JP2010199984A - Image processor, image reader, and image formation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately and rapidly determine attributes of respective divided regions which are divided in accordance with a connection component from an image. <P>SOLUTION: In an image processing part, an image is divided into a plurality of divided regions in accordance with a connection component; when a first feature quantity of the divided region is included in a feature quantity of a line drawing or chart, the attribute of the divided region is determined to be a line drawing or chart; and, when included in a feature quantity of a figure or photograph, the attribute of the divided region is determined to be a figure or photograph. A divided region where an extracted first feature quantity is included in a boundary part between the feature quantity of a line drawing or chart and the feature quantity of a figure or photograph is selected to extract a second feature quantity related to the width of a connection component formed by the selected divided region; when the extracted second feature quantity shows that the width of the connection component formed by the divided region is equal to or below a threshold, the attribute of the divided region is determined to be a line drawing or chart; and when showing that the width is larger than the threshold, the attribute of the divided region is determined to be a figure or photograph. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, an image reading apparatus, and an image forming apparatus.

  A technique for dividing an image into a character area, a graphic area, a photograph area, and the like is known. For example, in Patent Document 1, a connected region in which black pixels are connected is extracted from a binary image, and each connected region is represented by a line figure based on the shortest distance value from each pixel to the background pixel in the extracted connected region. A technique for classifying a region, a character region, or a photograph region is disclosed. Patent Document 2 discloses a technique for dividing an image into a character area, a graphic area, and a photographic area based on the size of a circumscribed rectangle of a connected component in the image. Further, in Patent Document 3, a binarized image is divided into divided regions for each component, and the number of black pixels in this divided region, the aspect ratio and thickness of the circumscribed rectangle, the rectangularity and thickness, and the size of the circumscribed rectangle Disclosed is a technique for classifying the attributes of each divided area into one of a character area, a graphic area, a photograph area, a table area, or a field separator area based on the ratio of the maximum run length to the length and the black-and-white inversion density in the area ing.

JP 05-189607 A JP 62-086465 A Japanese Patent Laid-Open No. 04-316178

  It is an object of the present invention to accurately determine the attribute of each divided region divided according to a connected component from an image with high accuracy.

  An image processing apparatus according to a first aspect of the present invention includes an area dividing unit that divides an image into a plurality of divided areas based on connected components included in the image, and each divided area divided by the area dividing unit. A first feature quantity extracting means for extracting a first feature quantity, and a first feature quantity extracted by the first feature quantity extraction means is included in a predetermined line drawing or table feature quantity; Determines the attribute of the divided region having the first feature amount as a line drawing or a table, and the first feature amount extracted by the first feature amount extraction means is a feature amount of a graphic or a photograph that is determined in advance. A first attribute determining unit that determines an attribute of the divided region having the first feature amount as a graphic or a photograph, and the first feature amount extracting unit among the plurality of divided regions. The first feature amount extracted by A region selection means for selecting a divided region included in a predetermined range from a boundary for discriminating between a line drawing, a table, and a figure or a photograph in a feature amount space composed of feature amounts of A second feature amount extracting unit that extracts a second feature amount relating to the thickness of the connected component constituting the divided region, and a second feature extracted by the second feature amount extracting unit. And a second attribute discriminating unit for discriminating an attribute of the divided area according to the thickness of the connected component constituting the divided area indicated by the quantity.

  An image processing apparatus according to a second aspect of the present invention is the configuration according to the first aspect, wherein the second attribute determination unit is configured such that the second feature amount extracted by the second feature amount extraction unit is When the thickness of the connected component constituting the divided area indicates that it is equal to or less than a threshold value, the attribute of the divided area is determined as a line drawing or a table.

  An image processing apparatus according to a third aspect of the present invention is the configuration according to the first or second aspect, wherein the second attribute discriminating means is the second feature quantity extracted by the second feature quantity extracting means. However, if the connected component constituting the divided area indicates that the thickness is larger than a threshold value, the attribute of the divided area is determined as a figure or a photograph.

