JP2006048180A - Image processor, image processing method and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce data amounts while maintaining the image quality of image data. <P>SOLUTION: This image processor is provided with an outline conversion part 213 for converting the outline of each of connected components included in image data configured of dots into vector data, a pattern holding part 215b for storing the connected components converted into the vector data by adding an index as the pattern of each of the connected components, a similarity determination part 215a for determining whether or not the connected components converted into the vector data are similar to the pattern stored by the pattern storing part, an index extracting part 215c for converting each of the connected components included in the image data into the index of the pattern whose similarity to the connected components is determined by the similarity judging part and position information on the image data and a vector data generating part 216 for outputting the index and the position information and the pattern as vector data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program for storing, printing, displaying, and transmitting a read image to another image processing apparatus.

  An image processing apparatus according to a conventional technique usually optically reads an image such as an office document or a handwritten manuscript with a scanner, holds the read image data as dot data, prints the dot data, Or sent to another image processing apparatus. This image processing apparatus is used in, for example, a digital copying machine, a facsimile machine, an electronic filing apparatus, and the like.

  Also, in such an image processing apparatus, the dot data is compressed and held in order to reduce the amount of data and efficiently hold the image data or transmit it to another image processing apparatus. As a method for compressing a raster image expressed by dots, methods such as run length compression and entropy compression are known.

  However, when the image composed of dot data has a low resolution at the time of reading, the outline of characters and figures will cause rattling and jaggedness (so-called dot feeling) peculiar to the image read by the optical scanner, and the original When compared with the image, there is a problem that the image appears to deteriorate.

  On the other hand, if the resolution at the time of reading is increased in order to avoid visual deterioration, the above problem can be avoided to some extent. However, the amount of data of the image increases, and the amount of data in the image processing apparatus increases. There has been a problem that it cannot be efficiently held or transmitted to another image processing apparatus. In particular, in the case of a color image, the amount of data increases compared to a monochrome image, and thus more efficient data compression is required. In addition, when transmitting image data to a small terminal, since the storage capacity of the small terminal is generally small, it is desirable that the file to be transmitted has as small a capacity as possible.

  In order to solve the above-described problems caused by the prior art, the present invention can reduce the amount of data without deteriorating the appearance of image data composed of dots compared to the original image, An object is to provide an image processing method and an image processing program.

  In order to solve the above-described problems and achieve the object, the image processing apparatus according to the first aspect of the present invention provides image data composed of dots for each connected component of a group of regions connected by the dots. Outline conversion means for respectively converting vector data representing outlines of connected components, pattern holding means for holding the vector data converted by the outline conversion means with an index for each predetermined pattern, and the outline conversion means Similarity determination means for determining similarity between the connected component converted into vector data and any of the patterns held in the pattern holding means; and the similarity determination means for each connected component included in the image data. The index of the pattern that is judged to be similar to the connected component And indexing means for converting to position information on the image data, the index and the position information converted by the indexing means, and the pattern held by the pattern holding means as vector data of the image data Vector data output means for outputting a file in a predetermined format.

  According to the first aspect of the present invention, when image data (raster image) composed of dots is converted into vector data, the connected components included in the image data can be patterned. Since the outline of the connected component included in the image data is specified by the pattern, the vector data does not include overlapping data. Therefore, the image data can be converted into vector data with a small amount of data.

  An image processing apparatus according to a second aspect of the present invention provides the image processing apparatus according to the first aspect, wherein the index of the pattern is extracted from the vector data in a predetermined file format output from the vector data output means. Extracting means, pattern extracting means for extracting the pattern in the vector data corresponding to the index extracted by the indexing means, and the pattern corresponding to the index extracted by the pattern extracting means, And image restoration means for restoring image data constituted by the dots by combining them based on the position information output by the indexing means.

  According to the second aspect of the present invention, the image data converted into vector data by the image processing apparatus according to the first aspect of the invention can be restored to image data composed of dots.

  According to a third aspect of the present invention, there is provided the image processing apparatus according to the second aspect, wherein the connected component converted into vector data by the outline converting means is connected to one side of a rectangle circumscribed by the outline of the connected component. Normalization means by scaling all the lengths of the two to be equal, and normalization means for outputting to the similarity determination means, wherein the indexing means is a magnification when the normalization means normalizes the connected components Information is output together with the index and the position information, and the image restoration means restores the magnification of the pattern based on the magnification information output by the indexing means.

  According to the third aspect of the present invention, since the connected components included in the image data are normalized, the number of connected components determined to be similar increases, and the number of patterns can be reduced. Therefore, the image data can be converted into vector data with a smaller data amount. In addition, the normalized pattern based on the magnification information at the time of normalization can be restored to the original magnification.

  According to a fourth aspect of the present invention, there is provided an image processing apparatus according to the first aspect, wherein the vector data output by the vector data output means is transmitted to another image processing apparatus via a network. Means are provided.

  According to the fourth aspect of the present invention, the image data converted into the vector data by the image processing apparatus according to the first aspect can be transmitted to another image processing apparatus. Since the image data has a small amount of data, efficient data transmission can be performed.

  An image processing apparatus according to a fifth aspect of the present invention is the image processing apparatus according to the second aspect, wherein the vector data output from the vector data output means is stored in the data storage means. Data reading means for reading out the vector data and outputting the vector data to the index extraction means.

  According to the fifth aspect of the present invention, the image data converted into vector data by the image processing apparatus according to the second aspect of the present invention can be stored. Furthermore, the stored image data can be restored and used.

  According to a sixth aspect of the present invention, there is provided an image processing method for converting image data composed of dots into vector data representing outlines of connected components for each connected component of a group of regions connected by the dots. An outline conversion step, a pattern holding step for indexing and holding the vector data converted by the outline conversion step for each predetermined pattern, and the connected component converted to vector data by the outline conversion step, A similarity determination step for determining similarity to any one of the patterns held in the pattern holding step, and each of the connected components included in the image data of the pattern that the similarity determination step determines to be similar to the connected components. Convert to position information on index and image data Vector data for outputting a file in a predetermined format as vector data of the image data, the indexing step, the index and the position information converted by the indexing step, and the pattern held by the pattern holding step And an output step.

  According to the sixth aspect of the present invention, when image data (raster image) composed of dots is converted into vector data, the connected components included in the image data can be patterned. Since the outline of the connected component included in the image data is specified by the pattern, the vector data does not include overlapping data. Therefore, the image data can be converted into vector data with a small amount of data.

  An image processing method according to a seventh aspect of the invention is the image processing method according to the sixth aspect, wherein the index of the pattern is extracted from the vector data of a predetermined file format output from the vector data output step. An extraction step, a pattern extraction step for extracting the pattern in the vector data corresponding to the index extracted by the indexing step, and the pattern corresponding to the index extracted by the pattern extraction step. An image restoration step of restoring image data constituted by the dots by combining them based on the position information output by the indexing step.

