resolution of the image is determined based on the character specifying information generated by the character generation unit.

  The image processing apparatus according to the present invention includes a character size determination unit that determines one or a plurality of character sizes based on the generated character string specifying information or character specifying information, and the resolution determining unit is the character size determining unit. The resolution is determined based on the determined character size.

  An image processing apparatus according to the present invention includes a setting unit that sets whether to give priority to image quality or image size, and the character size determination unit, when the setting unit sets priority to image quality, When the minimum character size is determined among one or a plurality of character sizes based on the column specifying information or the character specifying information, and the setting unit is set to prioritize the image size, the character string specifying information or the character specifying information A character frequency having a maximum frequency is determined from one or a plurality of character sizes based on the character size.

  An image processing apparatus according to the present invention includes a compression rate determination unit that determines a compression rate based on the resolution determined by the resolution determination unit, and a compression that compresses the image using the compression rate determined by the compression rate determination unit. And a section.

  In the image processing apparatus according to the present invention, the compression rate determination unit determines a compression rate based on the resolution determined by the resolution determination unit and the character size determined by the character size determination unit. And

  An image processing system according to the present invention includes an image processing device according to any one of the above-described inventions, an image input device that inputs an image to the image processing device, and an image that outputs an image converted by the image processing device. And an output device.

  A computer program according to the present invention is a computer program for converting the resolution of an image. The computer program stores character specifying information for specifying character candidates included in the image based on pixel values of a plurality of pixels constituting the image. Character generation means for generating, character string generation means for generating character string specifying information for specifying a character string candidate composed of a plurality of characters based on the generated character specifying information, and based on the generated character string specifying information It functions as a resolution determining means for determining the resolution of the image, and a converting means for converting the resolution of the image using the determined resolution.

  A recording medium according to the present invention records a computer program according to the above-described invention.

  According to the present invention, the optimum output resolution can be automatically set, and desired output data can be output without complicated settings by the user.

1 is a block diagram illustrating an example of a configuration of an image forming apparatus according to a first embodiment. 1 is a block diagram illustrating an example of a configuration of an image forming apparatus according to a first embodiment. It is a block diagram which shows an example of a structure of the process parameter determination part of 1st Embodiment. It is a flowchart which shows an example of the process sequence of an edge detection part. It is a schematic diagram which shows an example of the edge detection process by an edge detection part. It is a flowchart which shows an example of the process sequence of a character rectangle extraction part. It is a schematic diagram which shows an example of the character rectangle extraction process by a character rectangle extraction part. It is a flowchart which shows an example of the process sequence of a character string extraction part. It is a schematic diagram which shows an example of the character string extraction process by a character string extraction part. It is a flowchart which shows an example of the process sequence of a resolution determination part. It is a schematic diagram which shows an example of the resolution determination process by the resolution determination part. It is explanatory drawing which shows an example of the relationship between the rectangular size of a character string or a character, and output resolution. It is a block diagram which shows an example of a structure of the image forming apparatus of 2nd Embodiment. It is a block diagram which shows an example of a structure of the process parameter determination part of 2nd Embodiment. It is explanatory drawing which shows an example of the relationship between the rectangular size of a character string or a character, output resolution, and a compression rate. It is a block diagram which shows an example of a structure of the image forming apparatus of 3rd Embodiment. It is a schematic diagram which shows an example of the setting method of the resolution in 1st Embodiment thru | or 3rd Embodiment. It is a schematic diagram which shows the example of a display of default setting.

(First embodiment)
Hereinafter, the present invention will be described with reference to the drawings illustrating embodiments. 1 and 2 are block diagrams illustrating an example of the configuration of the image forming apparatus 100 according to the first embodiment. FIG. 1 shows signal processing when the read color image data is output from the color image output device 2, and FIG. 2 shows signal processing when the read color image data is transmitted from the transmission device 3. In the following description, the image forming apparatus 100 as an image processing system will be described.

  The image forming apparatus 100 is a digital multifunction machine having a color copy function, a color scanner function, and the like. The image forming apparatus 100 includes a color image input device 1 as an image input device that optically reads a color image from a document.

  Connected to the color image input device 1 is a color image processing device 50 as an image processing device that generates image data of a read color image and a compressed file.

  The color image processing device 50 includes a color image output device 2 as an image output device that outputs a color image based on the image data generated by the color image processing device 50, and a compressed file generated by the color image processing device 50 as an external device. Is connected to a transmission device 3 as an image output device.

  That is, the image forming apparatus 100 includes a color image input device 1, a color image processing device 50, a color image output device 2, a transmission device 3, and the like. An operation panel 4 is connected to the color image input device 1, the color image processing device 50, the color image output device 2, and the transmission device 3.

  The operation panel 4 includes a setting button for the user to set the operation mode of the image forming apparatus 100, an operation unit such as a numeric keypad, a display unit including a liquid crystal display and the like (all not shown). Various processes executed by the image forming apparatus 100 are controlled by a CPU (Central Processing Unit) (not shown). The CPU of the image forming apparatus 100 performs data communication with a computer and other digital multi-function peripherals connected to the network via a network card and a LAN cable (not shown).

  Hereinafter, each part of the image forming apparatus 100 will be described in detail.

  The color image input device 1 is composed of, for example, a color scanner having a CCD (Charge Coupled Device). The color image input device 1 reads a reflected light image from a document as an analog signal of RGB (R: red, G: green, B: blue) using a CCD and outputs it to the color image processing device 50.

  The color image processing device 50 performs an A / D conversion unit 11, a shading correction unit 12, a document type determination unit 13, an input tone correction unit 14, and an RGB analog signal input from the color image input device 1. Image data composed of RGB digital signals (hereinafter referred to as RGB signals) is generated by executing image processing to be described later in the region separation processing unit 15.

  In addition, the color image processing device 50 applies a color correction unit 16, a black generation and under color removal unit 18, a spatial filter processing unit 19, an output tone correction unit 21, and a floor to the RGB signals output from the region separation processing unit 15. The tone reproduction processing unit 22 executes image processing to be described later, thereby generating image data composed of digital signals of CMYK (C: cyan, M: magenta, Y: yellow, K: black) as a stream. Output to the color image output device 2. The image data may be temporarily stored in a storage unit (not shown) before being output to the color image output device 2. The storage unit is a non-volatile storage device (for example, a hard disk).

  The color image output device 2 forms a color image on a recording sheet (for example, recording paper) based on the image data input from the color image processing device 50 by a method such as thermal transfer, electrophotography, or inkjet. Output.

  The image forming apparatus 100 may include a monochrome image output apparatus that forms and outputs a monochrome image on a recording sheet instead of the color image output apparatus 2. In this case, the color image processing device 50 converts the image data of the color image into the image data of the monochrome image, and then outputs it to the monochrome image output device.

  The color image processing apparatus 50 also includes a color correction unit 16, a spatial filter processing unit 19, a resolution conversion processing unit 20, an output tone correction unit 21, and a compression processing unit for the RGB signals output from the region separation processing unit 15. By executing image processing to be described later at 23, compressed image data generated by compressing output image data composed of RGB digital signals, grayscale or monochrome output image data is generated and output to the transmission device 3.

  In the present embodiment, when the resolution conversion processing unit 20 performs the resolution conversion process, the processing parameter determination unit 17 determines the output resolution (resolution) to be converted as necessary, and determines the determined resolution. By outputting to the resolution conversion processing unit 20, it is possible to convert the input image (image data) to a desired output resolution.

  The output image data or the compressed data obtained by compressing the output image data may be temporarily stored in the storage unit before being output to the transmission device.

  As shown in FIG. 1, when the color image data read by the color image input device 1 is output from the color image output device 2, the resolution conversion processing unit 20 and the compression processing unit 23 do not operate. As shown in FIG. 2, when the color image data read by the color image input device 1 is transmitted from the transmission device 3, the black generation and under color removal unit 18 and the gradation reproduction processing unit 22 do not operate.

