JP5243578B2 - Image processing apparatus, image processing method, and image processing system - Google Patents

Description

The present invention relates to an image processing equipment check scanner device or the like capable of reading a document, such as small stamps and bills, an image processing method and an image processing system.

  Conventionally, as a binarization method of a multi-valued image, a binarized slice level (hereinafter simply referred to as “slice level”) for binarizing image data is changed in accordance with the entire multi-valued image and local density. A binarization method has been used (see, for example, Patent Document 1 and Patent Document 2). Here, a method of changing the slice level in accordance with the entire image and the local density will be described with reference to FIG.

  FIG. 8 is a graph in which the density change for each pixel of the image scanned by the check is graphed in the “main scanning direction−density” relationship.

  When changing the slice level according to the entire image and local density, as shown in FIG. 8, the slice level is calculated based on the slice level (floating slice level) calculated according to the local density and the density distribution of the entire image. The slice level (fixed slice level) is used. As can be seen from FIG. 8, when binarization is performed at the floating slice level, the slice level is such that not only the character portion but also the document surface information of the background portion remains.

JP-A-9-233326 Japanese Patent Laid-Open No. 11-4347

  However, in a manuscript such as a check or bill, the background of the paper surface may be covered with a thin pattern or letters as shown in FIG. For this reason, when binarization is performed at the floating slice level, a thin pattern or characters that could not be visually recognized as a background color emerge as shown in FIG. 10A, resulting in not only poor image readability. The file size saved after compression will also increase. As a result, as shown in FIG. 9, binarization by a fixed slice level is more suitable for a document such as a check whose paper background is covered with a light pattern or characters (FIG. 10B). ).

  On the other hand, when the floating slice level is binarized only with the fixed slice level without calculation, as shown in FIG. 11A, when the density of the image is locally different, the density change cannot be handled. The readability of the result becomes worse (FIG. 11 (b)). Thus, it becomes a problem when reading a wide variety of documents.

The present invention has been made in view of the above problems, improves character readability in an output result of image reading processing, prevents a reduction in the compression rate of read image data, and is provided in a host computer or an image reading apparatus main body. image processing equipment efficiency can be enhanced using such storage medium, and an object thereof is to provide an image processing method and an image processing system.

In order to achieve the above object, an image processing apparatus of the present invention includes an input unit that inputs character data and image data including noise information other than the character, and a first two that binarizes the input image data. As binarization processing different from the binarization means and the first binarization means, the second binarization means for binarizing the input image data, and the first binarization means Compression means for compressing each of the first image data binarized by the second image data and the second image data binarized by the second binarization means, and the first before and after compression by the compression means The first compressed image data obtained by compressing the first image data and the second compressed image data obtained by compressing the second image data based on the compression ratio obtained from the image data and the second image data. Control hand that stores or outputs either With the door, said control means is characterized by a higher compression ratio memorize or output by the first compressed image data and the second of said compressing means of the compressed image data.

In order to achieve the above object, an image processing method according to the present invention includes an input step for inputting image data including characters and noise information other than the characters, and a first two method for binarizing the input image data. As a binarization process different from the binarization process and the binarization process in the first binarization process, a second binarization process that binarizes the input image data; A compression step of compressing each of the first image data binarized by the binarization step 1 and the second image data binarized by the second binarization step, and the compression step Based on the compression ratio obtained from the first image data and the second image data before and after compression, the first compressed image data and the second image data obtained by compressing the first image data are compressed. one of the second compressed image data memorize or Characterized in that it comprises a force control process.

