JP4383157B2 - Image processing apparatus, program, and storage medium - Google Patents

Description

  The present invention relates to an image processing apparatus, a program, and a storage medium.

  Patent Document 1 discloses a technique for removing a character portion of an image and creating a background image filled with surrounding colors.

  Patent Document 2 includes an encoding technique that includes first means for extracting and encoding character information from a color image, and second means for encoding at a reduced resolution of image data, and synthesizes the respective data. Proposed.

  Japanese Patent Application Laid-Open No. H10-228667 proposes a technique for registering text data extracted from image data as search information and storing the image data at a low resolution.

  Patent Document 4 proposes a technique for dividing a color document image in which a natural image and characters are mixed into a character background block and an image data block, performing encoding suitable for each region, and setting the resolution of the pattern region low. ing.

Japanese Patent No. 2611012 JP-A-8-186711 JP 2001-297104 A Japanese Patent No. 3108133

  For image data of an image including a character image, the character image is separated, processed by image processing and compression processing suitable for the separated character image and background image, and then the character image is superimposed from above. It is possible to improve the visibility of the character image and the visibility of the background image by saving in a file format having a multi-layer function.

  On the other hand, the technique disclosed in Patent Document 1 discloses a technique for removing a character portion of an image and creating a background image filled with surrounding colors.

  However, the technique disclosed in Patent Document 1 has a problem in that it cannot reduce the amount of data and enable high-speed processing because the character portion of the image is removed and embedded with surrounding colors.

  An object of the present invention is to reduce the amount of data and enable high-speed processing when displaying a character image superimposed on a background image.

  According to the first aspect of the present invention, an extracting means for extracting a character part of an image from image data, a character image that is the extracted image, and a region in which the character part of the extracted image is extracted are filled with a predetermined image. An image processing apparatus comprising: a file creation unit that separately saves a background image as it is as one or a plurality of files.

  According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, the extraction unit performs the extraction at a predetermined sampling interval when the image data is a color image.

  According to a third aspect of the present invention, in the image processing apparatus according to the first or second aspect, the file creation unit is configured to convert the character image into one or a plurality of binary images when the image data is a color image. To the file.

  According to a fourth aspect of the present invention, in the image processing apparatus according to the third aspect, the extracting means extracts the character part after binarizing the image data, and a color image corresponding to the character part The character color is specified by extracting the color.

  According to a fifth aspect of the present invention, in the image processing device according to any one of the first to fourth aspects, the file creation means is a color obtained by reducing the color of the character image when the image data is a color image. The character image is stored in the file as an image.

  According to a sixth aspect of the present invention, in the image processing apparatus according to any one of the first to fifth aspects, the file creation unit performs reversible compression on the extracted character image, and the background image after the extraction is In lossy compression, each is compressed and encoded and then stored as the one or more files.

  According to a seventh aspect of the present invention, in the image processing apparatus according to the sixth aspect, the file creation means smoothes the extracted background image before the lossy compression.

  According to an eighth aspect of the present invention, in the image processing apparatus according to the sixth or seventh aspect, the file creation means performs an OCR process on the extracted character image, and the result of the OCR is converted into a text of transparent text. The file is saved as a layer-managed file with the data, the extracted character image, and the extracted background image as layers, and the text data layer is higher than the character image layer. A layer.

  According to a ninth aspect of the present invention, there is provided a character image that is an image obtained by extracting a character portion of an image from image data and a background image that is not filled with a predetermined image in a region where the character portion of the extracted image is extracted. Image processing means comprising image creating means for creating an image in which the character image is superimposed with an opaque color on an area where the character portion of the background image is extracted for one or a plurality of files stored separately Device.

  According to the tenth aspect of the present invention, an extraction process for extracting a character portion of an image from image data, and a character image that is the extracted image and a region in which the character portion of the extracted image is extracted are filled with a predetermined image. This is a computer-readable program that causes a computer to execute a file creation process for separately storing a background image as it is as one or more files.

  According to the eleventh aspect of the present invention, there is provided a character image that is an image obtained by extracting a character portion of an image from image data, and a background image that is not filled with a predetermined image in a region where the character portion of the extracted image is extracted. A computer that causes a computer to execute an image creation process for creating an image in which the character image is superimposed with an opaque color on an area where the character portion of the background image is extracted for one or a plurality of files stored separately Is a readable program.