  The transmission device 3 can be connected to a communication network such as a public network (not shown), a LAN (Local Area Network), or the Internet, and transmits a compressed file to the outside via the communication network by a communication method such as facsimile or e-mail. To do. For example, when the scan to e-mail mode is selected on the operation panel 4, the compressed file is attached to the e-mail by a transmission device using a network card, a modem, etc., and transmitted to the set transmission destination. The

  When performing facsimile transmission, the CPU of the image forming apparatus 100 performs a communication procedure with the other party in the transmission apparatus 3 using a modem, and compression is performed when a transmission possible state is secured. After performing necessary processing such as changing the compression format on the file, the file is sequentially transmitted to the other party via a communication line. When receiving a facsimile, the CPU of the image forming apparatus 100 receives a compressed file transmitted from the other party while performing a communication procedure at the transmitting apparatus 3 and outputs the compressed file to the color image processing apparatus 50.

  In the color image processing apparatus 50, the received compressed file is decompressed by an unillustrated decompression processing unit. The image data obtained by decompressing the compressed file is subjected to rotation processing and / or resolution conversion processing by a processing unit (not shown) as necessary, and the output gradation correction unit 21 outputs the output level. Tone correction is performed, and the gradation reproduction processing unit 22 performs gradation reproduction processing. The image data that has undergone various types of image processing is output to the color image output device 2, and an output image is formed on the recording sheet by the color image output device 2.

  Hereinafter, image processing in the color image processing apparatus 50 will be described in detail.

  The A / D conversion unit 11 receives an RGB analog signal input from the color image input device 1, converts the RGB analog signal into an RGB digital signal (that is, an RGB signal), and converts the converted RGB signal into a shading correction unit. 12 is output.

  The shading correction unit 12 performs processing for removing various distortions generated in the illumination system, the imaging system, and the imaging system of the color image input apparatus 1 on the RGB signal input from the A / D conversion unit 11. The shading correction unit 12 then outputs the RGB signal from which the distortion has been removed to the document type determination unit 13.

  The document type discrimination unit 13 converts the RGB reflectance signal input from the shading correction unit 12 into a density signal indicating the density of each RGB color, and discriminates the mode of the document such as a character, a printed photograph, or a photographic paper photograph. Document type determination processing is executed. When the user manually sets the document type determination process using the operation panel 4, the document type determination unit 13 outputs the RGB signal input from the shading correction unit 12 to the input tone correction unit 14 at the subsequent stage as it is. The processing result of the document type determination processing is reflected in the subsequent image processing.

  The input tone correction unit 14 performs image quality adjustment processing such as color balance adjustment, background density removal, and contrast adjustment on the RGB signals. Next, the input tone correction unit 14 outputs the processed RGB signal to the region separation processing unit 15.

  The region separation processing unit 15 separates each pixel in the image represented by the RGB signal input from the input tone correction unit 14 into one of a character region, a dot region, and a photo region. Further, the region separation processing unit 15 generates a region identification signal indicating which region each pixel belongs to based on the separation result, and generates a black generation and under color removal unit 18, a spatial filter processing unit 19, and a gradation reproduction processing unit. 22 and the processing parameter determination unit 17.

  The region separation processing unit 15 outputs the RGB signal input from the input tone correction unit 14 to the subsequent color correction unit 16 and the processing parameter determination unit 17 as they are.

  In the case of final output to the color image output device 2, the color correction unit 16 converts the RGB signal input from the region separation processing unit 15 into a CMY digital signal (hereinafter referred to as a CMY signal) to realize faithful color reproduction. Therefore, a process of removing color turbidity from the CMY signal based on the spectral characteristics of the CMY color material including unnecessary absorption components is performed. Then, the color correction unit 16 outputs the color-corrected CMY signal to the black color generation / under color removal unit 18. On the other hand, in the case of final output to the transmission device 3, the color correction unit 16 converts the RGB signal input from the region separation processing unit 15 into R′G′B ′ image data suitable for the display characteristics of the display device. Convert.

  In the case of final output to the color image output device 2, the black color generation under color removal unit 18 generates a black (K) signal from the CMY signal based on the CMY signal input from the color correction unit 16, and CMY A process of generating a new CMY signal by subtracting the K signal obtained by black generation from the signal is performed. As a result, the CMY3 color digital signals are converted into CMYK 4 color digital signals (hereinafter referred to as CMYK signals). Then, the black generation and under color removal unit 18 outputs the CMYK signal obtained by converting the CMY signal to the spatial filter processing unit 19.

  As an example of the black color generation process, a method of black color generation using skeleton black is generally used. In this method, the input / output characteristic of the skeleton curve is y = f (x), the input data is C, M, Y, the output data is C ′, M ′, Y ′, K ′, UCR (Under When the color removal rate is α (0 <α <1), the black color generation and under color removal processing is expressed by the following equations (1) to (4).

Formula (1): K ′ = f (min (C, M, Y))
Formula (2): C ′ = C−αK ′
Formula (3): M ′ = M−αK ′
Formula (4): Y ′ = Y−αK ′

  Here, the UCR rate α (0 <α <1) indicates how much CMY is reduced by replacing the portion where CMY overlaps with K. Equation (1) indicates that the K signal is generated in accordance with the smallest signal strength among the signal strengths of CMY.

  In the case of final output to the color image output device 2, the spatial filter processing unit 19 receives the region identification signal input from the region separation processing unit 15 with respect to the image data of the CMYK signal input from the black generation and under color removal unit 18. Based on the above, a spatial filter process using a digital filter is performed to correct a spatial frequency characteristic, thereby improving image blurring or graininess degradation.

  For example, for a region separated into characters by the region separation processing unit 15, the spatial filter processing unit 19 uses a filter with a large amount of high-frequency component enhancement in order to improve character reproducibility. To do. Further, the spatial filter processing unit 19 performs low-pass filter processing for removing the input halftone dot component on the region separated into halftone dots by the region separation processing unit 15.

  On the other hand, in the case of final output to the transmission device 3, the spatial filter processing unit 19 similarly receives the R′G′B ′ signal input from the color correction unit 16 and is input from the region separation processing unit 15. A spatial filter process using a digital filter is performed based on the region identification signal and the spatial frequency characteristics are corrected, thereby improving image blurring or graininess degradation.

  When the final output is to the color image output device 2, the spatial filter processing unit 19 outputs the processed CMYK signal to the output tone correction unit 21. On the other hand, in the case of final output to the transmission device 3, the spatial filter processing unit 19 outputs the processed R′G′B ′ signal to the resolution conversion processing unit 20.

  In the case of final output to the transmission device 3, the resolution conversion processing unit 20 as a conversion unit performs resolution conversion processing so that image data with a desired resolution determined by the processing parameter determination unit 17 is obtained. For example, in a simple example, when the input resolution of the scanner is 600 dpi × 300 dpi and the resolution determined by the processing parameter determination unit 17 is 300 dpi × 300 dpi, the resolution conversion processing unit 20 is 2 in the main scanning direction. By calculating an average value for each pixel and using it as an output value, resolution conversion from 600 dpi × 300 dpi to 300 dpi × 300 dpi is performed.

  In the case of final output to the color image output device 2, the output tone correction unit 21 outputs the CMYK signal input from the spatial filter processing unit 19 based on the dot area ratio that is a characteristic of the color image output device 2. The gradation correction process is performed, and the CMYK signal after the output gradation correction process is output to the gradation reproduction processing unit 22.

  On the other hand, in the case of final output to the transmission device 3, the output tone correction unit 21 makes the fog or highlight background disappear as necessary for the RGB signals input from the resolution conversion processing unit 20, or The output tone correction is performed so that the output tone is reduced, and the RGB signal after the output tone correction processing is output to the compression processing unit 23.

  The gradation reproduction processing unit 22 performs halftone processing corresponding to the region on the CMYK signal input from the output gradation correction unit 21 based on the region identification signal input from the region separation processing unit 15. For example, for a region separated into characters by the region separation processing unit 15, the gradation reproduction processing unit 22 performs binarization or multi-value conversion using a high-resolution screen suitable for reproducing high-frequency components. Perform processing. In addition, for a region separated into halftone dots by the region separation processing unit 15, the gradation reproduction processing unit 22 performs binarization or multi-value processing on the screen with an emphasis on gradation reproducibility. .

  The gradation reproduction processing unit 22 outputs the processed image data to the color image output device 2.

