JP2005208981A - Featured value extraction device, method for extracting featured value, and document filing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve problems that when pixel values are accumulated as they are in a projection direction, the ground color of paper, the background colors of an original, image components inserted into a document, ruled line components, etc. are also accumulated, character components in a projected waveform can not be clearly expressed and featured values are not correctly extracted. <P>SOLUTION: In a document filing device for extracting featured values such as character size and a line interval on a character image by using a projected waveform and managing document data in each document on the basis of the extracted featured values, whether respective pixels on a projection line are present in a character area or not is estimated in forming a projected waveform and whether the pixel values are to be accumulated or not is controlled in accordance with the estimated result, so that the influence of disturbance such as the ground color of paper, the background colors of the original, image components inserted into the document, and ruled line components is removed. Featured values are extracted by using the projected waveform generated by removing the influence of disturbance and the section of each document is decided from the extracted featured values. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a feature amount extraction device, a feature amount extraction method, and a document filing device, and more particularly to a feature amount extraction device that extracts feature amounts such as document layout, character size, and line spacing using a projected waveform of a document image. The present invention also relates to a document filing apparatus that determines a delimiter position of a document unit by using the feature quantity extraction method, and manages document data for each document.

  In recent years, when a document composed of one or more paper originals is efficiently digitized over a plurality of copies, a process of continuously reading the originals using a scanner device having an automatic paper feed function has been widely performed. . At this time, in order to manage the read document image data (document data) for each document, it is necessary to detect a break between the documents using some method.

  In order to detect a break between documents, conventionally, a character recognition area portion set in advance is cut out from image data read by a scanner device, and character recognition is performed. Based on the result of character recognition, The document separation is determined (for example, see Patent Document 1), a plurality of documents are read at once by a scanner device having an automatic paper feed function, and the feature amount of the read original image is calculated, and this calculation is performed. The document unit break is determined based on the feature amount (see, for example, Patent Document 2).

  In any of these conventional techniques, document separation is automatically determined without inserting a document separation document (for example, a blank document) between each document in advance or modifying the document to be document separation. Therefore, the burden on the user can be greatly reduced. In particular, with the conventional technique according to Patent Document 2, there is no restriction on the format of the target document, and it is possible to deal with a wider variety of documents. On the other hand, the prior art according to Patent Document 1 can only deal with a document of a specific format.

  By the way, in the prior art according to Patent Document 2, as examples of the feature quantity that is the basis for the delimiter position determination, the hue used for the image data, the number of screen lines, the layout of the document, the direction of character combination, etc. . However, in general office documents, there are many text-only originals that do not include images, and thus feature quantities such as the hue of the image and the number of screen lines cannot often be used to determine the separation position. In addition, since horizontal writing is common, the direction in which the characters are combined is often not useful for determining the document break position.

  On the other hand, feature quantities such as document layout, character size, and line spacing are generally different for each office document. Therefore, it is an effective technique to use these feature amounts for determination of the document break position. As a method for extracting feature quantities such as document layout, character size, and line spacing from a document image, it is common to form a projection waveform in the vertical and horizontal directions of the document image and use the projection waveform. .

  Specifically, the document image is binarized, projected in the vertical and horizontal directions, and a threshold value process is performed on these projection waveforms to determine a character area / blank area. The size and line spacing are detected (for example, see Patent Document 3). Also, binarization of the document image is performed, and then a line is cut out by projecting in the horizontal direction, and then each character is separated by projecting in the vertical direction in the image of each separated line. The character size is detected by correcting the separation of the characters by taking the horizontal projection again (see, for example, Patent Document 4).