  The invention described in claim 12 is a storage medium in which the program according to claim 10 or 11 is stored.

  According to the first and tenth aspects of the present invention, since the background image does not fill the region after the character image is extracted, an image obtained by superimposing the character image on the region where the character portion of the background image is extracted with an opaque color is used. If created, the above-mentioned file consisting of a character image and a background image can reduce the amount of data, and at the same time, an image that does not impair the readability of characters can be created at high speed. In addition, since the character image and the background image are stored separately as one or a plurality of files, the background image is finally provided to the user even if the entire image including the character image is smoothed. Can be prevented from having any influence of the smoothing process.

  The invention according to claim 2 is the invention according to claim 1, wherein the storage capacity required for processing is reduced, and high-speed processing is possible.

  The invention according to claim 3 is that, in the invention according to claim 1 or 2, it is common that the color of the character portion is generally written in one color or the number of colors not so many. The color of the part is specified by a clustering process or the like, and a character image can be created with one or a plurality of binary images of colors according to the clustering result.

  According to a fourth aspect of the present invention, in the third aspect of the present invention, a character image can be created with one or a plurality of binary images of colors according to the clustering result.

  In the invention according to claim 5, in the invention according to any one of claims 1 to 4, since the number of colors used for the character portion is usually not so much, one 4-bit or 8 It is sufficient to hold only a bit-reduced image or a plurality of color-reduced images, and this has the effect of increasing contrast and visibility for the user.

  The invention according to claim 6 is the invention according to any one of claims 1 to 5, wherein a binary image of characters is a compression format suitable for a character image, and a background image is a compression format suitable for a background. Can be compressed and encoded, and the file size can be reduced. In addition, by compressing a character image using a reversible compression method, it is possible to perform OCR processing without lowering the recognition rate when it is desired to perform character encoding in the future by processing such as OCR.

  According to the seventh aspect of the invention, in the sixth aspect of the invention, the image finally provided to the user can be prevented from having any influence of the smoothing process.

  The invention according to claim 8 can create a file having a multi-layer function capable of text search in the invention according to claim 6 or 7.

  According to the ninth and eleventh aspects of the present invention, the background image is an image obtained by superimposing the character image on the region where the character portion of the background image is extracted without filling in the region after the character image is extracted. Since it can be created, the above-mentioned file consisting of a character image and a background image can reduce the amount of data, and an image that does not impair the readability of characters can be created at high speed.

  An embodiment of the present invention will be described.

  FIG. 1 is a block diagram showing an electrical connection of an image processing apparatus 1 according to the present embodiment. As shown in FIG. 1, the image processing apparatus 1 is a computer such as a PC, which performs various calculations and centrally controls each part of the image processing apparatus 1, and a memory composed of various ROMs and RAMs. 12 are connected by a bus 13.

  The bus 13 is connected to a magnetic storage device 14 such as a hard disk, an input device 15 such as a keyboard and a mouse, a display device 16 and a storage medium reader 18 that reads a storage medium 17 such as an optical disk via a predetermined interface. An image reading device 21 that reads an image is connected, and a predetermined communication interface 19 that communicates with the network 4 is connected. As the storage medium 17, various media such as an optical disk such as a CD and a DVD, a magneto-optical disk, and a flexible disk can be used. As the storage medium reading device 18, specifically, an optical disk device, a magneto-optical disk device, a flexible disk device, or the like is used according to the type of the storage medium 17.

  The image processing apparatus 1 reads the image processing program 20 for executing the program of the present invention from the storage medium 17 for executing the storage medium of the present invention, and installs it in the magnetic storage device 14. These programs may be downloaded and installed via a network 4 such as the Internet. As a result of this installation, the image processing apparatus 1 is in a state in which predetermined processing described later can be executed. Note that the image processing program 20 may operate on a predetermined OS.

  FIG. 2 is a functional block diagram of functions realized by the image processing apparatus 1 based on the image processing program 20, and FIG. 3 is a flowchart of processing executed by the image processing apparatus 1. FIG. 5 is an explanatory diagram of processing executed by the image processing apparatus 1.

  As shown in FIGS. 2, 3, and 5, first, the image of the original is read by the image reading device 21 as a color image of 600 dpi, for example (step S1). The read image data (original image 41 in FIG. 5A) is subjected to the following two systems of processing.