  The compression processing unit 23 has a function as a compression unit, and performs compression processing such as JPEG on the image data including the R′G′B ′ signal according to the setting of the file format of the operation panel 4 as necessary. The compressed data is generated and output to the transmission device 3. Note that the compressed image data may be stored in a storage medium such as a USB memory or a hard disk, instead of being configured to transmit the image data to the transmission device 3.

  When the scan function is selected in the automatic setting mode on the operation panel 4, the processing parameter determination unit 17 determines an output resolution for converting the image data into an optimal resolution by the resolution conversion processing unit 20.

  Next, the processing parameter determination unit 17 will be described in detail. FIG. 3 is a block diagram illustrating an example of the configuration of the processing parameter determination unit 17 according to the first embodiment. The processing parameter determination unit 17 includes an edge detection unit 171, a character rectangle extraction unit 172, a character string extraction unit 173, a resolution determination unit 174, and the like. The processing parameter determination unit 17 performs edge detection on the input RGB signal by the edge detection unit 171, extracts a character rectangle based on the character edge detected by the character rectangle extraction unit 172, and performs a character string extraction unit 173. A character string is extracted based on the character rectangle extracted in step (1), and the resolution determination unit 174 determines the output resolution from the character string and the character rectangle.

  FIG. 4 is a flowchart illustrating an example of a processing procedure of the edge detection unit 171. As shown in FIG. 4, the edge detection unit 171 performs edge detection on the input RGB signal using an edge detection filter such as Sobel or Laplacian (S11). The edge detection unit 171 performs character edge expansion processing (expansion processing) on the detected edge (S12), extracts character edge candidates, and ends the processing.

  FIG. 5 (FIGS. 5A to 5E) is a schematic diagram showing an example of edge detection processing by the edge detection unit 171. FIG. FIG. 5A shows a part (10 × 10 pixels) of the G signal of the RGB image data. The numerical value in each frame is a pixel value. Although only the process for the G signal will be described here, the same process is performed for the R signal or the B signal. FIG. 5B shows a filter for edge detection. As the edge detection filter, for example, a Laplacian filter can be used as shown in FIG. 5B. Note that the filter is not limited to the example of FIG. 5B.

  FIG. 5C shows the result of filtering the 8 × 8 pixels surrounded by the thick frame in FIG. 5A with the filter shown in FIG. 5B. FIG. 5D shows an example in which only pixels having a pixel value of 100 or more are extracted as edge pixels (value is 1) and pixels having a pixel value of less than 100 are set to 0 with respect to the example of FIG. 5C. The thick frame (8 × 8 pixels) in FIG. 5A corresponds to the thick frame (8 × 8 pixels) in FIG. 5D.

  FIG. 5E shows the result of character edge expansion (expansion processing) performed on the example of FIG. 5D (the thick frame portion). In the expansion process, when the target pixel is 0, if there are edge pixels (pixels having a value of 1) in two or more of the eight neighboring pixels, the value of the target pixel is set to 1. Process. Let the example of FIG. 5E be a character edge candidate.

  As described above, by performing dilation processing after edge detection, as shown in FIG. 5, as a result of performing threshold processing on the filter processing result, pixels that are originally edges but are not determined to be edges are detected. In some cases, such pixels can be relieved and edges can be detected with high accuracy. Moreover, the connection of the pixels constituting the character can be improved.

  FIG. 6 is a flowchart illustrating an example of a processing procedure of the character rectangle extraction unit 172. The character rectangle extraction unit 172 has a function as a character generation unit. That is, the character rectangle extraction unit 172 generates character specifying information (also referred to as character rectangle information) that specifies character candidates (also referred to as character rectangles) included in the image based on the pixel values of a plurality of pixels constituting the image. . Character candidates can be specified for each character. The character specifying information (character rectangle information) is position information for defining a character candidate (character rectangle). For example, the upper left and lower right coordinate information of the character rectangle, the horizontal dimension (width) of the character rectangle, and Including vertical dimensions (height).

  More specifically, as shown in FIG. 6, the character rectangle extraction unit 172 performs a labeling process on the character edge candidates (for example, the example of FIG. 5E) output by the edge detection unit 171 (S21), Based on the connection of edges, the size of one character is extracted in units of rectangles as a character rectangle (also referred to as a character candidate or a character rectangle region).

  After extracting the character rectangle, the character rectangle extraction unit 172 compares each extracted rectangle with a neighboring rectangle size when the size difference is larger than a predetermined value, and a rectangle similar to the vicinity. A rectangle that does not exist is removed as a non-character rectangle (S22), and the process is terminated with the remaining rectangle as character rectangle information.

  FIG. 7 (FIGS. 7A to 7E) is a schematic diagram showing an example of character rectangle extraction processing by the character rectangle extraction unit 172. FIG. FIG. 7A is an example of the edge detection result (character edge candidate) extracted by the edge detection unit 171. 7A and 5E show character edge candidates, but the example of FIG. 7A and the example of FIG. 5E are different for convenience.

  First, the character rectangle extraction unit 172 performs a labeling process on the edge detection result (character edge candidate) extracted by the edge detection unit 171. FIG. 7B shows a target pixel and a reference pixel when performing a labeling process. If the pixel indicated by * in FIG. 7B is the target pixel, the initial label of the target pixel is determined with reference to the labeling values of four reference pixels (indicated by symbols a, b, c, and d) in the vicinity of the target pixel. . Specifically, when there is no pixel labeled with four reference pixels, a new label value is assigned to the target pixel. Further, when a plurality of label values exist in the four reference pixels, the minimum label value among them is given to the target pixel. FIG. 7C shows a result of adding an initial label to the edge detection result (character edge candidate) illustrated in FIG. 7A.

  Further, when there are a plurality of label values in the eight pixels around the target pixel to which the label value is assigned, the minimum label value among the initial label to which the target pixel is given and the label value of the eight pixels around the target pixel Is used as a correction label. FIG. 7D is an example showing the association between the initial label and the correction label. That is, in FIG. 7D, when an initial label is assigned to the pixel of interest, when there are a plurality of labels in the eight pixels around the pixel of interest, the label other than the minimum label value (initial label assigned to the pixel of interest) Is associated with the minimum label value (initial label).

  As shown in FIG. 7D, for example, label 3 (initial label 3) indicates that label 1 (correction label 1) is associated. In addition, label 6 (initial label 6) is related to label 5 (corrected label 5) at the initial association stage, but it is known that label 5 is related to label 4, Eventually, label 6 (initial label 6) is related to label 4 (correction label 4). FIG. 7D shows the relationship between the initial label and the final minimum label value (corrected label).

  FIG. 7E shows the result of performing the correction process so that the label value becomes the minimum label value by referring to the label of each pixel shown in FIG. 7C using the correspondence relationship between the initial label and the correction label shown in FIG. 7D. .

  In addition, the minimum value (upper left position of the rectangular area) and the maximum value (lower right position of the rectangular area) of the rectangular area surrounding the pixels with the same label value when correction processing is performed so that the minimum label value is obtained. By obtaining, the rectangular size of the rectangular area including the same label value is obtained. The thick line in FIG. 7E indicates a rectangular area. In FIG. 7E, four rectangular regions with label values of 1, 2, 4, and 7 are illustrated. The minimum value (upper left position of the rectangular area) and the maximum value (lower right position of the rectangular area) of each rectangular area are the rectangular size of each label (1, 2, 4, 7 in the example of FIG. 7E). Become.

  Next, items that cannot be determined as character rectangles for each label are excluded. Specifically, a rectangular area having 16 pixels in the horizontal direction and having 2 pixels in the vertical direction as shown in a label 4 in FIG. 7E, which is long in the horizontal direction or in the vertical direction, is horizontal compared to the length in the vertical direction. Because the length of the direction is long, there is a possibility of a horizontal line segment. Therefore, it is determined that the rectangular area of the label 4 is not a character rectangle and is excluded. In this case, in order to determine whether or not the rectangular region is a line segment, for example, a difference between the number of pixels in the horizontal direction and the number of pixels in the vertical direction is calculated, and the calculated difference is equal to or greater than a predetermined threshold (for example, , 12 pixels or more), it can be determined as a line segment.