  The image processing apparatus according to a fourth aspect of the present invention is the configuration according to any one of the first to third aspects, wherein the first feature amount extraction unit uses the respective divisions as the first feature amount. The size of the circumscribed rectangle of the region and the pixel density indicating the proportion of pixels that belong to the connected component constituting the divided region occupy the circumscribed rectangle are extracted, and the first attribute determining unit is configured to extract the first feature amount. The discrimination is performed using the size of the circumscribed rectangle and the pixel density extracted by the extraction unit.

  An image processing apparatus according to a fifth aspect of the present invention is the configuration according to any one of the first to fourth aspects, wherein the pixel positions belonging to the connected components constituting the divided region selected by the region selecting unit are turned on. The second feature amount extraction unit includes a contraction processing unit configured to perform a contraction process on the binary image in a state, and the second feature amount extraction unit after the contraction processing is performed by the contraction processing unit as the second feature amount. The number of on-state pixels in the divided area is extracted, and the second attribute determining unit performs the determination based on the number of pixels extracted by the second feature amount extracting unit.

  According to a sixth aspect of the present invention, in the configuration according to any one of the first to fourth aspects, the pixel position belonging to the connected component constituting the divided region selected by the region selecting unit is turned on. A contraction processing unit configured to perform a contraction process on the binary image in a state, and the second feature amount extraction unit includes, as the second feature amount, an on-state pixel in a divided region before and after the contraction process. A change amount of a number is extracted, and the second attribute determination unit performs the determination based on the change amount extracted by the second feature amount extraction unit.

  An image processing apparatus according to a seventh aspect of the present invention is the image processing apparatus according to any one of the first to sixth aspects, wherein the first feature amount extraction unit uses the respective divisions as the first feature amount. When the aspect ratio extracted by the first feature quantity extraction unit is included in a predetermined straight line feature quantity, the aspect ratio is extracted. It is characterized in that an attribute of a divided area having a line is determined as a line drawing, a table, or a straight line.

  The image reading apparatus according to claim 8 of the present invention performs processing using a reading unit that reads an image of a document and an image read by the reading unit. And a processing device.

  The image forming apparatus according to claim 9 of the present invention performs processing using a reading unit that reads an image of a document and an image read by the reading unit. The image processing apparatus includes: a processing apparatus; and an image forming unit that forms an image processed by the image processing apparatus on a recording material.

According to the first aspect of the present invention, compared with the case where the configuration of the present invention is not adopted, the attribute of each divided region divided from the image according to the connected component can be determined with high accuracy and at high speed.
According to the second aspect of the present invention, compared to a case where the configuration of the present invention is not adopted, it is possible to accurately discriminate the line drawing or the divided area of the table.
According to the third aspect of the present invention, compared to a case where the configuration of the present invention is not adopted, it is possible to discriminate the divided area of the figure or the photograph with high accuracy.
According to the fourth aspect of the present invention, as compared with the case where the configuration of the present invention is not adopted, it is possible to reduce the processing load for determining the attribute of the divided region based on the first feature amount.
According to the fifth aspect of the present invention, as compared with the case where the configuration of the present invention is not adopted, it is possible to efficiently perform the process of determining the attribute of the divided region based on the second feature amount.
According to the sixth aspect of the present invention, as compared with the case where the configuration of the present invention is not adopted, it is possible to efficiently perform the process of determining the attribute of the divided region based on the second feature amount.
According to the invention which concerns on Claim 7, based on the 1st feature-value, a linear division area can be discriminate | determined.
According to the eighth aspect of the present invention, compared with the case where the configuration of the present invention is not adopted, the attribute of each divided region divided from the image according to the connected component can be determined with high accuracy and at high speed.
According to the ninth aspect of the present invention, compared with the case where the configuration of the present invention is not adopted, the attribute of each divided region divided from the image according to the connected component can be determined with high accuracy and at high speed.

1 is a block diagram illustrating a configuration of an image forming apparatus according to an embodiment. FIG. 2 is a block diagram illustrating a functional configuration of an image processing unit of the image forming apparatus. It is a figure which shows the attribute discrimination | determination process of the said image forming apparatus. It is a figure which shows an example of the discrimination criterion used in the said attribute discrimination | determination process.