  According to the seventh aspect of the invention, the image data converted into the vector data by the image processing method according to the sixth aspect of the invention can be restored to image data composed of dots.

  An image processing method according to an eighth aspect of the present invention is the image processing method according to the seventh aspect, wherein the connected component converted into the vector data by the outline converting step is one side of a rectangle circumscribed by the outline of the connected component. And normalizing by scaling all the lengths to be equal, and outputting to the similarity determination step, the indexing step is a magnification when the normalization step normalizes the connected components Information is output together with the index and the position information, and the image restoration step restores the magnification of the pattern based on the magnification information output by the indexing step.

  According to the eighth aspect of the invention, since the connected components included in the image data are normalized, the number of connected components determined to be similar increases, and the number of patterns can be reduced. Therefore, the image data can be converted into vector data with a smaller data amount. In addition, the normalized pattern based on the magnification information at the time of normalization can be restored to the original magnification.

  According to a ninth aspect of the present invention, in the image processing method according to the sixth aspect, in the similarity determination step, a user can set a determination criterion as to whether or not the connected components are similar to each other. It is characterized by being.

  According to the ninth aspect of the present invention, the user can set a criterion for determining whether or not the connected components are similar, and the number of patterns included in the vector data can be adjusted. Therefore, the user can adjust the strictness of the restored image and the capacity of the vector data.

  An image processing method according to a tenth aspect of the present invention is the image processing method according to the sixth aspect, wherein the image data is color image data including a plurality of colors.

  According to the tenth aspect of the present invention, when color image data including a plurality of colors is converted into vector data, a connected component included in the color image data is a pattern regardless of the color of the connected component. Can be For similar connected components, the outline is specified by one pattern regardless of the color of the connected components, so that the vector data does not include overlapping data. Therefore, color image data can be converted into vector data with a small amount of data.

  An image processing program according to an eleventh aspect causes a computer to execute the image processing method according to any one of the sixth to tenth aspects.

  According to the eleventh aspect of the present invention, it is possible to cause a computer to execute the image processing method according to any one of the sixth to tenth aspects.

  According to the image processing apparatus, the image processing method, and the image processing program of the present invention, image data composed of dots can be converted into vector data with a small data amount. Therefore, efficient image processing can be performed, and the load on the communication line and the destination storage medium can be reduced when image data is transmitted to another image processing apparatus or the like.

  Further, since the image data is converted into vector data, the image data can be used without deteriorating the appearance of the optically read image as compared with the original image.

  Exemplary embodiments of an image processing apparatus, an image processing method, and an image processing program according to the present invention are explained in detail below with reference to the accompanying drawings.

(Embodiment)
(Contents of system configuration)
FIG. 1 is a system configuration diagram showing a network configuration including an image processing apparatus according to the present embodiment. An image processing apparatus 100 according to the present embodiment is connected to a network 101. In addition to the image processing apparatus 100 according to the present embodiment, a digital multifunction peripheral 102 having a scanner function and a printer function, personal computers (PCs) 103 and 104, a network printer 105, and a network scanner 106 are connected to the network 101. Has been. These can transmit / receive data to / from each other via a wired or wireless network 101.

  In addition, the electronic book terminal device 107 can be connected. The electronic book terminal device 107 has display means using electronic paper or the like, and displays contents downloaded via the network 101. The electronic book terminal device 107 may be always connected to the network 101 wirelessly or by wire, or may be connected to the network when content is downloaded, and then used offline. The electronic book terminal device 107 may be a dedicated terminal device or an information processing terminal device such as a PDA or a mobile phone.

(Functional configuration of image processing apparatus)
Next, the functional configuration of the image processing apparatus according to this embodiment will be described. FIG. 2 is a block diagram functionally showing the configuration of the image processing apparatus 100 according to the present embodiment.

  In FIG. 2, a reading unit 201 optically reads a document such as a document and outputs it to the conversion unit 202 as image data (binary raster image data) composed of dots. Specifically, the function is realized by an optical scanner having a CCD (Charge Coupled Device) or the like.

  The conversion unit 202 performs raster / vector conversion for each piece of image data output from the reading unit 201, converts the image data into vector data, and stores the vector data file as a storage unit 203, a printing unit 204, a communication unit 205, The data is output to the display unit 206. The conversion unit 202 restores the vector data file stored in the storage unit 203 or transmitted from another information processing apparatus via the communication unit 205 to raster image data by vector raster conversion, and stores the storage unit 203. , Output to the printing unit 204, the communication unit 205, and the display unit 206.

  Specifically, the conversion unit 202 realizes its function by executing a program stored in a ROM, RAM, hard disk (HD), etc. (not shown) by a CPU (not shown).

  The storage unit 203 stores the vector data output from the conversion unit 202. Specifically, the function is realized by a hard disk, a flexible disk, a CD-RW, or a USB-connected memory stick. In addition, the storage unit 203 also stores information relating to scaling of image data formed by the printing unit 204 described later based on the vector data.

  The printing unit 204 forms image data based on the vector data output from the conversion unit 202 or the vector data stored in the storage unit 203, and prints the image data on printing paper. Specifically, the function is realized by a printer such as a laser printer or an inkjet printer, a copier, or the like. Further, the printing unit 204 can print the image data in an arbitrary size by scaling the vector data when forming the image data.

  A communication unit 205 is connected to the network 101 and transmits vector data output from the conversion unit 202 or vector data stored in the storage unit 203 via the network 101 to the digital multifunction peripheral 102, the PCs 103 and 104, and the network printer 105. Send to etc. Specifically, the function is realized by a network interface or the like.

  The communication unit 205 receives data transmitted from the digital multi-function peripheral 102, the PCs 103 and 104, the network printer 105, and the like via the network 101. The received data is output to the conversion unit 202 and subjected to raster / vector conversion. The converted vector data is stored in the storage unit 203 or output to the printing unit 204, the display unit 206, and the like.

  When transmitting the vector data, the communication unit 205 can transmit the vector data by attaching it to a PDF file, for example. In this manner, by connecting the scanner function and the printer function via the network 101, a so-called facsimile function can be realized. In this case, the dot feeling peculiar to the output of the facsimile does not occur, and the appearance does not deteriorate as compared with the original image. Also, since the data amount is usually smaller than the compressed facsimile data, efficient image data communication can be realized.

  The display unit 206 forms image data based on the vector data output from the conversion unit 202 or the vector data stored in the storage unit 203, and displays the image data. Specifically, the function is realized by a display such as a liquid crystal display or a plasma display. Further, when forming the image data, the image data can be displayed in an arbitrary size by scaling the vector data.