  According to the present invention, even when a document that is often covered with a thin pattern or thin characters, such as a handwritten document, is read, the character readability for image data is improved and the image data is compressed. It is possible to prevent the rate from decreasing and increase the usage efficiency of the storage medium or the like.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a cross-sectional view illustrating an outline of an overall configuration of an image processing apparatus according to a first embodiment of the present invention as viewed from above. It is a figure which shows schematic structure of the principal part in the image processing apparatus 100 of FIG. FIG. 3 is a block diagram illustrating a schematic configuration of an image processing unit 212 in FIG. 2. 3 is a flowchart of processing for executing a binarization processing selection method in the image processing apparatus of FIG. 1. It is a flowchart of the process which performs the selection method of the binarization process of the image processing apparatus which concerns on the 2nd Embodiment of this invention. It is a figure which shows the optimal slice level calculated | required from the histogram in the image processing apparatus which concerns on the 3rd Embodiment of this invention. It is a figure for demonstrating the method of calculating the slice value between blocks. It is the figure which plotted the density change of a character and a background part. It is the figure which graphed the density | concentration change for every pixel of the image which scanned the check by the relationship of "main scanning direction-density". It is a figure which shows an example of the gray scale image of the original document covered with the background pattern, the character, etc. FIG. 10 is a diagram illustrating an example in which the gray scale image of FIG. 9 is binarized, where (a) is an example binarized at a floating slice level, and (b) is an example binarized at a fixed slice level. (A) is a figure which shows the gray scale image from which density differs locally, (b) is a figure which shows the example which binarized the gray scale image of FIG. 11 (a) at the fixed slice level.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a cross-sectional view schematically showing an overall configuration of the image processing apparatus according to the first embodiment of the present invention as viewed from above.

  In FIG. 1, an image processing apparatus 100 reads a bill, a check, or the like (hereinafter collectively referred to as “check 50”). The image processing apparatus 100 reads the MICR character printed on the check 50 and the double-sided image image by conveying the check 50 in the character string direction of the MICR character. Then, the image processing apparatus 100 transmits the read MICR character data and double-sided image data to a host computer (not shown) connected by an interface cable.

  The image processing apparatus 100 includes a hopper 1 for loading a check 50 and feeding the check 50 into the apparatus. In the hopper 1, a paper feed roller 2 for feeding the check 50 into the apparatus, a pressing plate 3 for biasing the check 50 against the paper feed roller 2, and the check 50 are stacked on the hopper 1. A load detection sensor 4 for detecting this is disposed.

  The check 50 fed by the paper feed roller 2 is sent to the paper transport path 5 by the feed roller 6 and separated one by one by the separation roller 7. Thereafter, the check 50 is conveyed to the position of the magnetic head 9 as magnetic ink character reading means for reading MICR characters by the conveying roller pair 8a, 8b.

  The portion of the check 50 on which the MICR characters are printed is brought into close contact with the magnetic head 9 by the pressing roller mechanism 10. On the upstream side of the magnetic head 9 in the paper transport path 5, a permanent magnet 11 for aligning the magnetization direction of the MICR characters printed on the check 50 and a paper leading edge detection sensor 12 for taking the timing for reading the MICR characters are arranged. ing. In addition, a paper trailing edge detection sensor 13 is disposed on the downstream side of the magnetic head 9 in the paper conveyance path 5.

  The check 50 that has passed the magnetic head 9 passes through the pair of transport rollers 8d and is transported by the transport roller pair 8c to the position of the image reading sensor 17 (image sensor 202) as image reading means. The image reading sensor 17 includes an LED, which is a light emitting element of three colors, a condenser lens, and a light receiving element, and can read the image images on both sides of the check 50 in both color / monochrome. The conveying roller pair 8c is provided with an electromagnetic clutch 14 that controls the rotation thereof. In addition, movable guides 16 and 16a constituting a part of the sheet conveyance path 5 are rotatably provided on the upstream side of the conveyance roller pair 8c.

  On the upstream side of the image reading sensor 17 in the paper conveyance path 5, a registration sensor 18 is provided for taking an image reading timing. On the other hand, on the downstream side of the image reading sensor 17 in the paper conveyance path 5, an endorser 19 for endorsing and printing the date, the name of the financial institution in which the check 50 was brought, etc. on the check 50 from which the image was read, A rear end detection sensor 20 that detects that the rear end of the check 50 has passed through the endorser 19 is disposed.