  Among the obtained rectangular areas, a rectangular area whose size is too small, for example, a label having a total of 4 pixels in each of the horizontal direction and the vertical direction as in the label 7 in FIG. May be a noisy pixel. Therefore, it is determined that the rectangular area of the label 7 is not a character rectangle and is excluded. In this case, in order to determine whether the size of the rectangular area is small, for example, the number of pixels in the horizontal direction and the number of pixels in the vertical direction are added, and the added value is equal to or less than a predetermined threshold (for example, 6 pixels or less). In some cases, it can be determined as a noise pixel.

  Although not illustrated in FIG. 7E, a label whose size of the rectangular area is too large, for example, a rectangular area whose size is about 1/4 of the size of the input image can be an image instead of a character. It can be eliminated because of its nature. In this case, when determining that the size of the rectangular area is large, the ratio of the number of pixels of the rectangular area to the number of pixels of the input image is obtained, and the ratio is equal to or larger than a predetermined threshold (for example, 0.25 or larger). In this case, it can be determined that the rectangular area is a label that is too large. By performing the non-character rectangle removal process as described above, only label 1 and label 2 in FIG. 7E are finally extracted as character rectangles.

  FIG. 8 is a flowchart illustrating an example of a processing procedure of the character string extraction unit 173. The character string extraction unit 173 functions as a character string generation unit. That is, the character string extraction unit 173 selects character string candidates (also referred to as character strings) composed of a plurality of characters based on the character specifying information (character rectangle information) generated by the character rectangle extraction unit 172 (character generation unit). Character string specifying information to be specified (also referred to as character string information) is generated. The character string specifying information (character string information) is position information for defining character string candidates (character strings). For example, upper left and lower right coordinate information of the character string candidates, horizontal dimensions of the character string candidates (Width) and vertical dimension (height). The character string specifying information includes the direction of the character string (horizontal direction, vertical direction).

  More specifically, as shown in FIG. 8, the character string extraction unit 173 calculates the distance between adjacent character rectangles with respect to the character rectangle information output by the character rectangle extraction unit 172 (S31), and is close to it. The character rectangles are integrated as a group of the same character string, and character string information is generated.

  The character string extraction unit 173 estimates the direction of the character string from the length of the character string with respect to the character string information in which the character rectangles are integrated (S32). Further, the character string extraction unit 173 further applies a character rectangle that is determined to be in a different direction, or a character rectangle that belongs to a character string that is determined to be in a different direction, or a character rectangle that is determined to be in a different direction. If the character rectangle can be corrected in the correct character string direction, the character string direction is corrected (S33).

  The character string extraction unit 173 extracts a character string area from the corrected character string information (S34), and the character string determined not to belong to the extracted character string area is a character string in a non-character string area. (S35), and the remaining information is used as final character string information, and the process is terminated.

  FIG. 9 (FIGS. 9A to 9D) is a schematic diagram showing an example of character string extraction processing by the character string extraction unit 173. FIG. 9A shows an example of a character rectangle extraction result extracted by the character rectangle extraction unit 172. In the example of FIG. 9A, 21 character rectangles (character candidates) are extracted.

  The distance between adjacent character rectangles is calculated for the character rectangle extraction result shown in FIG. 9A. First, a character string size having the same size as the target character rectangle is set as character string information for an arbitrary target character rectangle. As the method for obtaining the distance of the neighboring character rectangle, the upper left corner is close to the upper right vertex coordinate of the target character rectangle (for example, within ± 5 pixels in the vertical direction and within 10 pixels in the horizontal direction). Search from the coordinate information of all character rectangles to determine whether there is a character rectangle with vertex coordinates or whether there is a character rectangle with the lower left vertex coordinates close to the lower right vertex coordinate of the target character rectangle Then, a character rectangle at a close position is extracted as a nearby character rectangle.

  In addition, there is a character rectangle having an upper left vertex coordinate that is close to the lower left vertex coordinate of the target character rectangle in the vertical direction (for example, within ± 5 pixels in the horizontal direction and within 10 pixels in the vertical direction). Or whether there is a character rectangle with the upper right vertex coordinates close to the lower right vertex coordinates of the target character rectangle. Extract as a nearby character rectangle.

  When adjacent character rectangles exist in both the horizontal direction and the vertical direction, the closest character rectangle is selected as the adjacent character rectangle. Then, the extracted neighboring character rectangle is integrated as a character rectangle of the same group as the target character rectangle (labeled as the same character string), and the adjacent character rectangle is connected to the character string size of the target character rectangle The character string information is updated with the new size as the new character string size.

  If there is no neighboring character rectangle, the character string size of the character string information is left as the character rectangle size, and the process proceeds to the next character rectangle of interest. For the next attention character rectangle, a character rectangle that does not yet belong to any character string group is selected. Finally, no character string extraction processing has been performed, and any character string group is not selected. The character string extraction process is repeated until there is no character rectangle that does not belong.

  FIG. 9B illustrates the result of performing the above-described processing on the character rectangle illustrated in FIG. 9A. In FIG. 9B, a character string extracted from a portion surrounded by a thick frame is shown. In the example of FIG. 9B, six groups of character strings with reference numerals 1 to 6 are extracted.

  When the character string extraction process is completed, the horizontal and vertical sizes of the character strings of each group are calculated, and the longer direction is set as the direction of the character string. For example, in the case of a character string having 100 pixels in the horizontal direction and 10 pixels in the vertical direction, the character string is set as the horizontal character string.

  Further, when the size of the character string in the horizontal direction and the size in the vertical direction are approximately the same (for example, when the horizontal direction is 15 pixels and the vertical direction is 10 pixels), the direction of the character string is determined. Therefore, the character string direction is set as indefinite. In the example of FIG. 9B, the direction of the character strings 1, 2, and 3 is the horizontal direction (also referred to as a horizontal direction character string), and the character strings 4, 5, and 6 (including the character rectangle) are directionally indefinite characters. It becomes a column.

  After the character string direction is set for the character strings of all groups, the character string direction is indefinite, and is corrected in the correct character string direction based on the character string information related to the nearby character string. In the example of FIG. 9B, the directionally indefinite character string of reference numeral 4 is in substantially the same horizontal position (for example, the horizontal position is within ± 5 pixels) as the character string of the group with reference numeral 3 located near the upper side, and Since the distance in the vertical direction is also relatively short (for example, 8 pixels or less), it is determined that the character string is in the same direction as the character string of the group of 3. Thereby, the direction indefinite character string of the code | symbol 4 is correct | amended as a character string of a horizontal direction.

  In addition, the direction-indefinite character strings of reference numerals 5 and 6 are left as direction-indefinite character strings because no character string exists in the vicinity.

  FIG. 9C shows an example of the result of correcting the character string direction for the character string illustrated in FIG. 9B. In FIG. 9C, a character string (reference numerals 1, 2, 3, 4) surrounded by a thick frame has a fixed character string direction.

  Next, a character string located in the vicinity (also referred to as a neighborhood character string) and close in size, or the start position of the character string (the left end position of the character string) or the end position (the right end position of the character string) is close ( For example, within 8 pixels), grouping is performed. Thereby, a character string area is extracted. In the case of FIG. 9C, the character strings 1 to 4 are horizontal character strings, and since the start position or end position of the character string is close, the character strings 1 to 4 are character strings in the same group. Extract as a region. The character string area is a collection of one or more character strings having the same direction. Then, the direction indefinite character strings of reference numerals 5 and 6 that do not belong to the character string area are removed from the character string information as non-character string areas. The result of extracting the character string area and removing the non-character string area is illustrated in FIG. 9D. That is, the character string area illustrated in FIG. 9D includes a character string (rectangular area indicated by reference numerals 1 to 4) in which the direction of the character string is determined. If there is no character string area, it is determined that there is no character string and the character string information is cleared.

  FIG. 10 is a flowchart illustrating an example of a processing procedure of the resolution determination unit 174. The resolution determination unit 174 has a function as a resolution determination unit. That is, the resolution determination unit 174 determines the resolution of the image based on the character string specifying information (character string information) generated by the character string generation unit (character string extraction unit 173). Further, the resolution determination unit 174 determines the resolution of the image based on the character specifying information (character rectangle information) generated by the character generation unit (character rectangle extraction unit 172).