(Configuration of the conversion unit 202)
Next, a detailed configuration of the conversion unit 202 will be described. The conversion unit 202 includes an input / output unit 211, a color-specific binary image extraction unit 212, an outline conversion unit 213, a normalization processing unit 214, a dictionary processing unit 215, a vector data generation unit 216, an index reading unit 217, and a pattern reading unit. 218, a pattern restoration unit 219, a binary image generation unit 220, and a binary image synthesis unit 221.

  The input / output unit 211 receives data input from the reading unit 201, the storage unit 203, and the communication unit 205 to the conversion unit 202, and outputs the data to the color-specific binary image extraction unit 212 or the index reading unit 217. The data output to the conversion unit 202 is raster image data or vector data. When the received data is raster image data, the data is output to the color-specific binary image extraction unit 212, and when the received data is a vector file, the data is output to the index reading unit 217. .

  The input / output unit 211 also stores the vector data or raster image data output from the vector data generation unit 216 and the binary image synthesis unit 221 in accordance with the use of the data, the storage unit 203, the printing unit 204, and the communication unit 205. And output to one of the display units 206.

  When the raster image data output from the input / output unit 211 is color, the color-specific binary image extraction unit 212 extracts a component for each color used in the image, and outputs a binary raster image for each color. And The color-specific binary image extraction unit 212 outputs the extracted binary raster image for each color to the outline conversion unit 213, and outputs the number of extracted colors (number of colors) to the vector data generation unit 216.

  The outline conversion unit 213 performs outline conversion on the binary raster image for each color output from the color-specific binary image extraction unit 212, and approximates each connected component with a straight line or a curve. The outline conversion unit 213 outputs individual connected components approximated by a straight line or a curve to the normalization processing unit 214 and outputs the number of connected components to the vector data generation unit 216.

  The normalization processing unit 214 performs normalization processing on each connected component approximated by a straight line or a curve output from the outline conversion unit 213. Further, the normalization processing unit 214 outputs the normalized individual connected components to the dictionary processing unit 215, and outputs the coordinates and magnifications of the connected components before normalization to the vector data generation unit 216.

  The dictionary processing unit 215 performs dictionary processing on each normalized connected component output from the normalization processing unit 214, and creates an outline dictionary and extracts a dictionary reference index. The dictionary processing unit 215 includes a similarity determination unit 215a, a pattern holding unit 215b, and an index extraction unit 215c.

  The similarity determination unit 215a determines whether each normalized connected component output from the normalization processing unit 214 is similar to the pattern registered in the pattern holding unit 215b. Register in the unit 215b.

  The pattern holding unit 215b holds the outline of the connected component as an indexed pattern. The pattern registered in the pattern holding unit 215b is output to the vector data generation unit 216 as outline dictionary data each time one piece of image data is converted into vector data. Hereinafter, a pattern group registered in processing one image data is referred to as “outline dictionary data”.

  The index extraction unit 215 c extracts the index of the pattern registered in the corresponding pattern holding unit 215 b for each normalized connected component output from the normalization processing unit 214, and the vector data generation unit 216. Output to.

  The vector data generation unit 216 generates raster image vector data based on the information output from each unit. Further, the vector data generation unit 216 outputs the generated vector data to the input / output unit 211.

  The index reading unit 217 reads the number of colors from the vector data output from the input / output unit 211, and reads the reference pattern index for each color dictionary reference data. Further, the index reading unit 217 outputs the read reference pattern index to the pattern reading unit 218.

  The pattern reading unit 218 reads the pattern of normalized connected components in the outline dictionary data corresponding to the reference pattern index output from the index reading unit 217. The pattern reading unit 218 also outputs the read normalized connected component pattern to the pattern restoration unit 219.

  The pattern restoration unit 219 restores the normalized connected component pattern output from the pattern reading unit 218 to the original magnification based on the magnification information in the vector data. The pattern restoration unit 219 outputs the restored connected component to the binary image generation unit 220.

  The binary image generation unit 220 arranges the restored connected component output from the pattern restoration unit 219 with the original color added to the original coordinates based on the coordinate information and color information in the vector data. Thus, a binary raster image for each color is generated. In addition, the binary image generation unit 220 outputs the generated binary raster image for each color to the binary image synthesis unit 221.

  When there are a plurality of color-specific binary raster images output from the binary image generation unit 220, the binary image composition unit 221 combines them, and a raster image including a plurality of colors (hereinafter referred to as a color raster image). Is generated. In addition, the binary image composition unit 221 outputs the generated color raster image data to the input / output unit 211. When there is one binary raster image output from the binary image generating unit 220, the binary image is output to the input / output unit 211 as it is.

  With the above-described configuration, the conversion unit 202 converts raster image data and vector data. Below, the process of each part at the time of performing these conversions is demonstrated.

(Raster / vector conversion process)
First, processing when the conversion unit 202 converts color image data output from the reading unit 201 into vector data and outputs the vector data will be described with reference to FIGS. FIG. 3 is a flowchart showing a procedure of raster / vector conversion processing by the conversion unit 202. Details of the processing of each step will be described later.

  First, the input / output unit 211 of the conversion unit 202 receives an input of a color raster image composed of dots related to an image of one original read by the reading unit 201 (step S301). Therefore, one color raster image data is input for each original document. Here, depending on the data capacity of the original and the processing capability of the image processing apparatus 100, one original may be divided into a plurality of parts for output. Thereby, the efficiency of the reading process of the reading unit 201 can be improved.

  The color raster image data input to the input / output unit 211 is output to the color-specific binary image extraction unit 212. The color-specific binary image extraction unit 212 scans the input color raster image, and extracts a component for each color of the color raster image as a binary raster image for each color (step S302). Therefore, binary raster images corresponding to the number of colors used in the color raster image are extracted, and the number of colors in the original document serving as the header is specified among the constituent elements of vector data described later. And output to the vector data generation unit 216.

  The binary raster image for each color extracted by the color-specific binary image extraction unit 212 is output to the outline conversion unit 213. The outline conversion unit 213 performs outline conversion on the binary raster image for each color, extracts the outline of each connected component from the binary raster image, and approximates the outline of each connected component with a straight line or a curve ( Step S303). At this time, the number of connected components is specified from among the constituent elements of vector data described later, and is output to the vector data generation unit 216.

  The outline of each connected component extracted by the outline conversion unit 213 is output to the normalization processing unit 214. The normalization processing unit 214 normalizes each connected component by setting the length of one side of the rectangle circumscribed by the extracted outline of each connected component to 1 (step S304). At this time, coordinates and magnifications for each connected component among the constituent elements of vector data described later are specified and output to the vector data generating unit 216.