[Constitution]
FIG. 1 is a block diagram illustrating a configuration of an image forming apparatus 1 according to the present embodiment. As shown in FIG. 1, the image forming apparatus 1 includes a control unit 11, a communication unit 12, a storage unit 13, a display operation unit 14, an image reading unit 15, an image processing unit 16, and an image forming unit 17. And. The control unit 11 includes a CPU (Central Processing Unit), a memory, and the like, and controls each unit of the image forming apparatus 1 when the CPU executes a program stored in the memory or the storage unit 13. The communication unit 12 establishes communication with an external device connected via a communication line, and transmits and receives data. The storage unit 13 is a hard disk, for example, and stores various data. The display operation unit 14 is a touch panel, for example, and displays an image according to the control of the control unit 11 and accepts an operation of the operator. The image reading unit 15 is an example of a reading unit of the present invention, and is, for example, a scanner device. The image reading unit 15 reads an image on a sheet and generates image data corresponding to the image. The image processing unit 16 performs various types of image processing using the image data generated by the image reading unit 15 or the image data received by the communication unit 12. The image forming unit 17 is an example of the image forming unit of the present invention, and is, for example, a printer device that forms an image by an electrophotographic method. The image forming unit 17 forms an image corresponding to the image data processed by the image processing unit 16 on a sheet and outputs the sheet.

  Next, the functional configuration of the image processing unit 16 will be described. FIG. 2 is a diagram illustrating a functional configuration of the image processing unit 16. As shown in the figure, the image processing unit 16 includes, as a functional configuration, an area dividing unit 101, a first feature amount extracting unit 102, a first attribute determining unit 103, an area selecting unit 104, and a contraction process. Unit 105, second feature amount extraction unit 106, and second attribute determination unit 107. The area dividing unit 101 is an example of the area dividing means of the present invention, and divides the image into a plurality of divided areas based on the connected components included in the image. The first feature quantity extraction unit 102 is an example of a first feature quantity extraction unit according to the present invention, and extracts the first feature quantity of each divided region divided by the region division unit 101. For example, a circumscribed rectangle size and a black pixel density are used as the first feature amount. The circumscribed rectangle size means the size of a rectangle circumscribing the divided area. Further, the black pixel density refers to the density of on-state pixels in a divided region. This black pixel density indicates the ratio of pixels belonging to the connected components constituting the divided area occupying the circumscribed rectangle. The first attribute discriminating unit 103 is an example of the first attribute discriminating unit of the present invention, and the first feature amount extracted by the first feature amount extracting unit 102 is a feature of a line drawing or table in which the first feature amount is determined in advance. If it is included in the quantity, the attribute of the divided area having the first feature quantity is discriminated as a line drawing or a table, and the first feature quantity extracted by the first feature quantity extraction unit 102 is determined in advance. If it is included in the feature quantity of the graphic or photograph, the attribute of the divided area having the first feature quantity is discriminated as a graphic or photograph.

  The region selection unit 104 is an example of a region selection unit according to the present invention, and the first feature amount extracted by the first feature amount extraction unit 102 among the plurality of divided regions divided from the image is the first feature amount. In the feature amount space constituted by the feature amounts, a divided region included in a predetermined range is selected from a boundary for discriminating a line drawing or table from a figure or a photograph. The contraction processing unit 105 is an example of the contraction processing unit of the present invention, and performs contraction processing on a binary image in which pixel positions belonging to the connected components constituting the divided region selected by the region selection unit 104 are turned on. Do. The second feature quantity extraction unit 106 is an example of a second feature quantity extraction unit of the present invention, and for the divided areas selected by the area selection unit 104, the second feature quantity extraction unit 106 relates to the thicknesses of the connected components constituting the divided areas. 2 feature values are extracted. As the second feature amount, for example, the number of on-state pixels in the divided area after the contraction processing is performed by the contraction processing unit 105 is used. The second attribute discriminating unit 107 is an example of the second attribute discriminating unit of the present invention, and the connected component constituting the divided region indicated by the second feature quantity extracted by the second feature quantity extracting unit 106 is displayed. The attribute of the divided area is determined according to the thickness.