  The resolution determination unit 174 has a function as a character size determination unit. That is, the resolution determination unit 174 determines one or a plurality of character sizes based on the generated character string specifying information or character specifying information. For example, the character size (for example, 8 points, 10 points, 12 points, etc.) is determined according to the height or width of the character string candidate or the height or width of the character candidate. The resolution determination unit 174 determines the resolution based on the determined character size. For example, the larger the character size, the lower the resolution, and the smaller the character size, the higher the resolution. Thus, the optimum output resolution can be automatically set, and desired output data (output image after resolution conversion) can be output without complicated settings by the user.

  The resolution determination unit 174 has a function as a determination unit, and determines whether or not the character string specifying information (character string information) can be generated by the character string generation unit (character string extraction unit 173).

  More specifically, as shown in FIG. 10, the resolution determination unit 174 determines whether or not the condition 1 is satisfied (S41). Condition 1 is a condition that a character string exists or a condition that character string information exists.

  When the condition 1 is satisfied (YES in S41), the resolution determination unit 174 classifies and counts for each similar rectangular size based on the rectangular size of the character string of each group using the character string information, The character string information for each size is counted (S42). The resolution determination unit 174 determines an optimum output resolution based on the character string size from the counted value (S43), outputs an optimum output resolution determination result based on the character string size (S44), and ends the process.

  When the condition 1 is not satisfied (NO in S41), the resolution determination unit 174 determines whether or not the condition 2 is satisfied (S45). Condition 2 is a condition that a character rectangle exists, or a condition that character rectangle information exists.

  When the condition 2 is satisfied (YES in S45), the resolution determination unit 174 performs classification by classifying the rectangles for each similar rectangle size based on the character rectangle size using the character rectangle information, and character rectangles for each size. Information is counted (S46). The resolution determination unit 174 determines the optimum output resolution based on the character rectangle size from the counted value (S47), outputs the optimum output resolution determination result based on the character rectangle size (S48), and ends the process.

  When the condition 2 is not satisfied (NO in S45), the resolution determination unit 174 determines the optimum output resolution based on the document type information (S49). The document type information includes, for example, information such as whether the document is a printed photo document or a photographic paper photo document. The resolution determination unit 174 outputs the optimum output resolution determination result when there is no character (S50), and ends the process. The output optimum output resolution information is output to the resolution conversion processing unit 20, and the resolution conversion processing unit 20 converts the input image data (image) into a desired resolution.

  FIG. 11 (FIGS. 11A and 11B) is a schematic diagram illustrating an example of resolution determination processing by the resolution determination unit 174. FIG. 11A shows an example of character string information extracted by the character string extraction unit 173. When the character string information illustrated in FIG. 11A exists, the above condition 1 is satisfied (step S41 in FIG. 10). When the condition 1 is satisfied, the character string information for each size is counted in order to determine the optimum output resolution based on the character string information.

  As shown in FIG. 11A, the character string information includes the label number (character string label) of the character string extracted by the character string extraction unit 173, the upper left coordinate information (sx, sy), and the lower right corner of the character string. Coordinate information (ex, ey), character string direction (for example, reference numeral 1 is a horizontal direction and reference numeral 2 is a vertical direction). The character string is, for example, a rectangular area with reference numerals 1 to 4 as illustrated in FIGS. 9C and 9D.

  When the character string direction is the horizontal direction, the height (ey-sy) of the character string is calculated as a rectangular size, and when the character string direction is the vertical direction, the character string width (ex-sx) is calculated. Is calculated as a rectangle size.

  FIG. 12 is an explanatory diagram showing an example of the relationship between the character string or the rectangular size of the character and the output resolution. FIG. 12 illustrates the correspondence relationship between the character string or character rectangle size, character size, output resolution, and image compression rate. Note that the rectangle size, character size, output resolution, and the like are examples, and are not limited to the values shown in FIG. For example, a rectangular size can be set for all existing character sizes. Further, since the relationship between the rectangular size and the character size depends on the reading resolution of the scanner or the like, it is not always fixed to the size illustrated in FIG.

  It is determined which character size corresponds to the rectangular size calculated based on the height or width of the character string, and is counted for each character size. For example, in FIG. 11A, only the character string whose character string label is 1, the height (ey-sy) of the character string is 50, which corresponds to the rectangular size 52 ± 4 in FIG. Further, the height of the character strings of the other character string labels (2 to 9) is 30, which corresponds to the rectangular size 30 ± 2 in FIG. In this case, the value counted by character size results in one 14-point character string corresponding to the rectangular size 52 ± 4 and eight 8-point character strings corresponding to the rectangular size 30 ± 2.

  In FIG. 11A, for the sake of convenience, an example of the character string height of the same size is shown. However, in actuality, in the character string extraction result for the image read by the scanner, the rectangular size for each character size is not necessarily shown. Like 11A, it is not highly likely that the values will be exactly the same. For this reason, as illustrated in FIG. 12, an allowable range of several pixels is provided for the rectangular size. In addition, the allowable ranges of rectangular sizes corresponding to adjacent character sizes are overlapped. For this reason, if the height of the character string is 32, the character size may be 9 points and may be 8 points. Thereby, even when a character string of an ambiguous size is extracted by character string extraction, the influence can be reduced and the output resolution can be easily determined.

  As a method for determining the output resolution, a threshold is set, and a character size having a count value equal to or larger than the set threshold is targeted. For example, the output resolution can be determined based on the smallest character size among the target character sizes. This is a determination method for the purpose of outputting the smallest character at an easily readable resolution. In addition to this, there is a method of determining the output resolution based on the character frequency of the maximum frequency among the target character sizes. This is a determination method for the purpose of outputting the most existing characters at a resolution that is easy to read, and for the purpose of suppressing the file size while maintaining the character readability.

  For example, when the output resolution is determined based on the smallest character size among the target character sizes, the threshold is set to 5 in the example of FIG. In this case, since there are eight 8-point character strings as the minimum character size, 300 × 300 dpi corresponding to the character size of 8 points is determined as the output resolution (optimum output resolution) from FIG.

  As described above, when the direction of the character string is the horizontal direction, the resolution is determined according to the vertical dimension (height) of the character string candidate (character string). For example, the resolution is lowered as the height is increased, and the resolution is increased as the height is decreased. When the direction of the character string is the vertical direction, the resolution is determined according to the horizontal dimension (width) of the character string candidate (character string). For example, the larger the width, the lower the resolution, and the smaller the width, the higher the resolution. Thus, the optimum output resolution can be automatically set, and desired output data (output image after resolution conversion) can be output without complicated settings by the user.

  FIG. 11B shows an example of character rectangle information extracted by the character rectangle extraction unit 172. When the character string information illustrated in FIG. 11A does not exist, the above condition 1 is not satisfied (step S41 in FIG. 10). When the condition 2 is satisfied (step S45 in FIG. 10), the character string information for each size is counted in order to determine the optimum output resolution based on the character rectangle information.

  As shown in FIG. 11B, the character rectangle information includes the label number (character rectangle label) of the character rectangle extracted by the character rectangle extraction unit 172, the upper left coordinate information (sx, sy) of the character rectangle, and the lower right corner of the character rectangle. Coordinate information (ex, ey), a character string to which the character rectangle belongs (in the case of -1, it represents a character rectangle that does not belong to any character string). For convenience, the example of FIG. 11B shows a case where all the character rectangles do not belong to any character string, but actually, a character rectangle belonging to the character string and a character rectangle not belonging to the character string are mixed. .

  Then, the height (ey-sy) and width (ex-sx) of the character rectangle are calculated, and the longer of the calculated height or width is calculated as the character rectangle size. Counting is performed for each character size by determining which character size corresponds to the calculated rectangular size illustrated in FIG. Next, as in the case of the character string, the output resolution is determined based on the smallest character size with respect to the character size of the count value equal to or greater than the threshold value.

  As described above, the resolution is determined according to the vertical dimension (height) or horizontal dimension (width) of the character candidate (character rectangle). When the height and width of the character candidates are different, the longer (larger) can be used. The resolution is decreased as the height or width of the character candidate is increased, and the resolution is increased as the height or width is decreased. Thus, the optimum output resolution can be automatically set, and desired output data (output image after resolution conversion) can be output without complicated settings by the user.