Japanese Patent Laid-Open No. 10-21380 Japanese Patent Laid-Open No. 2002-24258 Japanese Patent Publication No. 7-1111738 JP-A-5-89283

  As described above, in each of the related arts according to Patent Documents 3 and 4 using the projection waveform, as shown in FIG. 12, the read document image is first binarized, and the projection waveform for the binarized image is obtained. The feature amount is extracted based on the projected waveform. Therefore, the processing result is greatly affected by the binarization threshold. However, even if the binarization threshold value is set to an appropriate value, the binarization threshold value still depends on the background color of the paper, the background color of the document, the presence / absence of an image inserted in the document, the character color / density. Therefore, it is difficult to obtain a stable processing result.

  As shown in FIG. 13, the above-described problem of the binarization level can be solved by taking a multi-value projection without binarization. However, if the background color of the paper or the background color of the document is dark, the image is inserted in the document, the character density is low, or the line is short, etc. The difference in projection is not clear. Further, in the case of a document image containing a lot of low density noise, if binarization processing is performed, the noise is determined to be a white level, and as a result, the influence of noise can be eliminated. However, if projection is performed with multiple values, noise components are also accumulated, which makes processing difficult in this respect.

  That is, in the conventional technology that uses a projection waveform to detect a feature amount such as a character size or line spacing on a character image, if the pixel values are accumulated as they are in the projection direction, the background color of the paper or the background color of the document Since the image components and ruled line components inserted in the document are also accumulated, the character components in the projected waveform are not clear. As described above, when the character component in the projected waveform is not clear and it is unclear which part of the projected waveform corresponds to the character part, the process of detecting the feature amount such as the character size and the line spacing from the projected waveform. Cannot be performed correctly.

  The present invention has been made in view of the above problems, and the object of the present invention is to influence the influence of disturbances such as the background color of the paper, the background color of the document, the image component inserted in the document, and the ruled line component. To provide a feature amount extraction device, a feature amount extraction method, and a document filing device using the feature amount extraction device, which can correctly extract feature amounts such as character size and line spacing without receiving them. .

  In order to achieve the above object, in the present invention, when forming a projection waveform, it is estimated whether each pixel on the projection line exists in the character area, and only about the pixel estimated to exist in the character area. The pixel values are accumulated to form a projection waveform. Then, a feature amount is extracted by comparing the formed projection waveform with a predetermined threshold value, or a document unit break is determined based on the extracted feature amount.

  When extracting feature quantities such as character size and line spacing on the character image using the projection waveform, the pixel values are accumulated according to the estimation result of whether each pixel on the projection line exists in the character area. By controlling whether or not, the influence of disturbance such as the background color of the paper, the background color of the document, the image component inserted in the document, and the ruled line component is eliminated. Then, using the projection waveform generated by eliminating the influence of disturbance, feature quantities such as character size and line spacing are extracted, and a document unit break is determined from the extracted feature quantities.

  According to the present invention, when forming a projection waveform, it is estimated whether or not each pixel on the projection line exists in the character area, and whether or not the pixel value is accumulated is controlled according to the estimation result. This eliminates the effects of disturbances such as paper background color, document background color, image components inserted into the document, and ruled line components, so that the character size and line spacing are not affected by disturbances. And the like can be correctly extracted, and document segment delimitation can be correctly performed.

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

  FIG. 1 is a block diagram illustrating a configuration example of a document filing apparatus according to an embodiment of the present invention. As is apparent from FIG. 1, the document filing apparatus according to the present embodiment includes a document input unit 11, a feature amount extraction unit 12, a similarity evaluation unit 13, a document storage unit 14, a document delimiter unit 15, and a document output unit 16. These components are connected to each other via a bus line 17. In the document filing apparatus having such a configuration, the feature quantity extraction unit 12 is the most characteristic part of the present invention.

  The document input unit 11 acquires document data from the input document and inputs it as document data to be registered in the document filing apparatus. Here, the input document data includes, for example, multi-value image data obtained by scanning from a printed material, or multi-value image data captured by a digital camera. Correspondingly, as the document input unit 11, for example, a scanner device provided with an ADF (Auto Document Feeder) and its control means, or a memory for continuously extracting images stored in a memory (card) of a digital camera A reader device and its control means are used.