[Operation]
Next, the operation of the image forming apparatus 1 according to the present embodiment will be described. In the image forming apparatus 1, an attribute determination process is performed in which an image read by the image reading unit 15 is divided into a plurality of divided areas and the attributes of the divided areas are determined. In the following description, this attribute determination process will be described. FIG. 3 is a diagram illustrating attribute discrimination processing. Here, it is assumed that the image reading unit 15 reads the image P0 in the drawing. As shown in the figure, this image P0 includes tables Ta to Tc. Table Ta is a table with a narrow ruled line and a medium size. Table Tb is a table with thick ruled lines and a medium size. The background colors of the tables Ta and Tb are both transparent. Table Tc is a table having a large ruled line interval and a large size. As for the background color of this table Tc, the item portion is colored.

  When the image P0 is read by the image reading unit 15 and image data is generated, the generated image data is input to the image processing unit 16. When the image data is input, the image processing unit 16 performs a binarization process that binarizes the image P0 with the background color using the image data (step S1). Specifically, in the image P0, the image processing unit 16 turns off pixels having pixel values representing the background color and turns on pixels having other pixel values. As a result, as in the binary image P1 in the figure, the image P0 is separated into a base region and a non-base region. Subsequently, the image processing unit 16 divides the binary image P1 into a plurality of divided regions based on the connected components included in the binarized image P1 (step S2). In this example, as shown in the figure, the binary image P1 is divided into a plurality of character areas C and frame areas Ra, Rb, Rc.

  Subsequently, the image processing unit 16 extracts the circumscribed rectangle size and the black pixel density of each divided region as the first feature amount (step S3). Note that the circumscribed rectangle size and the black pixel density are secondarily generated when the region is divided in step S2 described above. In this example, the circumscribed rectangle size “small” and the black pixel density “small” to “large” are extracted as the first feature amount of the character region C. Further, as the first feature amount of the frame region Ra, the circumscribed rectangle size “medium” and the black pixel density “small”, and as the first feature amount of the frame region Rb, the circumscribed rectangle size “medium” and the black pixel density “slightly”. The circumscribed rectangle size “large” and the black pixel density “slightly small” are extracted as the first feature amount of the frame region Rc. The black pixel density of the frame region Ra is “small” because the ruled line in the frame region Ra is thin, and the black pixel density of the frame region Rb is “slightly large” is the ruled line in the frame region Rb. Because it is thick. The reason why the black pixel density of the frame region Rc is “slightly small” is that the frame region Rc includes the item portion of the table Tc.

  Subsequently, the image processing unit 16 performs first attribute discrimination based on the extracted first feature amount (step S4). Specifically, when the extracted first feature amount is included in the range of “character” in the discrimination criterion K shown in FIG. 4, the image processing unit 16 sets the attribute of the divided region to “character”. If it is included in the “line drawing / table” range, the attribute of the divided area is determined as “line drawing / table”. If it is included in the “figure / photo” range, the attribute of the divided area is determined. The attribute is determined as “figure / photo”. This discrimination criterion K indicates a feature amount space constituted by the first feature amount. The range of “character” in the discrimination criterion K indicates a predetermined character feature amount, and the range of “line drawing / table” indicates a predetermined line drawing or table feature amount. The range of “figure / photograph” indicates a predetermined feature amount of a graphic or photograph. That is, when the extracted first feature quantity is included in the predetermined line drawing or table feature quantity, the image processing unit 16 sets the attribute of the area having the first feature quantity as a line drawing or table. If the first feature value that has been determined and included is included in the feature value of a predetermined graphic or photograph, the attribute of the region having the first feature value is determined to be a graphic or a photograph. For example, the circumscribed rectangle size “small” and the black pixel densities “small” to “large”, which are the first feature amounts of the character region C, are included in the range of “character” of the discrimination criterion K shown in FIG. Therefore, the attribute of the character area C is determined as “character”. Further, the circumscribed rectangle size “medium” and the black pixel density “small” which are the first feature amounts of the frame region Ra are included in the range of “line drawing / table” of the discrimination criterion K shown in FIG. Therefore, the attribute of the frame area Ra is determined as “line drawing / table”.