  If neither of the conditions 1 and 2 is satisfied, the optimum output resolution is determined based on the document type information such as whether it is a printed photo document or a photographic paper photo document, and is output. For a printed photo original, for example, the output resolution can be determined to be 200 × 200 dpi, and for a photographic paper photo original, the output resolution can be determined to be 300 × 300 dpi. The output resolution based on these document type information may always be a fixed value, or a default value may be set in advance according to the purpose.

  In the above example, the reason why the condition 1 is determined before the condition 2, that is, the reason why the character string information is given priority over the character rectangle information when determining (determining) the resolution is that the character string includes a plurality of character rectangles. This is because a stable character size can be determined.

  The resolution conversion can be performed by the following two methods by storing the read image data. The first method uses a one-pass scan, estimates the resolution, and sets the resolution in the main scanning direction and the sub-scanning direction by interpolation calculation. The second method is a method using two-pass scanning, in which the resolution is estimated, the reading speed of the scanner is changed, and the original is read again. In this case, the resolution in the sub-scanning direction is optically changed, and the resolution in the main scanning direction is set by interpolation calculation.

  Next, a case will be described in which which of the image quality and the image size (file size) to be given priority is set. The operation panel 4 has a function as a setting unit, and sets which of image quality or image size has priority. The operation panel 4 allows the user to select whether to prioritize image quality or image size.

  For example, when priority is given to image quality over image size, a lower threshold is set so that many target character sizes are included, and the output resolution is determined based on the smallest character size among them. . For example, if there are 100 12-point character strings and 10 9-point character strings, a threshold value is set to 3 in advance, and only character sizes with a count value of 3 or more are targeted. At this time, since there are 10 character strings of 9 points as the minimum character size, the optimum output resolution can be determined to be 200 × 200 dpi based on FIG.

  That is, when the resolution determination unit 174 is set to prioritize image quality, the resolution determination unit 174 determines the minimum character size among one or more character sizes based on character string information or character information. When determining the minimum character size, it is possible to set a threshold value in advance and determine the minimum character size among the character sizes that exist as many as the threshold value. By determining the resolution based on the minimum character size, an output result (output image) in which the smallest character can be sufficiently recognized can be obtained.

  In addition, when the image size is prioritized over the image quality, the threshold is set higher, the target character size is small, and the output resolution is determined based on the maximum frequency character size. For example, if there are 100 12-point character strings and 10 9-point character strings, a threshold value is set to 20 in advance, and only character sizes with a count value of 20 or more are targeted. At this time, since there are 100 12-point character strings having a character size equal to or greater than the threshold, the optimum output resolution can be determined to be 150 × 150 dpi based on FIG.

  That is, when the resolution determination unit 174 is set to prioritize the image size, the resolution determination unit 174 determines the most frequently used character size among one or more character sizes based on the character string information or the character information. When determining the maximum frequency character size, it is possible to set a threshold value in advance and determine the maximum frequency character size among the character sizes that exist as many as the threshold value. By determining the resolution based on the most frequent character size, the most characters in the image can be fully recognized, and the resolution is relatively low, reducing the visibility of the characters in the entire image. The file size can be reduced without any problems.

  It is also possible to provide option settings that are set as default values of the image forming apparatus 100 by administrator settings or the like.

  By the above-described processing, the optimum output resolution is finally determined, and the determined optimum output resolution information is output to the resolution conversion processing unit 20, so that the scan data automatically converted at the optimum output resolution corresponding to the document can be obtained. It will be available.

(Second Embodiment)
FIG. 13 is a block diagram illustrating an example of the configuration of the image forming apparatus 110 according to the second embodiment. The difference from the first embodiment is that information such as the optimum compression rate determined based on the parameters (output resolution, character size, etc.) determined by the processing parameter determination unit 17 is output to the compression processing unit 23. In addition, the same code | symbol is attached | subjected to the location similar to 1st Embodiment, and description is abbreviate | omitted.

  FIG. 14 is a block diagram illustrating an example of the configuration of the processing parameter determination unit 17 according to the second embodiment. The resolution / compression rate determination unit 175 has the same function as the resolution determination unit 174 of the first embodiment. Furthermore, the resolution / compression rate determination unit 175 has a function as a compression rate determination unit, and determines the compression rate based on the determined resolution.

  More specifically, the point counting for each character size based on the character string or the rectangular size of the character rectangle is the same as in the first embodiment. In the second embodiment, the optimum combination of output resolution and compression rate is further determined from the character size determined based on the rectangular size of the character string or character rectangle.

  FIG. 15 is an explanatory diagram showing an example of the relationship between the character string or character rectangle size, output resolution, and compression rate. FIG. 15 illustrates the correspondence relationship between the character string or character rectangle size, character size, output resolution, and image compression rate. Note that the rectangle size, character size, output resolution, and compression rate are examples, and are not limited to the values shown in FIG.

  In FIG. 15, when importance is attached to image quality, when there is a combination of the same character size and different resolutions based on the minimum character size (for example, 10 points, 12 points, 14 points), the higher resolution is selected. The compression rate associated with the selected resolution is selected as the optimum compression rate.

  In addition, when emphasizing the file size (image size), if there is a combination with the same character size and a different resolution based on the maximum frequency character size, select the lower resolution and support the selected resolution The attached compression rate is determined as the optimum compression rate.

  If the image quality and the file size are not particularly emphasized, the minimum character size is used as a reference in FIG. For example, when the character size of 10 points is the minimum character size, two optimum output resolutions of 200 × 200 dpi and 150 × 150 dpi are conceivable. Next, the maximum frequency character size is obtained. If the maximum frequency character size is also 10 points, it is determined that 200 × 200 dpi is the optimum output resolution and the optimum compression rate is set to a high setting. On the other hand, when the maximum frequency character size is 12 points or more, 150 × 150 dpi is set as the optimum output resolution, and the optimum compression rate is determined to be medium or high.

  Further, when there is no character string information and character rectangle information, and there is no setting for emphasizing image quality and file size (or no consideration), when the document type is a printed photo document, for example, the output resolution is 200 ×. It is determined that it is 200 dpi, and it is determined that the compression rate is medium. When the document type is a photographic paper photograph document, the output resolution is determined to be 300 × 300 dpi, the compression rate is determined to be medium, and the determination result is output.

  Further, when there is no character string information and character rectangle information and there is a setting that emphasizes image quality or file size, if the document type is a printed photo document and image quality is prioritized, for example, the output resolution is 200. X200 dpi is determined, and the compression rate is determined to be low. When the document type is a photographic paper photo document and image quality is prioritized, it is determined that the output resolution is 300 × 300 dpi and the compression rate is low.

  In addition, when there is no character string information and character rectangle information and there is a setting that emphasizes image quality or file size, if the document type is a printed photo document and file size is prioritized, the output resolution is 150 ×. It is determined that it is 150 dpi, and it is determined that the compression rate is high. When the document type is a photographic paper photo document and the file size is given priority, it is determined that the output resolution is 200 × 200 dpi, the compression rate is determined to be high, and the determination result is output.

  The compression rate can be changed to high, medium, or low as described above by changing the value of the quantization table when converting the image data into the JPEG (Joint Photographic Experts Group) format. . When the resolution is low, the compression rate is lowered to suppress the degree of compression, thereby preventing the visibility from being lowered. Further, when the resolution is high, the file size (image size) can be reduced without reducing visibility by increasing the compression rate and increasing the degree of compression.

  The optimum output resolution and compression rate are finally determined by the above-described processing, the optimum output resolution information is output to the resolution conversion processing unit 20, and the optimum compression rate information is output to the compression processing unit 23 to automatically Scan data converted with an optimum output resolution and compression rate according to the original can be obtained.

(Third embodiment)
FIG. 16 is a block diagram illustrating an example of the configuration of the image reading apparatus 120 according to the third embodiment. FIG. 16 shows a configuration of an image reading apparatus 120 as an image processing system. As shown in FIG. 16, the image reading device 120 includes a color image input device 1, a color image processing device 50, a transmission device 3, and the like.

  The color image processing apparatus 50 includes an A / D conversion unit 11, a shading correction unit 12, a document type determination unit 13, an input tone correction unit 14, a region separation processing unit 15, a color correction unit 16, a spatial filter processing unit 19, and a resolution. A conversion processing unit 20, an output tone correction unit 21, a compression processing unit 23, a processing parameter determination unit 17, and the like are provided.