  The feature amount extraction unit 12 extracts from the document data (multi-valued image data) input by the document input unit 11 an amount indicating a feature unique to the document data (hereinafter referred to as “feature amount”). In the feature quantity extraction unit 12 according to this example, as shown in FIG. 2, a projection waveform is formed by accumulating pixel values of each document data in the horizontal and vertical directions, and the height of a character representing the document data is determined from the projection waveform. For example, n pieces of height h and width w are extracted.

  As described in the description of the prior art, when the pixel values of the document image are accumulated as they are, the projected waveform includes the paper ground color, the background color of the document, the image in the document, the character density, the line length, and the like. An effect appears (see FIG. 13). In order to stably extract the characteristics related to the height h and width w of the character from the projection waveform, it is desirable to eliminate as much as possible the influence of components other than the character. Therefore, in the feature quantity extraction unit 12 according to this example, when forming a projection waveform, for each pixel on the projection line, it is estimated whether the pixel exists in the character region, and is estimated to exist in the character region. Only for the selected pixels, the pixel values are accumulated to form a projection waveform.

FIG. 3 shows an example of the configuration of a part in the document image. The document image shown in FIG. 3 includes a solid paint part 32 and a halftone dot image part 33 in addition to the area of the character part 31. Therefore, the projection waveform p (y) obtained by simply accumulating the pixel values includes the influence of the paper ground color, the solid coating portion 32, and the halftone image portion 33. The projection waveform p (y) can be expressed as the following formula (1). F (x, y) in Expression (1) represents a pixel value at the position (x, y).
p (y) = Σx {f (x, y)} (1)

  Here, the influence of the solid coating portion 32 and the halftone dot image portion 33 is eliminated from the projection waveform p (y) by using the property unique to the character portion 31. In the feature quantity extraction unit 12 according to this example, it is assumed that the region of the character part 31 is determined by paying attention to the degree of dispersion of pixel values on the projection line, and in particular, dispersion is used as the degree of dispersion.

  FIG. 4 shows the pixel value f (x, a) on the projection line indicated by the straight line y = a. When the pixel value variance v (x) in the interval [x−A / 2, x + A / 2] is obtained for an arbitrary position x in the pixel value f (x, a), the result is as shown in FIG. . Here, A is a predetermined specific section width.

  As can be seen from FIG. 5, the dispersion of the solid paint portion 32 is small except for both ends of the region, the dispersion of the shaded portion 33 is large, and the character portion 31 is intermediate between the solid paint portion 32 and the shaded portion 33. Accordingly, in the distribution of pixel values on the projection line, if only the pixel values of the region having a medium dispersion value are selectively accumulated, as shown in FIG. The projection waveform p1 (y) that better reflects the character component can be obtained by eliminating the influence.

The projection waveform p1 (y) obtained in this way can be expressed as the following equation (2).
p1 (y) = Σx {Th (v (x)) · f (x, y)} (2)
In the above equation (2), Th (x) is a threshold function.

The threshold function Th (x) selects only pixels that are estimated to exist in the character area, as shown in the following equation (3). In the following equation (3), vl and vh are threshold values corresponding to the range of the dispersion value in the region of the character part 31, as shown in FIG.
Th (v) = 1 (vl ≦ v ≦ vh),
Th (v) = 0 (v <vl or vh <v) (3)

  FIG. 7 is a block diagram illustrating an example of a specific configuration of the projection waveform forming unit in the feature quantity extracting unit 12. The projection waveform forming unit according to this example includes a dispersion degree detection unit 21, a threshold processing unit 22, and an accumulation processing unit 23, and is configured to process input image data in units of lines.

  The spread degree detection unit 21 detects the spread degree of image data input in units of lines. The threshold value processing unit 22 compares the detection result of the distribution degree detection unit 21 with the threshold values vl and vh (see FIG. 5). If the detected distribution degree is between the threshold values vl and vh, a logical “1” is obtained. "" Is output, otherwise logic "0" is output. The accumulation processing unit 23 accumulates pixel values when the output of the threshold processing unit 22 is logic “1”, and outputs it as a projection waveform pl (y).