  The individual connected components normalized by the normalization processing unit 214 are output to the similarity determination unit 215a of the dictionary processing unit 215. The similarity determination unit 215a compares each normalized connected component with all patterns registered in the outline dictionary data of the pattern holding unit 215b (step S305). In the pattern holding unit 215b, the connected components existing in the document are registered as an indexed pattern by the connected component dictionary processing described later. A pattern is a normalized connected component registered in outline dictionary data. The pattern registered in the outline dictionary data can be uniquely specified by the index.

  When the outline of the connected component to be compared matches the pattern registered in the outline dictionary data (step S305: Yes), the index extraction unit 215c extracts the index attached to the matched pattern (step S307). ). On the other hand, when the outline of the connected component to be compared does not match all the patterns stored in the outline dictionary data (step S305: No), the connected component to be compared is registered in the outline dictionary data as a new pattern. (Step S306). The new pattern is indexed at the same time as registration, and the index extraction unit 215c extracts the index (step S307). The extracted index is output to the vector data generation unit 216 and becomes a constituent element of vector data to be described later.

  Until the comparison with the outline dictionary data for all the connected components is completed, the processing from step S305 is repeated (step S308: No), and when the comparison with the pattern is completed for all the connected components (step S308: Yes), For binary raster images of other colors, the processing from step S303 is performed (step S309: No). When the processing is completed for binary raster images of all colors (step S309: Yes), the vector data generation unit 216 outputs the data extracted in step S301 and subsequent steps as a vector data file in a predetermined format (step S309). S310), the process according to this flowchart is terminated.

  Through the processing described above, the conversion unit 202 converts the color raster image into vector data and outputs it as a vector data file. Here, the configuration of the vector data generated by the vector data generation unit 216 will be described with reference to FIGS. FIG. 4 is a diagram schematically showing the configuration of a vector data file generated by the vector data generating unit 216. A vector data file 400 generated by the vector data generation unit 216 includes a header unit 401, color-specific dictionary reference data 402a to 402c, and outline dictionary data 403.

  FIG. 5 is a diagram showing the configuration of the vector data file in detail. The header portion 401 is the number of colors used in the image represented by the vector data specified in step S302 in the flowchart of FIG. Following the header portion 401, color-specific dictionary reference data 402a to 402c follow.

  The color-specific dictionary reference data 402a to 402c includes color information 501, the number of connected components 502, and connected component data 503a to 503c. One dictionary reference information for each color indicates the contents of one binary raster image. The color information 501 is information for specifying the color of the binary raster image such as an RGB value. The number of connected components 502 is the number of connected components included in the binary raster data specified in step S303 in FIG. The connected component data 503a to 503c is information about each connected component, and includes a reference pattern index 505, coordinate data 506, and magnification data 507 for the connected component.

  The reference pattern index 505 is an index of the pattern registered in the pattern holding unit 215b extracted in step S307 in FIG. The coordinate data 506 indicates the position of the connected component in the binary raster image, and the magnification data 507 is a coefficient for returning the normalized connected component to the original magnification. The coordinate data 506 and the magnification data 507 are specified in step S304 in FIG.

  The outline dictionary data 403 is a pattern of normalized connected components existing in a document registered in the pattern holding unit 215b by lexicization processing described later. The connected component in the document can be restored by specifying one pattern in the outline dictionary data 403 by the reference pattern index 505 and determining the coordinates and magnification in the document by the coordinate data 506 and the magnification data 507.

(Contents of vector / raster conversion process)
Next, processing for restoring a raster image from the vector data file output in this way will be described with reference to FIGS. FIG. 6 is a flowchart showing a procedure of vector / raster conversion processing by the conversion unit 202.

  First, a vector data file is input to the input / output unit 211 of the conversion unit 202 (step S601). The input / output unit 211 outputs the input vector data file to the index reading unit 217. The index reading unit 217 recognizes the number of color-specific dictionary reference data 402 (402a to 402c) by the header portion 401 of the vector data file output from the input / output unit 211, and concatenates one color-specific dictionary reference data 402. The reference pattern index 505 of the component data 503a to 503c is read (step S602). The read reference pattern index is output to the pattern reading unit 218.

  The pattern reading unit 218 reads the pattern corresponding to the reference pattern index 505 from the outline dictionary data 403 (step S603) and outputs it to the pattern restoration unit 219. The pattern restoration unit 219 restores the pattern output from the pattern reading unit 218 to the original size outline based on the magnification data 507 for each of the connected component data 503a to 503c (step S604). The restored outline is output to the binary image generation unit 220.

  When the magnification conversion processing is completed for all the connected component data 503 included in one color-specific dictionary reference data 402, the binary image generation unit 220 arranges the outline for each connected component at the position indicated by the coordinate data 506, and A binary raster image is generated with the color specified by the color information 501 (step S605). The generated binary raster image is output to the binary image composition unit 221.

  Hereinafter, until the processing is completed for all the color-specific dictionary reference data 402, the processing from step S602 is performed for each color-specific reference data 402 (step S606: No). When the processing is completed for all the color-specific reference data 402 (step S606: Yes), the binary image synthesis unit 221 restores the color raster image by synthesizing the binary raster image for each color (step S607). The process by is terminated.

  Through the processing as described above, the conversion unit 202 performs raster / vector conversion and vector / raster conversion of the image data to realize vector data conversion and raster image conversion of the image data.

(Contents of binary raster image extraction processing for each color)
Next, details of processing performed by the conversion unit 202 in the flowchart shown in FIG. 3 will be described. First, in step S302, the processing for when the binary image extraction unit 212 for each color extracts a binary raster image for each color from a color raster image in which a plurality of colors are used is described with reference to FIGS. I will explain. 7 to 13 are explanatory diagrams illustrating processing steps when a binary raster image is extracted from a color raster image.

  First, color dot data 700 is input to the color-specific binary image extraction unit 212. FIG. 7 shows an example of color dot data 700. In the figure, the same shaded areas indicate dots of the same color. Next, the color-specific binary image extraction unit 212 scans the dot data 700 in dot units sequentially from the end. FIG. 8 is an explanatory diagram schematically showing a scanning process for the dot data 700 of the color-specific binary image extraction unit 212. The color-specific binary image extraction unit 212 performs scanning in the direction of the arrow in the figure one dot at a time from the edge of the image. When the dot 801 including the color is found, the position of the dot having the same color as the dot 801 is recognized during the subsequent scanning. That is, the position of a dot having the same color as the colored dot found first during scanning is recognized.

  When the binary image extraction unit 212 for each color finishes scanning the entire dot data 700, it extracts a dot having the same color as the dot 801 based on the position information identified during the previous scan as shown in FIG. Since all the extracted dots have the same color, they can be handled as a binary raster image. FIG. 9 is a diagram illustrating an example of dot data of the same color extracted from the dot data 700. Only the dot group 901 having the same color as the dot 801 is extracted and can be handled as the binary raster image 900.