  Further, in the discrimination criterion K, a “boundary portion” is provided between the “line drawing / table” range and the “figure / photo” range. This boundary portion indicates a predetermined range from a boundary for discriminating a line drawing or table from a figure or a photograph in the feature amount space constituted by the first feature amount. For example, the circumscribed rectangle size “medium” and the black pixel density “slightly large”, which are the first feature amounts of the frame region Rb, are included in the “boundary portion” of the determination criterion K shown in FIG. In this case, the attribute of the frame region Rb is not determined here. Further, the circumscribed rectangle size “large” and the black pixel density “slightly small”, which are the first feature amounts of the frame region Rc, are also included in the “boundary portion” of the determination criterion K shown in FIG. Therefore, the attribute of the frame region Rc is not determined here. Here, the reason why the “boundary portion” is provided in the discrimination criterion K will be described. If the “border” is not provided in the discrimination criterion K shown in FIG. 4, the attribute of the frame region Rb described above should be discriminated as “line drawing / table”. / Photo "is misclassified. The attribute of the frame region Rc described above is also erroneously discriminated as “line drawing / table” where it should be discriminated as “figure / photo”. As described above, if an attribute is misidentified, the subsequent process of the attribute determination process is adversely affected. For this reason, a “boundary portion” is provided in the discrimination criterion K, and an attribute is discriminated by the following process for an area that may be erroneously discriminated.

  Subsequently, the image processing unit 16 selects a divided region in which the first extraction amount is included in the range of the “boundary portion” among the plurality of divided regions divided in step S2 described above (step in FIG. 3). S5). In this example, the frame region Rb and the frame region Rc are selected. Subsequently, the image processing unit 16 performs a contraction process on the binary image of the selected divided region (step S6). The contraction process is a process for examining the vicinity of the target pixel and turning off the target pixel if there is even one off pixel. The shrinking process is performed as many times as the ruled line portion included in the divided area disappears. For example, if the maximum thickness of the ruled line portion in the frame regions Rb and Rc is 5 pixels, the ruled line portion disappears when the shrinking process is performed three times using a 3 × 3 shrinking window. Therefore, in this case, the contraction process is performed three times. As a result of the shrinking process, the frame line portions included in the frame regions Rb and Rc disappear, and only the item portion included in the frame region Rc remains in a contracted state. The contraction image rb in the figure shows a binary image of the frame region Rb after the contraction process is performed, and the contraction image rc shows a binary image of the frame region Rc after the contraction process is performed. ing. In this case, the contracted image rb does not include an on-state pixel, and the contracted image rc includes an on-state pixel.

  Subsequently, the image processing unit 16 extracts the number of on-state pixels (hereinafter referred to as “the number of pixels after contraction”) in the divided area after the contraction process is performed as a second feature amount (step S7). . Since the number of pixels after the contraction becomes a value corresponding to the thickness of the binary image in the divided area, it can be said that it is a feature amount related to the thickness of the binary image in the divided area. In this example, since the contracted image rb does not include an on-state pixel, 0 after contraction is extracted as the second feature amount of the frame region Rb. On the other hand, since the contracted image rc includes ON-state pixels, the number of contracted pixels x (x> 0) is extracted as the second feature amount of the frame region Rc.

  Subsequently, the image processing unit 16 performs the second attribute discrimination on the divided region selected in Step S5 described above based on the extracted second feature amount (Step S8). Specifically, when the extracted number of pixels after contraction is 0, the image processing unit 16 determines that the attribute of the divided region is “line drawing / table”, and the extracted number of pixels after contraction is If it is greater than 0, the attribute of the divided area is determined as “figure / photo”. This is because when the number of pixels after contraction is 0, the thickness of the binary image in the divided area is equal to or smaller than the threshold value, and therefore, it can be said that only the ruled line portion is included in the divided area. On the other hand, when the number of pixels after contraction is larger than 0, the thickness of the binary image in the divided area is larger than the threshold value, and therefore it can be said that the divided area includes elements other than ruled lines. Because. That is, when the extracted second feature amount indicates that the thickness of the connected component that constitutes the divided area is equal to or smaller than the threshold, the image processing unit 16 sets the attribute of the divided area as a line drawing or If the extracted second feature quantity indicates that the thickness of the connected component constituting the divided area is larger than the threshold, the attribute of the divided area is determined as a figure or a photograph. To do. In this example, the attribute of the frame region Rb is determined as “line drawing / table” because the number of pixels after contraction is zero. On the other hand, the attribute of the frame region Rc is determined as “figure / photo” because the number of pixels after contraction is x (x> 0).