  In the first process, the image data read in step S1 is subjected to resolution conversion by the resolution conversion / compression unit 31 to reduce the data amount (for example, convert to a 300 dpi color image) (step S2). Each character color is specified from the color image by the color reduction unit 33, and binarization for each character color is performed by the binarization unit 34 (step S3), and the character extraction unit 32 performs at least a character region (pixel or area). ) (Extraction means, extraction process) (step S4), a character image 42 which is a binary image is created (FIG. 5B) (step S5), and this is converted by the resolution conversion / compression unit 31. Compression encoding is performed by a lossless compression method such as MMR (step S6), and the data amount is further reduced. Here, although the character is specified, information such as a ruled line may be extracted together with the character in addition to the character. Alternatively, an area that is highly likely to be a character may be extracted as a character area by paying attention to an edge or the like that is a feature of the character or ruled line. This is because the purpose is not necessarily to extract the character region accurately from the image.

  In the second process, the resolution conversion / compression unit 31 performs resolution conversion by an irreversible compression method such as JPEG for the image data read in step S1 (including a character image). The data amount is further reduced (for example, converted to a color image of 150 dpi) (step S7). The original image 41 to be processed in this case is used without extracting a character image 42 in step S3 and filling the region after the character extraction (giving a predetermined pixel value or the like). This is used as the background image 43 (FIG. 5B) (step S8), and this is compression-encoded by the resolution converting / compressing unit 31 (step S9) to further reduce the data amount.

  The character image 42 and the background image 43 created in this way are saved separately as one or a plurality of files (PDF files) (file creation means, file creation processing) (step S10). Specifically, for example, it is conceivable to save as a single file such as PDF or JPEG2000 having a multilayer function. In this case, a background image is arranged in the first layer (lower layer) of the layer-managed file, and a character image is arranged in the second layer (upper layer).

  Processing for displaying the file data thus created on the display device 16 will be described with reference to the functional block diagram of FIG. 2, the flowchart of FIG. 4, and the explanatory diagram of FIG.

  That is, when the image processing device 1 displays the file created in step S10 on the display device 16 (Y in step S11), the resolution conversion of the character image 42 is performed by the resolution conversion / compression unit 31, and the character image is displayed. 42 is a resolution-converted character image 44, and the resolutions of the character image 44 and the background image 43 are made uniform (step S12). Then, the character image 44 of the character corresponding to the character position on the image of the background image 43 is overwritten by being overlaid with an opaque color (image 45 in FIG. 5D) (step S13). If this superimposed image is displayed on the display device 16 (or printed, transmitted to the outside, etc.), an image in which the character image 44 and the background image 43 are overlapped can be viewed as a single image.

  As described above, since the background image 43 does not perform hole filling in the area after the character image 42 is extracted, the amount of data of the file including the character image 42 and the background image 43 can be reduced, and An image that does not impair the readability of characters can be created at high speed.

  In step S3, by extracting an image from the original image 41 at a predetermined sampling interval, it is possible to reduce the storage capacity necessary for processing and perform high-speed processing. That is, in the technique disclosed in Patent Document 1 described above, at least a memory for calculating the background of the character is required to extract the character portion and obtain the background color. This is unnecessary. In other words, in the present embodiment, the character image 44 is displayed so as to overlap with the background image 43 in an opaque color, so it is not necessary to specify the background color of the original image 41 and erase the character portion from the original image 41. The extraction process of the character image 42 is performed at a predetermined interval, the background color extraction process becomes unnecessary, and the processing time and the memory required for it are unnecessary.

  As described above, one or a plurality of binary images are created as the character image 42. This is because the color of the character part is usually written in one color such as black, red, blue, etc. or the number of colors that is not so many, so the color of the character part is specified by clustering processing etc. Then, a plurality of binary images having colors according to the clustering result are created.

  An example of this clustering technique is shown. That is, in steps S3 and S4, the character image is binarized first, and the character portion is extracted from the binary image. Here, a frequency histogram is created in the RGB three-dimensional space with reference to the pixel value (RGB value) of the color original image corresponding to the extracted character portion. Next, clustering is performed to divide the frequency histogram into several classes. As the clustering method, a special method is not used, and a generally known K-Mean method or the like may be used.