  The color-specific binary image extraction unit 212 scans again the dot data 700 after the dot group 901 having the same color as the dot 801 is extracted as shown in FIG. FIG. 10 is an explanatory diagram schematically showing a scanning process for the color-specific binary image extraction unit 212 dot data 700. The color-specific binary image extraction unit 212 starts scanning from the position of the colored dot 801 that is first found during the first scanning. This is because it is already known by the first scanning that there is no colored dot before this. When the colored dot 1001 is first found during the second scanning, the position of the dot having the same color as the dot 1001 is recognized during the subsequent scanning.

  When the color-specific binary image extraction unit 212 finishes the second scan of the dot data 700, it extracts a dot having the same color as the dot 1001 as shown in FIG. To do. FIG. 11 is a diagram illustrating an example of dot data of the same color extracted from the dot data 700. Only a dot group 1101 having the same color as the dot 1001 is extracted and can be handled as a binary raster image 1100.

  Further, the color-specific binary image extraction unit 212 scans the dot data 700 again after the dot group 1101 having the same color as the dot 1001 is extracted as shown in FIG. FIG. 12 is an explanatory diagram schematically showing scanning processing for the dot data 700 of the binary image extracting unit 212 for each color. The color-specific binary image extraction unit 212 starts scanning from the position of the colored dot 1001 first found during the second scanning, and when the colored dot 1201 is found first during the third scanning. During the subsequent scanning, a dot having the same color as the dot 1201 is recognized.

  After completing the third scan of the dot data 700, the color-specific binary image extraction unit 212 extracts dots of the same color as the dot 1201 based on the information of the positions identified during the previous scan as shown in FIG. To do. FIG. 13 is a diagram illustrating an example of dot data of the same color extracted from the dot data 700. Only a dot group 1301 having the same color as the dot 1201 is extracted and can be handled as a binary raster image 1300. By repeating the above procedure, the color-specific binary image extraction unit 212 extracts a binary raster image for each color from the color raster image.

(Contents of outline conversion)
Next, outline conversion performed by the outline conversion unit 213 in step S304 will be described. Outline conversion is performed by first extracting an outline for each connected component from a binary raster image and then approximating the extracted outline with a straight line. First, the process of extracting the outline for each connecting line segment from the binary raster image will be described with reference to FIGS.

  14 to 21 are explanatory diagrams showing outline extraction processing steps. First, as shown in FIG. 14, the vertices of a dot rectangle including a figure (hereinafter simply referred to as “figure”) 1401 whose outline is to be obtained are extracted. For example, the vertices of the dot rectangle are extracted as 1402, 1403, 1404 in the figure. In the figure, for the sake of explanation, the figure 1401 is shown as a shaded portion, but it is actually a colored portion of a binary image. In addition, the portions other than the shaded portions are colorless portions of the binary image.

  Next, as shown in FIG. 15, the extracted vertices are connected by a line segment along the side of the dot rectangle. For example, vertices 1402 and 1403 are connected by a line segment 1501. One vertex 1402 and 1404 are connected by a line segment 1502.

  Subsequently, as shown in FIG. 16, the color on both sides of the line segment is compared for each line segment connecting the vertices. If it belongs to the figure for which an outline is desired, it will be colored, otherwise it will be colorless. If the colors on both sides of the line segment are the same color, the line segment is removed. That is, if both sides of the line segment are colored as indicated by 1701 in FIG. 17, the line segment is removed. Further, when one side is colored and the other side is colorless as in 1702 and 1703, the line segment is left.

  In FIG. 16, a line segment 1501 remains because the colors on both sides thereof are different. Further, the line segment 1502 (see FIG. 15) is removed because both sides thereof are the same color. Note that the process of connecting the vertices with line segments and the process of removing line segments of the same color on both sides of the line segments may be performed simultaneously.

  Next, as shown in FIG. 18, the direction is set so that the right side is colored toward each vertex. For example, the vertex 1402 is oriented in the direction of the vertex 1403 as indicated by the arrow 1801. That is, each vertex is oriented in the direction as indicated by arrows 1901, 1902, 1903, and 1904 shown in FIG. Then, the directions of adjacent vertices are compared, and as shown in FIG. 20, if the directions are the same, line segment reduction is performed for that portion, and the straight line is degenerated. That is, the vertex 1403 remains because it has a different direction from the adjacent vertex, but the vertex 1402 is surrounded by the straight line 2001 because it has the same direction as the next vertex.

  Then, as shown in FIG. 21, the line segment is traced along the direction attached to the degenerated straight line, and the starting point and the outline are extracted. In FIG. 21, two outlines are extracted from the graphic 1401. These two outlines form a figure 1401 by closed regions starting from points 2101 and 2102 and ending points. By the procedure as described above, in step S302, an outline is extracted for each connected line segment of the binary raster image input to the outline conversion unit 213.

(Contents of straight line approximation process)
Next, outline linear approximation by broken line approximation will be described with reference to FIGS. 22 and 23 are diagrams showing processing steps for polygonal line approximation. First, as shown in FIG. 22, a straight line 2201 that approximates the arrangement of the vertices of the outline dot rectangle is obtained from the position of each vertex of the outline dot rectangle. Next, as shown in FIG. 23, a start point 2301 and an end point 2302 of the obtained straight line are determined, and information on corner points and curves is given to these points. That is, the outline from the start point to the end point can be expressed by the position information of the start point and the end point and the information of the corner points and curves given thereto. By repeating this, the outline of the connected component can be expressed by a straight line by a plurality of vertices (start point / end point). In the following description, the outline of the connected component is approximated by a straight line. However, the approximation is also performed using information on each vertex (start point / end point and control point) by curve approximation using a Bezier curve or the like. Can do.

(Contents of connected component normalization)
Next, the processing of the normalization processing unit 214 when normalizing the connected components in step S304 will be described with reference to FIGS. 24 to 26 are explanatory diagrams illustrating processing steps of the normalization processing unit 214 when normalizing connected components. The normalization of the connected components is performed by setting the length of one side of the rectangle circumscribed by the outline to 1 for each extracted connected component.

First, the coordinates of each vertex of the outline approximated by a broken line for each connected component are obtained. As shown in FIG. 24, each connected component is linearly approximated by a plurality of vertices. In FIG. 24, the vertexes of the connected component 2400 are seven points 2411 to 2417. Further, the vertexes of the connected component 2420 are 15 points of vertices 2431 to 2445. Similarly, the vertexes of the connected component 2460 are 7 points of vertices 2471 to 2477, and the vertexes of the connected component 2480 are 7 points of vertices 2491 to 2497. The coordinates of these vertices are expressed as (x ₀ , y ₀ ). For example, the coordinates of the vertex 2411 are represented as (x ₂₄₁₁ , y ₂₄₁₁ ), and the coordinates of the vertex 2412 are represented as (x ₂₄₁₂ , y ₂₄₁₂ ).