  In the projection waveform forming unit according to the present example, the pixel values are accumulated only for the pixels estimated to exist in the character area, and the pixel values are not accumulated for other pixels. As an example, the binary accumulation method is used. However, it is also possible to adopt a method of accumulating after weighting the pixel values according to the character area likeness.

  By comparing the projection waveform p1 (y) thus obtained with an appropriate threshold value ty as shown in FIG. 6, an area larger than the threshold value ty is a black area, and a small area is a white area. Can be determined. The same processing as in the above procedure may be performed for the vertical projection lines, and the projection waveform taken in the vertical direction may be obtained to determine the white region / black region.

  Then, from the determination result of the white region / black region, as shown in FIG. 8, the binary data series is output as the binarization result of the projection waveform data, and the length of the continuous black region is obtained for all, For example, in these values, n values centered on the peak of the appearance frequency are set as the feature amounts of the horizontal projection of the document data.

  Note that the feature amount is not limited to the length of the black region, and a portion determined to be a white region can be regarded as a line interval or a character interval, and these values can be included in the feature amount. Furthermore, since the line spacing is generally larger than the character spacing, the direction in which the white spacing is large may be added to the feature amount as the character set direction.

  Referring back to FIG. 1, the document storage unit 14 stores and stores the input document data in association with the n types of feature values extracted by the feature value extraction unit 12, and includes a hard disk drive and a DVD (Digital Versatile Disc). -Realized by a mass storage device such as a RAM / ± RW / ± R drive.

  The similarity evaluation unit 13 extracts a plurality of feature amounts if a plurality of feature amounts are accumulated for the feature amounts extracted by the feature amount extraction unit 12 and associated with the document data and accumulated in the document accumulation unit 14. Are respectively compared to obtain the similarity between them. The similarity referred to here is, for example, the Euclidean distance between feature vectors associated with each document data if the feature value is a vector expression (hereinafter, this feature value is referred to as “feature vector”). It is a degree evaluated based on.

  Specifically, as shown in FIG. 9A, the case where the Euclidean distance between feature vectors is smaller than a predetermined threshold is evaluated as similarity: large, and as shown in FIG. 9B. The case where the Euclidean distance between feature vectors is equal to or greater than the predetermined threshold is evaluated as similarity: small. However, the definition of the distance between feature vectors is not limited to the Euclidean distance.

  The document delimiter 15 obtains a similarity between a series of document data by the similarity evaluation unit 13 and puts a delimiter into the series of document data based on the obtained similarity. As a method of accumulating in document units (consisting of document data of one page or more), for example, a method of creating a file folder for each document and storing corresponding document data with a file name having a sequential number in the input order Alternatively, there is a method of using an image file format such as a multi-page TIFF (Tagged Image File Format) that can hold a plurality of pages.

  The document output unit 16 outputs document data instructed to be output in a predetermined format. For example, a CRT (Cathode Ray Tube) and its control means, a printer device and its control means, a magnetic disk, a memory card, etc. The read / write device and its control means, or a data transfer device that exchanges data via a network or the like. That is, the document output unit 17 outputs, for example, a document printed on paper, image data output to a CRT, or a file formatted by HTML (Hyper Text Markup Language) as an output result.

  Next, the procedure of document data separation processing in the document filing apparatus according to this embodiment having the above-described configuration will be described with reference to the flowchart of FIG.

  For example, the user sets a paper document including one page or more on the ADF. At this time, the paper document may be either a single document (consisting of one or more pages) or a plurality of documents (comprising one or more pages). Also, when setting a paper document, the user need not be aware of document separation. The paper document set in the ADF is sent to the scanner device page by page by the ADF. At this time, the scanner device functions as the document input unit 11 in FIG. That is, from the document input unit 11, the same number of document data as the number of pages of the paper document set in the ADF is input to the document filing device.