  Further, when creating one or a plurality of character images 42 as the character image 42, a color multi-value image may be used instead of a binary image. Since the number of colors used for the character portion is usually not so large, 160,000 colors or the like are unnecessary as in a general color original. In other words, it is sufficient to hold an image that has been reduced to 4 bits or 8 bits, or an image that has been reduced by a plurality of sheets, and this has the effect of increasing contrast and visibility for the user.

  In the above example, the character recognition (OCR) process is performed on the extracted character image 42 (step S5), and the character recognition result is used as the text data of the transparent text, so that the uppermost layer of the layer-managed file is used. It may be arranged in (third layer), a background image in the first layer (lower layer), and a character image in the second layer (upper layer). Thereby, an electronic file having a multi-layer function capable of text search can be created.

  Further, since the character image 42 has been binarized (step S3), the contrast with the background image 43 is high and the visibility is high. Causes moiré due to the characteristics of these scanner devices. As means for solving this, it is generally known that the occurrence of moire can be suppressed by performing a smoothing process on a scanned image.

  However, when smoothing processing is performed on the entire scanned image, there is a problem in that characters are blurred and visibility is deteriorated.

  Accordingly, after the character image 42 is extracted (step S4), the entire background image 43 or a region not recognized as the character image 42 in the background image 43 is selectively smoothed and γ corrected. If image processing such as processing is performed, the visibility other than the character image 42 can be enhanced.

  Furthermore, since the background image 43 is generally composed of an image such as a photograph other than characters, it may not require a high resolution like characters, so even if the image size is reduced by resolution conversion processing. Good.

1 is a block diagram of electrical connection of an image processing apparatus according to an embodiment of the present invention. It is a functional block diagram of an image processing apparatus. It is a flowchart explaining the process which an image processing apparatus performs. It is a flowchart explaining the process which an image processing apparatus performs. It is explanatory drawing explaining the process which an image processing apparatus performs.

Explanation of symbols

1 Image processing device 17 Storage medium 20 Program

Claims (10)

A first resolution conversion means for converting the resolution of the original image data;
Extraction means for extracting the character portion of the image from the image data whose resolution is converted by the first resolution conversion means ;
Second resolution conversion means for converting the resolution of the original image data to a lower resolution than the first means;
File creation means for separately saving the image data of the character portion extracted by the extraction means as a character image and the image data whose resolution is converted by the second resolution conversion means as a background image, separately as one or a plurality of files It equipped with a door,
The file creation means, the lossy compression of the character image, the lossy compression of the background image, respectively, and compression-encoded, and then stored as the one or more files,
An image processing apparatus.   The image processing apparatus according to claim 1, wherein the extraction unit performs the extraction at a predetermined sampling interval when the image data is a color image.   The image processing apparatus according to claim 1, wherein the file creation unit stores the character image as one or a plurality of binary images in the file when the image data is a color image.   4. The image according to claim 3, wherein the extraction unit binarizes the image data, extracts a character part, extracts a color of a color image corresponding to the character part, and specifies a character color. 5. Processing equipment.   The image processing according to any one of claims 1 to 4, wherein when the image data is a color image, the file creation unit stores the character image in the file as a color image obtained by reducing the color of the character image. apparatus. The file creating unit, wherein the smoothing the pre-xenon scene image in front of the non-reversible compression, the image processing apparatus according to one of claim 1. The file creating unit, before Kibun letter image by OCR processing, those with text data of the text in transparent color The results of this OCR, and pre Kibun shaped images, and pre-xenon scene image, respectively Layer and save the file as layer management file, the layer of the text data and the character image of a layer above the layer, the image processing apparatus according to one of any claims 1 to 4. A first resolution conversion procedure for converting the resolution of the original image data;
An extraction procedure for extracting the character portion of the image from the image data whose resolution has been converted by the first resolution conversion means;
A second resolution conversion procedure for converting the resolution of the original image data to a lower resolution than the first means;
A file creation procedure for storing the image data of the character part extracted by the extraction procedure as a character image and the image data whose resolution is converted by the second resolution conversion procedure as a background image and separately storing them as one or a plurality of files And
In the file creation procedure, the character image is reversibly compressed, the background image is lossy compressed, and is compressed and encoded, and then stored as the one or more files.
An image processing method characterized by the above. A program for causing a computer to execute the image processing method according to claim 8 . A computer-readable recording medium on which the program according to claim 9 is recorded.

Images