  The check 50 that has been read is discharged to the discharge tray 21 a or the discharge tray 21 b by the discharge roller pair 22. On the downstream side of the paper discharge roller pair 22 is a flapper 23 for switching the paper conveyance path so as to distribute the check 50 to be discharged to the paper discharge trays 21a and 21b according to the MICR character data read by the magnetic head 9. Is provided.

  The image processing apparatus 100 includes a drive motor 24 that drives a group of conveyance rollers, a controller unit 25 that controls the entire image processing apparatus 100 including these components, and communicates with a computer, and an operation switch panel 26. ing. On the operation switch panel 26, the operator can select an image reading resolution of the check 50 and send an instruction to start reading the check 50 to the controller unit 25.

  The sheet conveyance path distance between the conveyance roller pair 8b and the conveyance roller pair 8c is set to be shorter than the length in the conveyance direction of the check 50 to be conveyed.

  Next, image processing performed by the image processing apparatus 100 in FIG. 1 will be described with reference to FIGS. 2 and 3.

  FIG. 2 is a diagram showing a schematic configuration of a main part in the image processing apparatus 100 of FIG.

  In FIG. 2, the image processing apparatus 100 includes an image reading unit 200 corresponding to the image reading sensor 17 and a control unit 210 corresponding to the controller unit 25. The image reading unit 200 includes a lens 201 and an image sensor 202. The control unit 210 includes an image memory 211, an image processing unit 212, and a CPU 213. The control unit 210 can communicate with an external host computer 220.

  An image read by the image reading sensor 17 is converted into a grayscale image of 256 gradations (8 bits) by an A / D converter (not shown) and stored in the image memory 211. In response to an image read instruction from the host computer 220, the CPU 213 sequentially reads out images from the image memory 211 by a DMA transfer unit (not shown) and transfers them to the image processing unit 212.

  FIG. 3 is a block diagram illustrating a schematic configuration of the image processing unit 212 of FIG.

  In FIG. 3, the image processing unit 212 includes a binarization selection unit 301, a binarization processing unit A302, a binarization processing unit B303, a compression unit 304, a compression rate determination unit 305, and a transfer unit 306. Is provided.

  In the image processing unit 212, either the binarization processing unit A 302 or the binarization processing unit B 303 is selected by the binarization selection unit 301, and the read image is transferred to the selected binarization processing unit 2. It is priced. The binarized image data (binary image data) is compressed by the compression unit 304 and transferred to either the compression rate determination unit 305 or the transfer unit 306.

  When transferred to the compression rate determination unit 305, first, a compression rate when the binary image data is compressed by the compression unit 304 is calculated, and binarization is performed by the binarization selection unit 301 based on the calculated compression rate. Either the processing unit A302 or the binarization processing unit B303 is selected. Thereafter, the image data is read again from the image memory 211 and transferred to the transfer unit 306 via the binarization processing unit and the compression unit. The compressed image data transferred to the transfer unit 306 is transmitted to the host computer 220 by general-purpose transmission means (not shown). The CPU 213 controls all these operations.

  A binarization selection method according to the present embodiment will be described with reference to the flowcharts of FIGS.

  FIG. 4 is a flowchart of processing for executing the binarization processing selection method in the image processing apparatus of FIG.

  In FIG. 4, the CPU 213 performs an image reading (scanning) operation in accordance with an instruction from the host computer 220, and stores grayscale image data in the image memory 211 (step S201).

  Next, the CPU 213 selects the compression rate determination unit 305 as the output destination of the image data from the compression unit 304 (step S202), and instructs the binarization selection unit 301 to select the binarization processing unit A302 (step S202). S203). Next, the CPU 213 starts reading gray scale image data from the image memory 211. The read grayscale image data is converted into binary (1-bit) image data by the binarization processing unit A302 and transferred to the compression unit 304 (step S204).