  The color image input device 1 is composed of, for example, a color scanner having a CCD (Charge Coupled Device). The color image input device 1 reads a reflected light image from a document as an analog signal of RGB (R: red, G: green, B: blue) using a CCD and outputs it to the color image processing device 50.

  An analog signal read by the color image input device 1 is converted into an A / D conversion unit 11, a shading correction unit 12, a document type determination unit 13, an input tone correction unit 14, an area separation process in the color image processing device 50. 15, color correction unit 16, processing parameter determination unit 17, spatial filter processing unit 19, resolution conversion processing unit 20, output tone correction unit 21, and compression processing unit 23. In addition, since the process of each part is the same as that of the case of 1st Embodiment, description is abbreviate | omitted.

  Although not shown, since the compression processing unit 23 performs optimal compression processing, it is possible to determine the same compression rate as in FIGS. 12 and 15. A digital camera can also be used as the image reading device.

  FIG. 17 is a schematic diagram illustrating an example of a resolution setting method according to the first to third embodiments. In the image forming apparatus 100, the image forming apparatus 110, and the image reading apparatus 120 according to the first to third embodiments, in the manual setting mode in the scan function, for example, the user needs to set the resolution as illustrated in FIG. 17A. is there. In addition, there are various setting items such as color settings, document mode settings, file format settings, and compression rate settings.

  In the automatic setting mode of the present embodiment, an operation panel as shown in FIG. 17B is prepared, and other settings including the resolution setting shown in FIG. Setting) is automatically performed, and optimal scan data can be acquired. For the automatic setting, for example, a function called ACS (Auto Color Select) has already been generally used for color, and color setting can be automatically performed by using the function. As for the document mode, the document mode can be automatically set by using the determination result of the document type determination unit 13. The compression rate can be automatically set together with the output resolution as described above. Regarding the file format, a frequently used file format (for example, PDF) can be preset as a default.

  FIG. 18 is a schematic diagram showing a display example of default settings. As shown in FIG. 17B, when there is a default setting button, by pushing the default setting button, only items that can be default set as shown in FIG. 18 are displayed and the user can arbitrarily select them. Here, the file format or priority item can be set in advance if desired. Further, when there is no default setting button, it is possible to set in advance only items that should be set by default, such as administrator settings. Thus, according to the present embodiment, it is possible to provide a function that allows the user to acquire optimal scan data with a single push without performing complicated settings.

  Each of the embodiments described above can also be applied to image data (images) downloaded via a network. Even if the resolution of the downloaded image data is unknown, the character size can be determined to determine the output resolution. Further, when downloading image data from a server such as a cloud, it is also possible to determine the resolution and / or compression rate on the server side and download the compressed image data. In this case, the server corresponds to the image processing apparatus of the present embodiment. In this case, the amount of image data to be downloaded can be reduced.

  In each of the above-described embodiments, the above-described resolution determination processing or compression rate determination is performed on a computer-readable recording medium in which program code (execution format program, intermediate code program, source program) of a program to be executed by a computer is recorded. It is also possible to record image processing to be processed. As a result, a recording medium on which a program code for performing the resolution determination process or the compression rate determination process described above is recorded can be provided in a portable manner.

  In the present embodiment, the recording medium may be a memory (not shown) such as a program medium such as a ROM because processing is performed by a microcomputer, or an external storage device (not shown). As a program medium, a program code reader may be provided, and the program medium can be read by inserting a recording medium into the reader.

  In any case, the stored program may be configured to be accessed and executed by the microprocessor, or in any case, the program code is read and the read program code is displayed on the microcomputer. The program code may be downloaded to a program storage area that has not been executed and the program code executed. It is assumed that this download program is stored in the main device in advance.

  Here, the program medium is a recording medium configured to be separable from the main body, such as a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a flexible disk or a hard disk, a CD-ROM / MO / MD / DVD, or the like. Optical disk system, IC card (including memory card) / optical card system, mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory: registered trademark), flash ROM, etc. It may be a medium that carries a fixed program including the semiconductor memory according to the above.

  In the present embodiment, since the system configuration is such that a communication network including the Internet can be connected, the medium may be a medium that dynamically carries a program so as to download a program code from the communication network. When the program code is downloaded from the communication network as described above, the download program may be stored in the main device in advance, or may be installed from another recording medium. This embodiment can also be realized in the form of a computer data signal embedded in a carrier wave, in which the above-described program code is embodied by electronic transmission. Further, the above-described image processing method is executed by reading the above-mentioned recording medium by a digital color image forming apparatus or a program reading apparatus provided in a computer system.

  The computer system includes an image input device such as a flatbed scanner, a film scanner, and a digital camera, a computer that performs various processes such as the image processing method by loading a predetermined program, and a CRT display that displays the processing results of the computer. An image display device such as a liquid crystal display and a printer that outputs the processing results of the computer to paper or the like. Furthermore, a network card or a modem is provided as communication means for connecting to a server or the like via a network.

  The technical features described in the above embodiments of the present invention can be combined with each other to form a new technical plan.

  The image processing apparatus (50) of the present embodiment includes a character generation unit (172) that generates character specifying information that specifies character candidates included in the image based on the pixel values of a plurality of pixels constituting the image; A character string generation unit (173) for generating character string specifying information for specifying a character string candidate composed of a plurality of characters based on the character specifying information generated by the character generation unit, and a character string generating unit A resolution determination unit (174, 175) that determines the resolution of the image based on character string specifying information, and a conversion unit (20) that converts the resolution of the image using the resolution determined by the resolution determination unit. It is characterized by that.

  The computer program according to the present embodiment is a computer program for converting the resolution of an image. The character specifying information for specifying a character candidate included in the image based on pixel values of a plurality of pixels constituting the image. Based on the generated character string specifying information, the character string generating means for generating character string specifying information for specifying a character string candidate composed of a plurality of characters based on the generated character specifying information, and the generated character string specifying information And a resolution determining unit that determines the resolution of the image, and a conversion unit that converts the resolution of the image using the determined resolution.

  In the present embodiment, the character generation unit generates character specifying information for specifying character candidates included in the image based on the pixel values of a plurality of pixels constituting the image. In order to specify a character candidate, for example, an edge pixel is extracted based on the pixel value of each pixel of the image, and among the extracted edge pixels, for example, labeling is performed by using adjacent edge pixels as one lump. The character rectangle surrounding the edge pixels of the same label can be used as a character candidate. Character candidates can be specified for each character. The character specifying information is position information for defining a character candidate (character rectangle). For example, the upper left and lower right coordinate information of the character rectangle, the horizontal dimension (width) and the vertical dimension ( Height).

  The character string generating unit generates character string specifying information for specifying a character string candidate composed of a plurality of characters based on the character specifying information generated by the character generating unit. The character string candidates are, for example, a plurality of character candidates arranged in the horizontal direction or the vertical direction. The character string specifying information is position information for defining character string candidates (character strings). For example, the upper left and lower right coordinate information of the character string area, the horizontal dimension (width) and vertical of the character string area Including direction dimensions (height). The character string specifying information includes the direction of the character string (horizontal direction, vertical direction).

  The resolution determining unit determines the resolution of the image based on the character string specifying information generated by the character string generating unit. The conversion unit converts the resolution of the image using the resolution determined by the resolution determination unit. For example, when the direction of the character string is the horizontal direction, the resolution is determined according to the vertical dimension (height) of the character string candidate (character string). For example, the resolution is lowered as the height is increased, and the resolution is increased as the height is decreased. When the direction of the character string is the vertical direction, the resolution is determined according to the horizontal dimension (width) of the character string candidate (character string). For example, the larger the width, the lower the resolution, and the smaller the width, the higher the resolution. Thus, the optimum output resolution can be automatically set, and desired output data (output image after resolution conversion) can be output without complicated settings by the user.

  The image processing apparatus (50) of the present embodiment includes a determination unit (174, 175) that determines whether the character string specifying information can be generated by the character string generation unit (173), and the resolution determination unit (174). 175), when the determination unit determines that the generation of the character string specifying information is not possible, the resolution of the image is determined based on the character specifying information generated by the character generating unit (172). It is characterized by being.