  As described above, in the attribute determination process described above, the attributes of the frame regions Ra and Rb included in the image P0 are determined as “line drawing / table”, and the attribute of the frame region Rc is determined as “figure / photo”. By identifying the attribute of the frame region Rc as “figure / photograph”, the table Tc including the background color of the item portion can be reproduced. Further, by distinguishing the attributes of the frame regions Ra and Rb as “line drawing / table”, the tables Ta and Tb can be converted into an image format different from that of the table Tc. For example, the tables Ta and Tb can be expressed by vector data. Then, the contents can be easily processed later.

[Modification]
The above is the description of the embodiment, but the contents of this embodiment can be modified as follows. Further, the following modifications may be combined as appropriate.
(Modification 1)
In the above-described embodiment, the number of contracted pixels is extracted as the second feature amount, but the second feature amount is not limited to this. For example, the amount of change in the number of pixels in the ON state of the divided region before and after the contraction process may be used as the second feature amount. In this case, the image processing unit 16 changes the number of on-state pixels in the divided area before and after the contraction process based on the number of on-state pixels in the divided area before the contraction process and the number of pixels after contraction. Calculate the amount. The image processing unit 16 determines that the attribute of the divided area is “line drawing / table” when the calculated change amount is larger than the threshold value, and when the calculated change amount is equal to or less than the threshold value, the divided area. Is determined as “figure / photo”. This is because, when the amount of change in the number of pixels is large, the thickness of the binary image in the divided area is equal to or less than the threshold value, so that it can be said that the divided area includes only the ruled line portion, This is because when the amount of change in the number of pixels is small, the thickness of the binary image in the divided area is larger than the threshold value, so that it can be said that the divided area includes elements other than the ruled line.
Further, the image processing unit 16 may extract the thickness of the binary image of the divided region as the second feature amount.
In this case, the image processing unit 16 performs distance conversion, for example, and extracts the thickness of the binary image. Then, when the thickness of the extracted binary image is equal to or smaller than the threshold value, the image processing unit 16 determines that the attribute of the divided area including the binary image is “line drawing / table” and extracts the extracted 2 When the thickness of the value image is larger than the threshold value, the attribute of the divided area including the binary image is determined as “figure / photo”. In this case, it is not necessary to perform the contraction process described above.

(Modification 2)
In the embodiment described above, the image processing unit 16 may extract the aspect ratio of the divided area as the first feature amount. In this case, when the extracted aspect ratio is included in the feature amount of the predetermined straight line, the image processing unit 16 determines that the attribute of the area having the aspect ratio is “line drawing / table” or “straight line”. .

(Modification 3)
In the embodiment described above, the configuration in which the image processing unit 16 is provided in the image forming apparatus 1 has been described as an example. However, the image processing unit 16 may be provided in another apparatus. For example, the image processing unit 16 may be provided in an image reading apparatus having a function equivalent to that of the image reading unit 15 described above. Further, the image processing unit 16 may be specified as an image processing apparatus that is a single apparatus.

(Modification 4)
In the embodiment described above, the processing performed by the image processing unit 16 may be realized by a single or a plurality of hardware resources, or may be realized by the CPU executing one or a plurality of programs. . The program is provided in a state stored in a computer-readable recording medium such as a magnetic recording medium such as a magnetic tape or a magnetic disk, an optical recording medium such as an optical disk, a magneto-optical recording medium, or a semiconductor memory. obtain. It is also possible to download the program via a communication line such as the Internet.