  The compression unit 304 compresses the transferred binary image data, and transfers the compressed binary image data (compressed image data) to the compression rate determination unit 305 (step S205). At this time, the compressed image data is not transferred to the transfer unit 306.

  The compression rate determination unit 305 counts the size (number of bytes) of the compressed image data sent from the compression unit 304. Then, the compression rate determination unit 305 or the CPU 213 calculates a compression rate defined as a value obtained by dividing the size (number of bytes) of the binary image data before compression by the result of counting by the compression rate determination unit 305, and the work rate is not shown. Store as “CompRateA” in the memory (step S206). It is assumed that the work memory is constituted by a storage device such as a RAM installed in the control unit 210, for example.

  Next, the CPU 213 instructs the binarization selection unit 301 to select the binarization processing unit B303 (step S207). Next, the CPU 213 starts reading gray scale image data from the image memory 211. The read grayscale image data is converted into binary (1-bit) image data by the binarization processing unit B303 and transferred to the compression unit 304 (step S208).

  The compression unit 304 compresses the transferred binary image data, and transfers the compressed binary image data (compressed image data) to the compression rate determination unit 305 (step S209). At this time, the compressed image data is not transferred to the transfer unit 306.

  The compression rate determination unit 305 counts the size (number of bytes) of the compressed image data sent from the compression unit 304. Then, the compression rate determination unit 305 or the CPU 213 calculates the compression rate by dividing the size (number of bytes) of the binary image data before compression by the result of the compression rate determination unit 305 counting, and stores “CompRateB” in the work memory. (Step S210).

  Next, the compression rate determination unit 305 or the CPU 213 compares the value of “CompRateA” stored in the work memory with the value of “CompRateB” (step S211), and the binarization selection unit 301 determines whether or not the comparison results. The binarization processing unit is selected by commanding (steps S212 and S213).

i) When CompRateA> CompRateB (YES in step S211)
The CPU 213 or the compression rate determination unit 305 instructs the binarization selection unit 301 to select the binarization processing unit A302 (step S212). Next, the CPU 213 selects the transfer unit 306 as the output destination from the compression unit 304 (step S214), and starts reading grayscale image data from the image memory 211. The read grayscale image data is converted into binary (1-bit) image data by the binarization processing unit A302 and transferred to the compression unit 304 (step S215). The compression unit 304 compresses the transferred binary image data into compressed image data, transfers the compressed image data to the transfer unit 306 (step S216), and transmits the compressed image data to the host computer 220 by general-purpose communication means (not shown). .

ii) When CompRateA ≦ CompRateB (NO in step S211)
The CPU 213 or the compression rate determination unit 305 instructs the binarization selection unit 301 to select the binarization processing unit B303 (step S213). Next, the CPU 213 selects the transfer unit 306 as the output destination from the compression unit 304 (step S214), and starts reading gray scale image data from the image memory 211. The read grayscale image data is converted into binary (1-bit) image data by the binarization processing unit B303 and transferred to the compression unit 304 (step S215). The compression unit 304 compresses the transferred binary image data into compressed image data, transfers the compressed image data to the transfer unit 306 (step S216), and transmits it to the host computer 220 by general-purpose communication means (not shown).

  According to the first embodiment, the image data binarized by the binarization processing unit A302 and the image data binarized by the binarization processing unit B303 are respectively compressed by the compression unit 304. Two compression rates obtained from the compressed image data and binary image data before compression are compared with each other. In accordance with the comparison result, binarization is performed by either the binarization processing unit A302 or the binarization processing unit B303, and the image data compressed by the compression unit 304 is output. Further, when a storage device such as a hard disk is built in or connected to the image reading device, the compressed image data may be stored in the storage device. This improves character readability in the output result of the image reading process, even when reading a document that is often covered with thin patterns or thin characters, such as handwritten documents, and improves the readability of the scanned image data. The use efficiency of a storage medium or the like provided in the host computer or the image reading apparatus main body can be improved by preventing a reduction in the compression rate.