  In the present embodiment, the determination unit determines whether or not the character string specifying information can be generated by the character string generation unit. That is, the determination unit determines whether a character string exists in the image. If the determination unit determines that the character string specifying information is not generated, the resolution determining unit determines the resolution of the image based on the character specifying information generated by the character generating unit. That is, when there is no character string in the image, the resolution is determined based on the character specifying information.

  For example, the resolution is determined according to the vertical dimension (height) or horizontal dimension (width) of the character candidate (character rectangle). When the height and width of the character candidates are different, the longer (larger) can be used. The resolution is decreased as the height or width of the character candidate is increased, and the resolution is increased as the height or width is decreased. Thus, the optimum output resolution can be automatically set, and desired output data (output image after resolution conversion) can be output without complicated settings by the user.

  The image processing apparatus (50) of the present embodiment includes a character size determining unit (174, 175) that determines one or a plurality of character sizes based on the generated character string specifying information or character specifying information, and determining the resolution. The sections (174, 175) are characterized in that the resolution is determined based on the character size determined by the character size determination section.

  In the present embodiment, the character size determination unit determines one or more character sizes based on the generated character string specifying information or character specifying information. For example, the character size (for example, 8 points, 10 points, 12 points, etc.) is determined according to the height or width of the character string candidate or the height or width of the character candidate. The resolution determining unit determines the resolution based on the character size determined by the character size determining unit. For example, the larger the character size, the lower the resolution, and the smaller the character size, the higher the resolution. Thus, the optimum output resolution can be automatically set, and desired output data (output image after resolution conversion) can be output without complicated settings by the user.

  The image processing apparatus (50) of the present embodiment includes a setting unit (4) for setting which of image quality or image size is to be prioritized, and the character size determination unit (174, 175) is the setting unit. When the image quality is set to be prioritized, a minimum character size is determined from one or more character sizes based on the character string specifying information or the character specifying information, and the setting unit is set to prioritize the image size. In this case, the character frequency having the highest frequency is determined from one or a plurality of character sizes based on the character string specifying information or the character specifying information.

  In the present embodiment, the setting unit sets whether to give priority to image quality or image size. The setting unit is, for example, an operation panel for the user to set, and the user can select and set whether to give priority to image quality or image size. The character size determination unit determines a minimum character size among one or a plurality of character sizes based on the character string specifying information or the character specifying information when the image quality is set to be prioritized. When determining the minimum character size, it is possible to set a threshold value in advance and determine the minimum character size among the character sizes that exist as many as the threshold value. By determining the resolution based on the minimum character size, an output result (output image) in which the smallest character can be sufficiently recognized can be obtained.

  Further, the character size determination unit determines the most frequently used character size among one or a plurality of character sizes based on the character string specifying information or the character specifying information when the image size is set to be prioritized. When determining the maximum frequency character size, it is possible to set a threshold value in advance and determine the maximum frequency character size among the character sizes that exist as many as the threshold value. By determining the resolution based on the most frequent character size, the most characters in the image can be fully recognized, and the resolution is relatively low, reducing the visibility of the characters in the entire image. The file size can be reduced without any problems.

  The image processing apparatus (50) of the present embodiment includes a compression rate determining unit (175) that determines a compression rate based on the resolution determined by the resolution determining unit (175), and the compression determined by the compression rate determining unit. And a compression unit (23) for compressing the image using a rate.

  In the present embodiment, the compression rate determination unit determines the compression rate based on the resolution determined by the resolution determination unit. The compression unit compresses the image using the compression rate determined by the compression rate determination unit. The compression rate can be changed to, for example, high, medium, or low by changing the value of the quantization table when converting the image data into the JPEG (Joint Photographic Experts Group) format. When the resolution is low, the compression rate is lowered to suppress the degree of compression, thereby preventing the visibility from being lowered. Further, when the resolution is high, the file size can be reduced without reducing the visibility by increasing the compression rate and increasing the degree of compression.

  In the image processing apparatus (50) of the present embodiment, the compression rate determining unit (175) is based on the resolution determined by the resolution determining unit (175) and the character size determined by the character size determining unit (175). Thus, the compression rate is determined.

  In the present embodiment, the compression rate determination unit determines the compression rate based on the resolution determined by the resolution determination unit and the character size determined by the character size determination unit. For example, it is assumed that the determined character size is large and small for the same resolution. In this case, the compression rate for the larger character size is increased, and the compression rate for the smaller character size is decreased. Thereby, the file size can be reduced without reducing the visibility. Further, it is assumed that the determined resolution has a high and low resolution for the same character size. In this case, the compression rate of the higher resolution is increased, and the compression rate of the lower resolution is decreased. Thereby, the file size can be reduced without reducing the visibility.

  An image processing system (100, 110, 120) according to the present embodiment includes an image processing device (50) according to any one of the aforementioned inventions, and an image input device (1) that inputs an image to the image processing device. And an image output device (2, 3) for outputting an image converted by the image processing device.

  In the present embodiment, an image processing system including an image processing apparatus can be provided.

  A recording medium according to the present invention records a computer program according to the above-described invention.

  In the present embodiment, a recording medium on which a computer program is recorded can be provided.

DESCRIPTION OF SYMBOLS 1 Color image input device 2 Color image output device 3 Transmission device 4 Operation panel 50 Color image processing device 11 A / D conversion part 12 Shading correction part 13 Document type discrimination | determination part 14 Input gradation correction part 15 Area | region separation process part 16 Color correction Section 17 Processing parameter determination section 171 Edge detection section 172 Character rectangle extraction section 173 Character string extraction section 174 Resolution determination section 175 Resolution / compression ratio determination section 18 Black color generation and under color removal section 19 Spatial filter processing section 20 Resolution conversion processing section 21 Output Gradation correction unit 22 Gradation reproduction processing unit 23 Compression processing unit

Claims (9)

A character generation unit that generates character specifying information for specifying character candidates included in the image based on pixel values of a plurality of pixels constituting the image;
A character string generating unit that generates character string specifying information for specifying character string candidates composed of a plurality of characters based on the character specifying information generated by the character generating unit;
A resolution determining unit that determines the resolution of the image based on the character string specifying information generated by the character string generating unit;
An image processing apparatus comprising: a conversion unit that converts the resolution of the image using the resolution determined by the resolution determination unit. A determination unit that determines whether the character string generation information can be generated by the character string generation unit;
The resolution determination unit
The resolution of the image is determined based on the character identification information generated by the character generation unit when the determination unit determines that the generation of the character string identification information is negative. The image processing apparatus according to 1. A character size determining unit that determines one or more character sizes based on the generated character string specifying information or character specifying information;
The resolution determination unit
The image processing apparatus according to claim 1, wherein the resolution is determined based on the character size determined by the character size determination unit. A setting unit is provided for setting whether to give priority to image quality or image size.
The character size determining unit
If the setting unit is set to prioritize image quality, the minimum character size is determined among one or more character sizes based on the character string specifying information or the character specifying information,
When the setting unit is set to prioritize the image size, the character frequency of the maximum frequency is determined among one or a plurality of character sizes based on the character string specifying information or the character specifying information. The image processing apparatus according to claim 3. A compression rate determination unit that determines a compression rate based on the resolution determined by the resolution determination unit;
The image processing apparatus according to claim 3, further comprising: a compression unit that compresses the image using the compression rate determined by the compression rate determination unit. The compression rate determining unit
6. The image processing apparatus according to claim 5, wherein a compression rate is determined based on the resolution determined by the resolution determination unit and the character size determined by the character size determination unit.   An image processing apparatus according to any one of claims 1 to 6, an image input apparatus that inputs an image to the image processing apparatus, and an image output apparatus that outputs an image converted by the image processing apparatus. An image processing system comprising: In a computer program for converting the resolution of an image,
Computer
Character generating means for generating character specifying information for specifying character candidates included in the image based on pixel values of a plurality of pixels constituting the image;
A character string generating means for generating character string specifying information for specifying a character string candidate composed of a plurality of characters based on the generated character specifying information;
Resolution determining means for determining the resolution of the image based on the generated character string specifying information;
A computer program that functions as conversion means for converting the resolution of the image using the determined resolution.   A computer-readable recording medium on which the computer program according to claim 8 is recorded.

Images