  When document data is input from the document input unit 11 (step S11), whether the input document data has already been extracted with a feature amount and is stored in the document storage unit 14 in association with the feature amount. It is determined whether or not (step S12). If the input document data is document data for which feature amounts have not yet been extracted, the feature amount is extracted from the input document data by the feature amount extraction unit 12 (step S13), and then the process proceeds to step S14. To do. If the input document data is already extracted document data, the process directly proceeds to step S14. That is, the process for extracting the feature amount is not performed on the document data in which the feature amount has already been extracted from the input document data and is stored in the document storage unit 14 in association with the feature amount.

  The above processing is performed for all input document data (step S14). Subsequently, the feature amount associated with the document data stored in the document storage unit 14 and the feature amount associated with the input document data are compared with each other with respect to the document data stored in the document storage unit 14. Evaluation is performed by the evaluation unit 13 (step S15). Then, based on the evaluation result, a document unit separation result is determined for the document data, separated into document units, and stored in the document storage unit 14 (step S16).

  In the above-described embodiment, an example has been described in which the feature amount extraction unit 12 directly processes input multi-value document data and performs character region determination using variance as a distribution of pixel values. However, the present invention is not limited to this.

  That is, the feature quantity extraction unit 12 does not process the input multi-value document data as it is, but binarizes it in advance and converts it into image data consisting of only white and black pixels, and further distributes the dispersion of pixel values. Instead of the degree, it is also possible to determine the character area by using the number of subsequent black pixels, that is, the run length, as the dispersion degree for each black pixel on the projection line. In this case, in FIG. 3, the run length of the solid paint portion 32 is large, the run length of the halftone image portion 33 is small, and the run length of the character portion 31 is intermediate between the solid paint portion 32 and the halftone image portion 33.

  In the above embodiment, the case where the threshold function Th (v) is used for the character region determination has been described as an example. However, the weighting function w (v) is used instead of the threshold function Th (v). It is also possible to use. This will be specifically described below.

  The threshold function Th (v) in Expression (2) is a function that takes only two types of values “0” or “1”. That is, the threshold function Th (v) plays a role of selecting all pixels on the projection line as pixels in the character area. The two threshold values vl and vh in FIG. 5 are criteria for selecting whether or not a certain pixel is in the character area. Accordingly, the shape of the obtained projection waveform pl (y) is greatly influenced by these two threshold values vl and vh. However, since the variance v (x) of pixel values in the character area varies greatly depending on the font and character size, it may be difficult to set the two threshold values vl and vh appropriately.

  Therefore, it is possible to easily set the two threshold values vl and vh by using a weighting function w (v) as shown in FIG. 11 instead of the threshold function Th (v) in the equation (2). Can do. In the weight function w (v), as is clear from FIG. 11, the weight changes gently in the vicinity of the two threshold values vl and vh. Therefore, for example, even if the threshold value vl is slightly changed to become vl + α, the value w (vl + α) of the weight function w (v) does not change greatly. Therefore, the shape of the obtained projection waveform is not greatly affected by the two threshold values vl and vh.

The projection waveform p2 (y) obtained using the weight function w (v) can be expressed as the following equation (4).
p2 (y) = Σx {w (v (x)) · f (x, y)} (4)

  In the above embodiment, the feature amount extraction unit 12 extracts the image feature amount using the pixel value dispersion degree as the character-specific property, but the frequency characteristic of the pixel value as the character-specific property. Can also be used.

  As can be seen from FIG. 4, the frequency of the solid fill portion 32 is small (low), the frequency of the halftone image portion 33 is large (high), and the character portion 31 is intermediate between the solid fill portion 32 and the halftone dot image portion 33. Therefore, in the pixel value distribution on the projection line, if only the region having a medium frequency is selectively accumulated, the influence of the solid paint portion 32 and the dot image portion 33 is eliminated, and the character component is reflected better. A projection waveform can be obtained.