[Second Embodiment]
The image processing apparatus according to the second embodiment of the present invention has the same configuration (FIGS. 1 to 3) as the image processing apparatus 100 according to the first embodiment, and a description thereof will be omitted. Only differences from the first embodiment will be described below.

  In the second embodiment, a binarization method is adopted in which the binarization processing unit A302 has a higher character reading quality and a lower compression rate than the binarization processing unit B303.

  A binarization processing selection method according to the second embodiment will be described with reference to the flowcharts of FIGS.

  FIG. 5 is a flowchart of processing for executing the binarization processing selection method of the image processing apparatus according to the second embodiment of the present invention.

  In FIG. 5, the CPU 213 performs an image reading (scanning) operation according to an instruction from the host computer 220, and stores grayscale image data in the image memory 211 (step S301). Subsequently, the CPU 213 selects the compression rate determination unit 305 as the output destination of the image data from the compression unit 304 (step S302).

  Next, the CPU 213 instructs the binarization selection unit 301 to select the binarization processing unit A302 (step S303), and starts reading grayscale image data from the image memory 211 (step S304). The read image data is converted into binary (1-bit) image data by the binarization processing unit A302 and transferred to the compression unit 304 (step S304).

  The compression unit 304 compresses the transferred binary image data, and transfers the compressed binary image data (compressed image data) to the compression rate determination unit 305 (step S305). At this time, the compressed image data is not transferred to the transfer unit 306.

  The compression rate determination unit 305 counts the size (number of bytes) of the compressed image data sent from the compression unit 304. Then, the CPU 213 or the compression rate determination unit 305 calculates the compression rate by dividing the size (number of bytes) of the binary image data before compression by the result of counting by the compression rate determination unit 305, and stores “ CompRateA "is stored (step S306). Then, the value of “CompRateA” stored in the work memory is compared with a predetermined value set in advance (step S307), and the binarization selection unit 301 is instructed according to the comparison result to perform binarization processing. Parts are selected (steps S308 and S309). The predetermined value compared in step S307 is set in advance. In step S307, instead of the predetermined value, the value obtained by a predetermined method from the input image data may be compared with the value of “CompRateA”. For example, in the case of an image with a large variance, the value to be compared may be reduced.

i) When CompRateA> predetermined value (YES in step S307)
The CPU 213 or the compression rate determination unit 305 instructs the binarization selection unit 301 to select the binarization processing unit A302 (step S308). Next, the CPU 213 selects the transfer unit 306 as an output destination from the compression unit 304 (step S310), and starts reading grayscale image data from the image memory 211. The read grayscale image data is converted into binary (1-bit) image data by the binarization processing unit A302 and transferred to the compression unit 304 (step S311). The compression unit 304 compresses the transferred binary image data into compressed image data, transfers the compressed image data to the transfer unit 306 (step S312), and transmits the compressed image data to the host computer 220 by general-purpose communication means (not shown). .

ii) When CompRateA ≦ predetermined value (NO in step S307)
The CPU 213 or the compression rate determination unit 305 instructs the binarization selection unit 301 to select the binarization processing unit B303 (step S309). Next, the CPU 213 selects the transfer unit 306 as an output destination from the compression unit 304 (step S310), and starts reading grayscale image data from the image memory 211. The read grayscale image data is converted into binary (1-bit) image data by the binarization processing unit B303 and transferred to the compression unit 304 (step S311). The compression unit 304 compresses the transferred binary image data into compressed image data, transfers the compressed image data to the transfer unit 306 (step S312), and transmits the compressed image data to the host computer 220 by general-purpose communication means (not shown). .