Next, as shown in FIG. 25, a rectangle circumscribing the vertex of the outline of each connected component (a circumscribed rectangle) is obtained. The coordinates of the circumscribed rectangle are {(x _min , y _min ), (x _min , y _Max ), (x _Max , y _Max ), (x _Max , y _min )}, where x _Max is the value of each vertex of the connected component The maximum value of the x coordinate, x _min is the minimum value of the x coordinate of each vertex of the connected component, y _Max is the maximum value of the y coordinate of each vertex of the connected component, and y _min is the minimum value of the y coordinate of each vertex of the connected component It is expressed as

For example, the circumscribed rectangle 2500 of the connected component 2400 is {(x ₂₄₁₁ , y ₂₄₁₂ ), (x ₂₄₁₁ , y ₂₄₁₆ ), (x ₂₄₁₄ , y ₂₄₁₆ ), (x ₂₄₁₄ , y ₂₄₁₂ )} (see FIG. 24). expressed. Similarly, a circumscribed rectangle 2520 of the connected component 2420, a circumscribed rectangle 2560 of the connected component 2460, and a circumscribed rectangle 2580 of the connected component 2480 are obtained.

Then, the circumscribed rectangle of each connected component is normalized so that the length of each side is 1. That is, the coordinates of each circumscribed rectangle are {(x _min , y _min ), (x _min , y _Max ), (x _Max , y _Max ), (x _Max , y _min )}, where x _Max −x _min = 1. , Y _Max −y _min = 1.

  FIG. 26 shows an example of normalized connected components. The circumscribed rectangle 2600 is obtained by normalizing the circumscribed rectangle 2500 (see FIG. 25). Similarly, the circumscribed rectangle 2620 is obtained by normalizing the circumscribed rectangle 2520, the circumscribed rectangle 2660 is obtained by normalizing the circumscribed rectangle 2560, and the circumscribed rectangle 2680 is obtained by normalizing the circumscribed rectangle 2580. The length of each side of all the circumscribed rectangles after normalization is 1. Through the above processing, the connected components are normalized in step S303.

(Content of connected component dictionary processing)
Next, the connected component lexicization processing performed by the lexicization processing unit 215 in steps S305 to S307 in FIG. 3 will be described with reference to FIGS. 27 and 28 and FIG. The dictionary of connected components refers to registering normalized connected components as patterns in the pattern holding unit 215b, that is, the outline dictionary in the image data, and when there are a plurality of the same patterns in one document. The object is to reduce the amount of data without holding duplicate information.

  FIGS. 27 and 28 are diagrams schematically showing processing when a connected component is dictionaryd. Here, as an example, normalized connected components 2701 to 2704 are dictionaryd. As shown in FIG. 27, no pattern is initially registered in the outline dictionary 2700. Therefore, in S305 of the flowchart of FIG. 3, when the similarity determination unit 215a compares whether the same pattern as the connected component 2701 is registered in the outline dictionary, there is no pattern registered in the outline dictionary 2700 (step S305: No). Therefore, the connected component 2701 is registered in the outline dictionary 2700 as a new pattern as shown in FIG. 28 (step S306). At this time, an index is attached as means for specifying the pattern in the outline dictionary 2700. Here, “0001” is assigned as an index. Then, the index 0001 is extracted by the index extraction unit 215c as means for specifying the shape of the connected component 2701 (step S307).

  Next, when the similarity determination unit 215a compares whether the same pattern as the connected component 2702 is registered in the outline dictionary 2700 (step S305), the connected component 2701 is registered as the pattern of the index 0001. Since the connected component 2702 and the connected component 2701 match, it is determined that the same pattern as the connected component is registered in the outline dictionary 2700 in the dictionary (step S305: Yes). The index extraction unit 215c extracts the index 0001 as means for specifying the shape of the connected component 2702 (step S307). Similarly, since it is determined that the connected component 2703 matches the connected component 2701, the index 0001 is extracted as a means for specifying the shape of the connected component 2703.

  On the other hand, when it is compared whether the same pattern as the connected component 2704 is registered in the outline dictionary 2700 (step S305), it is not determined that the connected component 2704 and the connected component 2701 match (step S305: No). Therefore, the connected component 2704 is registered as a new pattern in the outline dictionary 2700 (step S306), and an index “0002” for identifying the pattern is attached. Then, the index 0002 is extracted as means for specifying the shape of the connected component 2704 (step S307).

  In this way, by patternizing connected component patterns used in a document and specifying the pattern with an index, the data of the pattern that is used redundantly in one document can be discarded. For this reason, compared to the image data subjected to normal raster / vector conversion, duplicate data is not retained, and the amount of data after conversion can be further reduced.

  Next, processing when the similarity determination unit 215a compares whether the pattern registered in the outline dictionary 2700 matches the connected component to be compared will be described with reference to FIGS. Whether the pattern registered in the outline dictionary 2700 matches the connected component to be compared is determined by comparing both vector components. Note that since the patterns registered in the outline dictionary are normalized connected components, the terms “pattern” and “connected component” are not particularly distinguished in the following description.

FIG. 29 is a diagram illustrating an example of a vector component of connected components. The connected component 2900 is represented by 7 points of vertices 2911 to 2917 by broken line approximation. The connected component 2920 is represented by 15 points of vertices 2931 to 2945 by broken line approximation. Here, the vector positioned in the top direction among the vertices that contact the left side of the circumscribed rectangle of the connected component, that is, the vector that goes clockwise to the vertex 2913 in the connected component 2900 and the vertex 2934 in the connected component 2920 is v. _10, v and _20. Then, the depth of these vectors is set to 0.

Next, a vector directed from the vertex 2913 clockwise to the next vertex 2914 is v ₁₁ , a vector directed clockwise from the vertex 2934 to the next vertex vertex 2935 is v _21, and these depths are 1. Further, a vector that goes from the vertex 2914 clockwise to the next vertex 2915 is v ₁₂ , a vector that goes clockwise from the vertex 2935 to the next vertex vertex 2936 is v _22, and these depths are 2. Similarly, while increasing the depth by 1 in the clockwise direction, the depths of all vectors connecting the vertices are determined.

  Then, based on the flowchart of FIG. 30, it is determined whether or not the connected components to be compared match, and whether or not the vector components of the connected components match. FIG. 30 is a flowchart showing a process of determining whether or not connected components match each other. First, of the connected components to be compared, one Depth is set as p, and the other Depth is set as q, and p = q = 0 is set (step S3001).

  Next, the vectors of Depth p and Depth q are compared (step S3002). If the angle difference between the two vectors is 10 ° or more (step S3003: No), it is determined that the two connected components do not match (step S3011), and the processing according to this flowchart is terminated. When the angle difference between both vectors is less than 10 ° (step S3003: Yes), the lengths of both vectors are compared (step S3004). When the length ratio of both vectors is less than 1.2 (step S3004: Yes), the process proceeds to step S3007.