DESCRIPTION OF SYMBOLS 1 ... Image forming apparatus, 11 ... Control part, 12 ... Communication part, 13 ... Memory | storage part, 14 ... Display operation part, 15 ... Image reading part, 16 ... Image processing part, 17 ... Image forming part, 101 ... Area division part , 102 ... 1st feature quantity extraction unit, 103 ... 1st attribute discrimination unit, 104 ... Area selection unit, 105 ... Shrinkage processing unit, 106 ... 2nd feature quantity extraction unit, 107 ... 2nd attribute discrimination unit .

Claims (9)

Area dividing means for dividing the image into a plurality of divided areas based on the connected components included in the image;
First feature quantity extraction means for extracting a first feature quantity of each divided area divided by the area division means;
When the first feature quantity extracted by the first feature quantity extraction unit is included in the predetermined line drawing or table feature quantity, the attribute of the divided area having the first feature quantity is set to the line drawing or When the first feature value extracted from the first feature value extraction unit is included in the feature value of a predetermined figure or photograph, the divided area having the first feature value is determined. First attribute determination means for determining the attribute of
Among the plurality of divided areas, the first feature quantity extracted by the first feature quantity extraction unit is a line drawing or table and a figure or photograph in a feature quantity space constituted by the first feature quantity. Area selection means for selecting a divided area included in a predetermined range from the boundary to be determined;
Second feature quantity extraction means for extracting a second feature quantity related to the thickness of the connected component constituting the divided area for the divided area selected by the area selection means;
Second attribute discrimination means for discriminating the attribute of the divided area according to the thickness of the connected component constituting the divided area indicated by the second feature quantity extracted by the second feature quantity extracting means; An image processing apparatus comprising: In the case where the second attribute determination unit indicates that the second feature amount extracted by the second feature amount extraction unit indicates that the thickness of the connected component constituting the divided region is equal to or less than a threshold value. The image processing apparatus according to claim 1, wherein the attribute of the divided area is determined as a line drawing or a table. When the second attribute determination unit indicates that the second feature amount extracted by the second feature amount extraction unit indicates that the thickness of the connected component constituting the divided region is larger than a threshold value The image processing apparatus according to claim 1, wherein the attribute of the divided area is determined as a graphic or a photograph. The first feature quantity extraction means, as the first feature quantity, indicates a size of a circumscribed rectangle of each divided region and a ratio indicating that a pixel belonging to a connected component constituting the divided region occupies the circumscribed rectangle. Extract the density and
The said 1st attribute discrimination | determination means performs the said discrimination | determination using the magnitude | size of the said circumscribed rectangle and the said pixel density which were extracted by the said 1st feature-value extraction means. The image processing apparatus according to any one of the above. A contraction processing unit that performs contraction processing on a binary image in which pixel positions belonging to a connected component constituting a divided region selected by the region selection unit are turned on;
The second feature amount extraction unit extracts, as the second feature amount, the number of on-state pixels in the divided area after the contraction processing is performed by the contraction processing unit,
The image according to any one of claims 1 to 4, wherein the second attribute determining unit performs the determination based on the number of pixels extracted by the second feature amount extracting unit. Processing equipment. A contraction processing unit that performs contraction processing on a binary image in which pixel positions belonging to a connected component constituting a divided region selected by the region selection unit are turned on;
The second feature amount extraction unit extracts, as the second feature amount, an amount of change in the number of on-state pixels in the divided region before and after the contraction process,
5. The image according to claim 1, wherein the second attribute determination unit performs the determination based on a change amount extracted by the second feature amount extraction unit. Processing equipment. The first feature amount extraction unit extracts an aspect ratio of each of the divided regions as the first feature amount,
When the aspect ratio extracted by the first feature amount extraction unit is included in a predetermined straight line feature amount, the first attribute determination unit displays the attribute of the divided region having the aspect ratio as a line drawing. The image processing apparatus according to claim 1, wherein the image processing apparatus is identified as a table or a straight line. Reading means for reading an image of a document;
An image reading apparatus comprising: the image processing apparatus according to claim 1, which performs processing using an image read by the reading unit. Reading means for reading an image of a document;
The image processing apparatus according to claim 1, wherein processing is performed using an image read by the reading unit;
An image forming apparatus comprising: an image forming unit that forms an image processed by the image processing apparatus on a recording material.

Images