  According to the second embodiment, the image data binarized by the binarization processing unit A302 is compressed by the compression unit 304, and the compression rate obtained from the compressed image data and the binary image data before compression is calculated. Compared with a predetermined value set in advance. In accordance with the comparison result, either the binarization processing unit A302 or the binarization processing unit B303 binarizes the image data, and outputs or stores the image data compressed by the compression unit 304. This improves character readability in the output result of the image reading process, even when reading a document that is often covered with thin patterns or thin characters, such as handwritten documents, and improves the readability of the scanned image data. The use efficiency of a storage medium or the like provided in the host computer or the image reading apparatus main body can be improved by preventing a reduction in the compression rate. In addition, the image processing itself can be made efficient.

[Third Embodiment]
The image processing apparatus according to the third embodiment of the present invention has the same configuration (FIGS. 1 to 3) as the image processing apparatus 100 according to the first embodiment, and a description thereof will be omitted. Only differences from the first embodiment will be described below.

  In the third embodiment, a binarization processing method when the binarization processing unit A302 in the first and second embodiments is a binarization processing unit using a floating slice value will be described. I do.

  FIG. 6 is a diagram showing the optimum slice level obtained from the histogram in the image processing apparatus according to the third embodiment of the present invention.

  First, the binarization processing unit A302 using the floating slice value will be described.

  In the binarization processing unit A302, the above-described grayscale image data is divided into blocks of 64 pixels × 64 pixels, and a histogram is created in each block. Then, the slice value of each block can be calculated from the created histogram by the discriminant analysis method.

  Binarization by discriminant analysis means that the histogram of an image is divided into two as shown in FIG. 6, and when the divided histograms are called class 1 and class 2, the inter-class variance value between class 1 and class 2 Is determined to be the maximum, and the determined division position (density) is used as the slice value. The slice value obtained by the discriminant analysis method from the histogram in the block is set as the slice value at the center of the block.

  FIG. 7 is a diagram for explaining a method of calculating a slice value between blocks.

  A, B, C, and D in FIG. 7 represent slice values obtained from the histogram in each block as the slice value of the point at the center position of the 64 × 64 pixel block. Interpolated slice values at all pixels surrounded by the four center points are obtained by the following equation [Equation 1].

[Formula 1]
A, B, C, and D are slice values obtained from the histogram of each block.
Slice value between AB: SAB (X) = (B−A) X / 64 + A
Slice value between CDs: SCD (X) = (D−C) X / 64 + C
Slice value at point (X, Y) in ABCD: S (X, Y) = {SCD (X) −SAB (X)} Y / 64 + SAB (X)
The block division, the calculation of the slice value at the center of the block, and the calculation of the slice value of all the pixels between the blocks are performed on the entire image, and the binarization processing of the grayscale image data is performed using the obtained slice value. If this method is used, an optimum slice value can be selected even for an image having partially different shades such as a blurred character, and the output result of the image reading process is also highly readable.

  However, when the binary image data obtained by the binarization processing method using such a floating slice value is compressed, if the paper surface is covered with a thin pattern or thin characters, the binary image data is converted into binary image data. The compression rate decreases because thin patterns and thin letters emerge. Readability also deteriorates. Therefore, in such a case, the binarization processing unit B303 using a single slice value is used.

  Next, the binarization processing unit B303 will be described.

  In the binarization processing unit B303, first, a histogram of the entire image is created. A single slice value can be calculated from the created histogram by discriminant analysis. Then, binarization processing is performed on the entire image of the grayscale image data using the obtained slice value. By using this method, it is possible to output an image with low noise and high readability with respect to an image with a lot of noise on the background. However, the readability of faint characters and the like decreases. The compression rate when compressing binary image data generated by this method is relatively high.