  On the other hand, when the length ratio of both vectors is 1.2 or more (step S3004: No), a vector obtained by adding one Depth to the shorter vector is added (step S3005). For example, if it is a depth 0 vector, the vector of depth 1 is added. In addition, if the angle difference from the comparison target vector is less than 10 ° and the length ratio is less than 1.2 (step S3006: Yes), the process proceeds to step S3007. If the angle difference is 10 ° or more or the length ratio is 1.2 or more (step S3006: No), it is determined that the two connected components do not match (step S3011), and the processing according to this flowchart ends.

A method for comparing the angles and lengths of such vectors will be described with reference to FIGS. 31 and 32 are diagrams for explaining comparison processing between vectors. The vector v ₁ shown in FIG. 31 is represented by v ₁ = (v _1x , v _1y ). Then, the angle alpha is _{^{α = tan -1 (v 1y /}} v 1x), length | is expressed by ^{_{= √ (v 1x 2 + v}} 1y 2) | v 1 | is | v _1. Specifically, v ₁ = (0.3, 0.4), α = tan ⁻¹ (0.4 / 0.3) = 53 °, | v ₁ | = √ (0.4 ² +0.3 ² ) = 0.5.

This vector v ₁ is compared with the vector v ₂ shown in FIG. The vector v ₂ is represented by v ₂ = (0.2, −0.2), an angle β = −45 °, and | v ₂ | = 0.28. The angle difference | α−β | between the vector v ₁ and the vector v ₂ is | α−β | = | 53 ° − (− 45 °) | = 98 °, and the length ratio Max (| v ₁ | / | v ₂ | , | V ₂ | / | v ₁ |) is Max (| v ₁ | / | v ₂ |, | v ₂ | / | v ₁ |) = Max (0.5 / 0.28, 0.28 / 0.5) = 1.7.

Therefore, the vectors v ₁ and v ₂ are determined not to match in the flowchart of FIG. In the flowchart of FIG. 30, 10 ° is used as the threshold value for the angle difference and 1.2 is used as the threshold value for the length ratio. However, the threshold value can be set to an arbitrary value. By bringing the threshold value close to 0, more strict comparison processing can be performed.

  Returning to the description of the flowchart of FIG. 30, in step S3007, it is determined whether vectors of Depth p + 1 and Depth q + 1 exist. If both Depth p + 1 and Depth q + 1 exist (step S3007: Yes), 1 is added to each of Depth p and Depth q (step S3008), and the process returns to step S3002 and the subsequent processing is repeated. On the other hand, when both Depth p + 1 and Depth q + 1 are not present or only one of them is present (Step S3007: No), if only one of Depth p + 1 or Depth q + 1 is present (Step S3010: Yes), It is determined that the two connected components do not match (step S3011), and the processing according to this flowchart ends. If neither Depth p + 1 nor Depth q + 1 exists (step S3010: No), it is determined that the two connected components match (step S3012), and the processing according to this flowchart ends.

  When it is determined that the connected component matches the pattern registered in the outline dictionary (step S3012), the connected component is not registered in the outline dictionary, and only the index of the pattern is extracted (FIG. 3). Step S305). On the other hand, if it is determined that they do not match (step S3011), the connected component is registered in the outline dictionary as a new pattern (step S306 in FIG. 3). By the processing as described above, the similarity determination unit 215a determines whether the pattern registered in the outline dictionary matches the connected component to be compared, and creates a dictionary of the connected components.

  The vector data file output by the image processing apparatus 100 according to the present embodiment by such a lexicographic process does not hold duplicate pattern data, and therefore more data than normal raster vector conversion. The amount is small and light data.

(Contents of image data retention / use)
Next, with reference to FIGS. 2, 33, and 34, processing in a case where the image processing apparatus 100 according to the present embodiment reads a document and holds and uses the read image data will be described.

  First, the reading unit 201 reads a document to be used. FIG. 33 is an example of a document to be read by the image processing apparatus 100, in which black and white binary bitmap data is printed out. Three circles are shown, and solid lines in the figure are image lines. is there. For example, solid lines 3301 and 3302 indicate the image lines of the outermost circle. However, the smooth image line of the circle cannot be expressed, and the jaggedness of the dots has occurred.

  The reading unit 201 optically reads the original, and outputs the original to the conversion unit 202 as one binary raster image data file for each sheet. Depending on the volume of image data of the document, it may be divided into a plurality of files for output. The converting unit 202 performs raster / vector conversion for each of the binary raster image data files output from the reading unit 201 and outputs the converted data to the storage unit 203 as a vector data file.

  FIG. 34 is an explanatory diagram showing the result of raster / vector conversion of the document shown in FIG. As shown in FIG. 33, the jagged edges of the dots are eliminated, and circles 3401 to 3403 having smooth outlines are realized. Further, since it is vector data, even if the image data is scaled by each functional unit such as the printing unit 204 and the display unit 206, the appearance does not deteriorate compared to the original image.

  Further, since the circles 3401 to 3403 are similar in shape, they are reproduced by a single dictionary pattern. For this reason, the capacity of vector data output as a result of raster-vector conversion by the image processing apparatus 100 according to the present embodiment is smaller than that of normal raster-vector conversion.

  For convenience of explanation, the original is assumed to be a monochrome binary. However, even for a color image, only color information 501 (see FIG. 5) and color boundary value information included in the original are added. The amount of data increase is negligible.

  As described above, according to the present embodiment, the entire image read for each sheet (document) is regarded as one image and raster-vector conversion is performed, so that the image is drawn on one document. The entire image can be made into one vector data. Therefore, when printing or displaying based on the vector data, there is no rattling or jaggedness peculiar to the image read by the optical scanner, and the appearance does not deteriorate as compared with the original image. Furthermore, since the entire image drawn on one document is made into one vector data, only a part of the image cannot be changed / tampered.

  Further, according to the present embodiment, raster-vector conversion is performed on a color image including a plurality of colors in the same manner as a monochrome image, and the entire original drawn on one original is converted into one vector data. can do. Further, even a document using a plurality of colors can be efficiently converted to vector data without increasing the data amount.

  Furthermore, according to the present embodiment, since the connected components included in the original are normalized and the similar connected components are converted into a dictionary, the output vector data does not include duplicate data. Therefore, the amount of data can be further reduced compared to vector data obtained by normal raster / vector conversion.

  The image processing method described in the present embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. Further, this program may be a transmission medium that can be distributed via a network such as the Internet.

  As described above, the image processing apparatus, the image processing method, and the image processing program according to the present invention are suitable for processing an image read by an optical method, and in particular, an electronic filing apparatus, a digital copying machine (multifunction machine), a printer, and a network. As a printer, the present invention can be applied to an alternative to a facsimile.