In order to extract the image feature amount using the frequency characteristic of the pixel value as a property unique to the character part, a band pass filter that passes only the frequency band corresponding to the character part 31 with respect to the pixel value on the projection line. Apply. The projection waveform p3 (y) obtained through this band pass filter can be expressed as the following equation (5).
p3 (y) = Σx {BPF (f (x, y)) (5)
In the above equation (5), BPF (x) is a band pass filter, and the pass band of the BPF (x) is a frequency band corresponding to the character part 31.

It is a block diagram which shows the structural example of the document filing apparatus which concerns on one Embodiment of this invention. It is a conceptual diagram of the feature-value extraction in a feature-value extraction part. It is a figure which shows an example of a structure of the part in a document image. It is a figure which shows the pixel value f (x, a) on the projection line shown by the straight line y = a. It is a figure which shows dispersion | distribution v (x) of the pixel value calculated | required about arbitrary positions x in pixel value f (x, a). It is a figure which shows the projection waveform p1 (y) when only the area | region which has a medium dispersion | distribution value selectively accumulates in distribution of the pixel value on a projection line. It is a block diagram which shows an example of the specific structure of the projection waveform formation part in a feature-value extraction part. An example in which the lengths of continuous black areas are all obtained from the determination results of white areas and black areas, and n values centering on the peak of the appearance frequency are used as the feature values of the horizontal projection is shown. FIG. It is a conceptual diagram of similarity evaluation. It is a flowchart which shows the procedure of the division | segmentation process of document data. It is a characteristic view of the weight function w (v) used for determination of a character area. It is FIG. (1) with which it uses for description of the subject of a prior art. It is FIG. (The 2) with which it uses for description of the subject of a prior art.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11 ... Document input part, 12 ... Feature quantity extraction part, 13 ... Similarity evaluation part, 14 ... Document storage part, 15 ... Document delimiter part, 16 ... Document output part, 21 ... Scatter degree detection part, 22 ... Threshold value Processing unit, 23... Cumulative processing unit

Claims (7)

A projection waveform forming means for estimating whether or not each pixel on the projection line is present in the character region, and accumulating pixel values only for the pixels estimated to be present in the character region to form a projection waveform;
A feature quantity extraction apparatus comprising: extraction means for extracting a feature quantity by comparing the projection waveform formed by the projection waveform forming means with a predetermined threshold value. The feature amount extraction apparatus according to claim 1, wherein the projection waveform forming unit estimates whether or not the character area is present based on a distribution degree of the pixel values on the projection line. The projection waveform forming means obtains a variance of pixel values in a predetermined section for the pixel values on the projection line, and selectively accumulates only pixel values in a region having a predetermined variance value. The feature amount extraction apparatus according to claim 2. The projection waveform forming means binarizes the input multi-valued image data and converts it into image data consisting of only white pixels and black pixels, and determines the number of black pixels following each black pixel on the projection line. The feature quantity extraction device according to claim 2, wherein the feature quantity extraction device is used as the degree of dispersion. The feature amount extraction apparatus according to claim 1, wherein the projection waveform forming unit estimates whether or not the character waveform area exists based on a frequency characteristic of the pixel value on the projection line. A first step of estimating whether or not each pixel on the projection line is present in the character area, and accumulating pixel values only for the pixels estimated to be present in the character area to form a projection waveform;
And a second step of extracting a feature amount by comparing the projection waveform formed in the first step with a predetermined threshold value. A projection waveform forming means for estimating whether or not each pixel on the projection line is present in the character region, and accumulating pixel values only for the pixels estimated to be present in the character region to form a projection waveform;
Extraction means for extracting a feature amount by comparing the projection waveform formed by the projection waveform forming means with a predetermined threshold;
A document filing apparatus comprising: a determination unit that determines a break of a document unit based on the feature amount extracted by the extraction unit.