  The above-described binarization processing unit A302 and binarization processing unit B303 are employed as the binarization processing unit of the first embodiment or the second embodiment. As a result, when reading a document whose paper surface is covered with a thin pattern or thin characters, the binarization processing unit A302 using a floating slice value that reduces the compression rate when compressing binary image data is selected. First, the binarization processing unit B303 is selected. For this reason, although the readability of the blurred characters is somewhat deteriorated, it is possible to avoid the characters becoming illegible due to the pattern. Further, it is possible to avoid the compressed image data from becoming unnecessarily large.

  On the other hand, when reading a document that is not covered with a thin pattern or thin characters on the paper surface, since the compression rate is difficult to decrease, the binarization processing unit A302 using the floating slice value is selected, and the readability of the blurred characters is selected. Will improve. In addition, the size of the compressed image data is also appropriate.

  The binarization processing method of the binarization processing unit A302 in the present embodiment is performed by the discriminant analysis method, but is not limited to this, and the binarization processing method using various floating slice value determination methods is used. It may be changed. Further, the binarization processing method of the binarization processing unit B303 is performed by the discriminant analysis method, but is not limited to this, and even if it is changed to the binarization processing method using various slice value determination methods. Good.

  When comparing compression ratios in the first or third embodiments, a predetermined coefficient in "CompRateB", a coefficient set by the user, or a coefficient obtained by a predetermined method based on input image data You may multiply.

  When the binarization processing unit A employs a binarization method in which the character reading quality is higher than the binarization processing unit B but the compression rate tends to be low, “CompRateB” is applied. More preferably, the predetermined coefficient is smaller than 1.

  In each of the above embodiments, part or all of the functions of the image processing unit 212 may be realized by a program executed by the CPU 213 or another CPU.

  In the processing method described with reference to FIGS. 4 and 5, the binarization process and the compression process are performed twice or three times, but the compressed image data or the binarized image data generated when calculating the compression rate is performed. May be stored in a memory or the like, and binarization processing (step S215, step S311) may be omitted, or both binarization processing and compression processing (step S216, step S312) may be omitted. . Thereby, a series of processes can be performed efficiently.

  Further, two or more binarization processing units may be provided instead of two. Thereby, it is possible to cope with a plurality of types of binarization processing. Further, it may be configured to perform processing while switching processing functions by one binarization processing unit. Thereby, the number of parts can be reduced and cost reduction can be aimed at. Further, the binarization processing unit may be realized in the form of a program executed by the CPU. Thereby, the weight reduction and efficiency improvement of an apparatus can be achieved.

  Needless to say, the image processing apparatus 100 may be configured to include or connect a hard disk device, a removable disk device, or the like without using the host computer 220. The present invention also includes a case where binary data or image data transferred from an image reading apparatus is compressed by a host computer or binarized and compressed.

  The object of the present invention is also achieved by supplying a storage medium storing software program codes for realizing the functions of the above-described embodiments to a system or apparatus. In that case, the computer (or CPU, MPU, etc.) of the system or apparatus reads and executes the program code stored in the storage medium. The program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code and the storage medium storing the program code constitute the present invention.

  As a storage medium for supplying the program code, for example, a removable disk, a hard disk, a magneto-optical disk, or the like can be used. Further, optical disks such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD ± R, DVD-RAM, DVD ± RW, magnetic tape, nonvolatile memory card, ROM, and the like can be used. The program code may be downloaded via a network.

  The functions of the above-described embodiments are not only realized by executing the program code read by the computer. That is, when the OS (operating system) running on the computer performs part or all of the actual processing based on the instruction of the program code, and the functions of the above-described embodiments are realized by the processing. Is also included.

  Further, the program code read from the storage medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer. When the CPU of the function expansion board or function expansion unit performs part or all of the actual processing based on the instruction of the program code, and the functions of the above-described embodiments are realized by the processing. Is also included.

  When reading a document covered with thin patterns and characters, such as a handwritten document, not only is the character readability of the output result high, but the compression rate is also high, so the use efficiency of storage media such as disks can be increased. Therefore, industrial applicability is high.