1 is a system configuration diagram showing a network configuration including an image processing apparatus according to an embodiment. 1 is a block diagram functionally showing the configuration of an image processing apparatus 100 according to the present embodiment. 10 is a flowchart illustrating a procedure of raster / vector conversion processing by a conversion unit 202. It is the figure which showed typically the structure of the vector data file which the vector data generation part 216 produces | generates. It is the figure which showed the structure of the vector data file in detail. 5 is a flowchart showing a procedure of vector / raster conversion processing by a conversion unit 202. It is explanatory drawing which shows the process at the time of extracting a binary raster image from a color raster image. It is explanatory drawing which shows the process at the time of extracting a binary raster image from a color raster image. It is explanatory drawing which shows the process at the time of extracting a binary raster image from a color raster image. It is explanatory drawing which shows the process at the time of extracting a binary raster image from a color raster image. It is explanatory drawing which shows the process at the time of extracting a binary raster image from a color raster image. It is explanatory drawing which shows the process at the time of extracting a binary raster image from a color raster image. It is explanatory drawing which shows the process at the time of extracting a binary raster image from a color raster image. It is explanatory drawing which shows the process process of outline extraction. It is explanatory drawing which shows the process process of outline extraction. It is explanatory drawing which shows the process process of outline extraction. It is explanatory drawing which shows the process process of outline extraction. It is explanatory drawing which shows the process process of outline extraction. It is explanatory drawing which shows the process process of outline extraction. It is explanatory drawing which shows the process process of outline extraction. It is explanatory drawing which shows the process process of outline extraction. It is a figure which shows the process process of a broken line approximation. It is a figure which shows the process process of a broken line approximation. It is explanatory drawing which shows the process of the normalization process part 214 at the time of normalizing a connection component. It is explanatory drawing which shows the process of the normalization process part 214 at the time of normalizing a connection component. It is explanatory drawing which shows the process of the normalization process part 214 at the time of normalizing a connection component. It is a figure which shows typically the process at the time of making a connected component into a dictionary. It is a figure which shows typically the process at the time of making a connected component into a dictionary. It is a figure which shows an example of the vector component of a connection component. It is a flowchart which shows the process of the judgment process of whether a connection component corresponds. It is a figure explaining the comparison process between vectors. It is a figure explaining the comparison process between vectors. 2 is a diagram illustrating an example of a document that is read by an image processing apparatus. FIG. It is explanatory drawing which shows the result of carrying out raster vector conversion of the original of FIG. 33 with the image processing apparatus 100 of this Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Image processing apparatus 101 Network 102 Digital multifunction machine 103,104 Personal computer 105 Network printer 106 Network scanner 107 Electronic book terminal device 201 Reading part 202 Conversion part 203 Storage part 204 Printing part 205 Communication part 206 Display part 211 Input / output part 212 Color Another binary image extraction unit 213 Outline conversion unit 214 Normalization processing unit 215 Dictionary processing unit 215a Similarity determination unit 215b Pattern holding unit 215c Index extraction unit 216 Vector data generation unit 217 Index reading unit 218 Pattern reading unit 219 Pattern restoration unit 220 Binary image generation unit 221 Binary image composition unit

Claims (11)

Outline conversion means for converting image data constituted by dots into vector data representing outlines of the connected components for each connected component of a group of regions connected by the dots;
Pattern holding means for holding the vector data converted by the outline conversion means with an index for each predetermined pattern; and
Similarity determination means for determining similarity between the connected component converted into vector data by the outline conversion means and any pattern held in the pattern holding means;
Indexing means for converting each connected component included in the image data into an index of the pattern that the similarity determining means determines to be similar to the connected component and position information on the image data;
A vector data output means for outputting the index and the position information converted by the indexing means and the pattern held by the pattern holding means as a file in a predetermined format as vector data of the image data;
An image processing apparatus comprising: Index extraction means for extracting an index of the pattern from the vector data in a predetermined file format output from the vector data output means;
Pattern extraction means for extracting the pattern in the vector data corresponding to the index extracted by the indexing means;
Image restoration means for restoring image data constituted by the dots by combining the pattern corresponding to the index extracted by the pattern extraction means based on the position information output by the indexing means;
The image processing apparatus according to claim 1, further comprising: The connected components converted into vector data by the outline converting means are normalized by scaling the lengths of the sides of the rectangle circumscribed by the outline of the connected components to be equal, and output to the similarity determining means. With normalization means,
The indexing means outputs magnification information when the normalizing means normalizes the connected component together with the index and the position information,
The image processing apparatus according to claim 2, wherein the image restoration unit restores the magnification of the pattern based on the magnification information output by the indexing unit.   The image processing apparatus according to claim 1, further comprising a data transmission unit configured to transmit the vector data output by the vector data output unit to another image processing apparatus via a network. Data storage means for storing the vector data output by the vector data output means;
Data reading means for reading the vector data stored in the data storage means and outputting to the index extraction means;
The image processing apparatus according to claim 2, further comprising: An outline conversion step of converting image data constituted by dots into vector data representing outlines of the connected components for each connected component of a group of regions connected by the dots;
A pattern holding step for holding the vector data converted by the outline conversion step with an index for each predetermined pattern; and
A similarity determination step of determining the similarity between the connected component converted into vector data by the outline conversion step and any of the patterns held in the pattern holding step;
An indexing step of converting each connected component included in the image data into an index of the pattern that the similarity determining step determines to be similar to the connected component and position information on the image data;
A vector data output step of outputting the index and the position information converted by the indexing step and the pattern held by the pattern holding step as a file of vector data of the image data in a predetermined format;
An image processing method comprising: An index extraction step of extracting an index of the pattern from the vector data in a predetermined file format output from the vector data output step;
A pattern extraction step for extracting the pattern in the vector data corresponding to the index extracted by the indexing step;
An image restoration step for restoring image data constituted by the dots by combining the patterns corresponding to the indexes extracted by the pattern extraction step based on the position information output by the indexing step;
The image processing method according to claim 6, further comprising: The connected component converted into the vector data by the outline converting step is normalized by scaling the lengths of all sides of the rectangle circumscribed by the outline of the connected component to be equal, and is output to the similarity determining step. Including a normalization step,
In the indexing step, magnification information when the normalizing step normalizes the connected component is output together with the index and the position information,
The image processing method according to claim 7, wherein in the image restoration step, the magnification of the pattern is restored based on the magnification information output by the indexing step.   The image processing method according to claim 6, wherein in the similarity determination step, a user can set a determination criterion as to whether or not the connected components are similar.   The image processing method according to claim 6, wherein the image data is color image data including a plurality of colors. An image processing program for causing a computer to execute the image processing method according to claim 6.

Images