DESCRIPTION OF SYMBOLS 100 Image processing apparatus 200 Image reading part 210 Control part 211 Image memory 212 Image processing part 213 CPU
220 Host computer 301 Binary selection unit 302 Binary processing unit A
303 Binary processing unit B
304 Compression unit 305 Compression rate determination unit 306 Transfer unit

Claims (12)

Input means for inputting image data including characters and noise information other than the characters;
First binarizing means for binarizing the input image data;
As a binarization process different from the first binarization means, a second binarization means for binarizing the input image data;
Compression means for compressing each of the first image data binarized by the first binarization means and the second image data binarized by the second binarization means;
First compressed image data and second image obtained by compressing the first image data based on a compression rate obtained from the first image data and the second image data before and after compression by the compression unit. Control means for storing or outputting any one of the second compressed image data obtained by compressing the data,
It said control means, the image processing device characterized by a higher compression ratio by the compression means memorize or outputs of the first compressed image data and the second compressed image data. The control means stores or outputs the first compressed image data when the compression rate of the first compressed image data is higher than a predetermined value or a value obtained by a predetermined method based on the input image data. And
When the compression rate of the first compressed image data is lower than a predetermined value or a value obtained by a predetermined method based on the input image data, the second compressed image data is stored or output. the image processing apparatus according to claim 1 Symbol mounting to. The first binarization means binarizes input image data by locally changing a slice level for binarizing image data,
It said second binarization means, the image processing apparatus according to claim 1 or 2, characterized in that binarizing the input image data in a single slice level. The control means compares the compression rate of the first compressed image data with a value obtained by multiplying the compression rate of the second compressed image data by a coefficient, and the compression rate of the first compressed image data is large. When the first compressed image data is stored or output, and when the value obtained by multiplying the compression rate of the second compressed image data by a coefficient is large, the second compressed image data is stored or output. The image processing apparatus according to claim 3 .   The image processing apparatus according to claim 1, wherein the input unit includes a reading unit that reads an image of a document. An input step of inputting image data including characters and noise information other than the characters;
A first binarization step for binarizing the input image data;
As a binarization process different from the binarization process in the first binarization process, a second binarization process for binarizing the input image data;
A compression step of compressing each of the first image data binarized by the first binarization step and the second image data binarized by the second binarization step;
The first compressed image data and the second image obtained by compressing the first image data based on the compression ratio obtained from the first image data and the second image data before and after the compression in the compression step. an image processing method characterized in that it comprises a one control step of one remembers or output of the second compressed image data by compressing the data.   7. The image processing method according to claim 6, wherein the control step stores or outputs one of the first compressed image data and the second compressed image data which has a higher compression rate in the compression step. The control step stores or outputs the first compressed image data when the compression ratio of the first compressed image data is higher than a predetermined value or a value obtained by a predetermined method based on the input image data. And
When the compression rate of the first compressed image data is lower than a predetermined value or a value obtained by a predetermined method based on the input image data, the second compressed image data is stored or output. The image processing method according to claim 6 or 7 . In the first binarization step, the slice level for binarizing the image data is locally changed to binarize the input image data,
Said second binarization process, the image processing method according to any one of claims 6 to 8, characterized in that binarizing the input image data in a single slice level. The control step compares the compression rate of the first compressed image data with a value obtained by multiplying the compression rate of the second compressed image data by a coefficient, and the compression rate of the first compressed image data is large. When the first compressed image data is stored or output, and when the value obtained by multiplying the compression rate of the second compressed image data by a coefficient is large, the second compressed image data is stored or output. The image processing method according to claim 9 . Wherein the input step, the image processing method according to any one of claims 6 to 10, characterized in that it comprises a reading step of reading an image of an original. Image processing system characterized in that it is constituted by an information processing apparatus that receives data output from the image processing apparatus and the image processing apparatus according to any one of claims 1 to 5.