Image processing device

Abstract

To properly extract and exclude unnecessary parts such as headers and footers even when there is an error in the reading position during scanning or an OCR misjudgment.SOLUTION: An image processing device includes extraction means 901 that extracts objects from the image of each page of a document, analysis means 902 that uses an image of one page of the document as a reference page and other pages as analysis pages, and analyzes each object on the reference page and each object on the analysis page, and determination means 903 that determines whether each object on the reference page is a necessary part or an unnecessary part based on the result of the analysis means 902.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an image processing device. [Background technology]

[0002] 2. Description of the Related Art Printing devices that read documents from paper media using a scanner or the like and print based on the read document data are well known.

[0003] In some cases, the printing device reads a multi-page document from a paper medium and digitizes the document data for reuse. When reusing the document data in this way, the header and footer information present on each page is often unnecessary.

[0004] For example, Patent Document 1 describes that an image processing device performs a similarity determination for each page after the first page, searches for similar header images that are similar to the reference header image and similar footer images that are similar to the reference footer image, and deletes the image data of pages from the image data of each page after the first page that contain similar footer images and whose areas other than the area where the similar footer images are present are blank. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-134647 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in Patent Document 1, when performing the similarity determination, color histograms are generated for each of the reference image and the target image, and a color similarity is calculated that indicates the degree of similarity between the color histograms of the reference image and the target image. The image processing unit then determines that the predetermined condition is not met if the color similarity is at least equal to or less than a predetermined threshold (for example, 80% to 90%). However, simply extracting the header image and footer image based on the color similarity may make it difficult to properly extract the header image and footer image, for example, if there is an error in the reading position during scanning or an OCR determination error.

[0007] The present invention has been made in consideration of these problems, and aims to provide an image processing device that can properly extract and properly remove unnecessary parts such as headers and footers, even if there is an error in the reading position during scanning or an OCR judgment error. [Means for solving the problem]

[0008] In order to achieve the above object, the image processing device according to the present invention is characterized by: an extraction means for extracting an object from an image of each page of a document; an analysis means for analyzing an image of one page of a document as a reference page and other pages as analysis pages, and analyzing each object on the reference page and each object on the analysis page; a determining means for determining whether each object on the reference page is a necessary part or an unnecessary part according to the result of the analyzing means; The reason is that it is equipped with the following. [Effects of the Invention]

[0009] According to the features of the image processing device of the present invention, unnecessary parts such as headers and footers can be appropriately extracted and removed. [Brief explanation of the drawings]

[0010] [Figure 1]1 is a diagram showing a schematic configuration of a printer network system according to a first embodiment of the present invention. [Figure 2] 1A and 1B are explanatory diagrams illustrating an object extraction process performed by an analysis unit included in the inkjet printing device 1 according to the first embodiment of the present invention, where (a) shows image data of the first page of the image data, and (b) shows image data of the second page of the image data. [Figure 3] 3 is an explanatory diagram for schematically explaining the actions of an analysis unit and a determination unit provided in the inkjet printing apparatus according to the first embodiment of the present invention. FIG. [Figure 4] FIG. 10 is an explanatory diagram for schematically explaining the actions of an analysis unit and a determination unit provided in the inkjet printing apparatus according to the modified example of the first embodiment of the present invention. [Figure 5] 4 is a flowchart showing the processing content of the inkjet printing apparatus according to the first embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The same or equivalent parts and components are designated by the same or equivalent reference numerals throughout the drawings. However, it should be noted that the drawings are schematic and may differ from the actual product. Furthermore, the drawings may include parts with different dimensional relationships and ratios.

[0012] Furthermore, the embodiments shown below are merely examples of devices that embody the technical concept of the present invention, and the technical concept of the present invention does not limit the arrangement of each component to that shown below. Various modifications can be made to the technical concept of the present invention within the scope of the claims.

[0013] Example 1 As shown in FIG. 1, the printer network system according to the first embodiment of the present invention includes an inkjet printing apparatus 1 and a client terminal 14.

[0014] Each client terminal 14 is configured by a PC (personal computer) or the like. This client terminal 14 has a CPU 16 that executes various processes based on a control program stored in a ROM 17. The CPU 16 is also connected to a RAM 18 that functions as a working area, an input unit 19 that includes a keyboard, a mouse, etc., an output unit 20 that includes a liquid crystal display, etc., and an external storage device 21.

[0015] The external storage device 21 has a storage area for application programs for generating image data including printed images such as documents and pictures, a storage area for the printer driver program of the inkjet printing device 1, and a storage area for various other application programs.

[0016] The CPU 16 is an arithmetic unit configured with a processor, memory, and other peripheral devices, and starts an application program in the external storage device 21 in response to a start request input from the input unit 19. The CPU 16 also generates, on the started application program, original information indicating an image specified by parameters input from the input unit 19, etc. The generated original information is displayed on the output unit 20, and is stored in the external storage device 21 when a save request is input from the input unit 19.

[0017] When a print request for document information generated by a document application program or the like is input, CPU 16 executes a printer driver program on CPU 16, thereby virtually constructing printer driver 110 on CPU 16. Then, printer driver 110 causes inkjet printing device 1 to output a print job.

[0018] The printer driver 110 includes a print control program, acquires image data generated by an original application, etc., and generates a print job. The generated print job is then output from the external interface unit 15 to the external interface unit 11 of the control unit 10.

[0019] The inkjet printing device 1 is a line printer in which an inkjet head is arranged in a direction perpendicular to the paper transport direction, and printing is performed by ejecting ink from the nozzles of the inkjet head onto the transported paper.

[0020] 1, the inkjet printing device 1 includes a scanner unit 101 that reads image data from an original, a printer unit 102 that prints (records) a print image on a recording paper (one side or both sides) based on the image data read by the scanner unit 101 or the image data included in the print job, and a control unit 10 for overall control. When image data is read from the scanner unit 101, the scanner unit 101 acts as an acquisition unit, and when the print job is generated by the client terminal 14, the external interface unit 11 acts as an acquisition unit.

[0021] Furthermore, external interface units 15 of a plurality of client terminals 14 (only one representative client terminal 14 is shown in FIG. 1) are connected to external interface unit 11 of control unit 10 via a local area network LAN. Control unit 10 receives print jobs from each client terminal 14. Inkjet printing device 1 executes print processing in printer unit 102 under conditions specified in the print job.

[0022] Furthermore, an operation panel 80 is connected to the control unit 10. This operation panel 80 is disposed on top of the inkjet printing apparatus 1.

[0023] 1, the control unit 10 of the inkjet printing device 1, which causes the printer section 102 to perform a printing operation, includes a CPU 90. This CPU 90 controls the operations of the scanner section 101 and the printer section 102 in accordance with the contents input and set from the operation panel 80, based on the programs and setting information stored in the ROM 91.

[0024] The control unit 10 is provided with a RAM 92, which temporarily stores various settings such as print jobs input to the control unit 10 from the client terminal 14, the number of copies input from the operation panel 80, and the resolution when scanning by the scanner unit 101.

[0025] Furthermore, the control unit 10 is provided with an external storage device 93, which is connected to the CPU 90. The external storage device 93 has an area for storing print jobs transferred from the above-mentioned RAM 92. The external storage device 93 also stores various types of data, etc.

[0026] The CPU 90 functionally comprises an extraction means 901 , an analysis means 902 , and a determination means 903 .

[0027] The extraction means 901 extracts objects from image data read by the scanner unit 101 or image data acquired by the external interface unit 11 via the local area network LAN. It is assumed that the image data includes multiple pages.

[0028] Fig. 2 is an explanatory diagram illustrating an object extraction process performed by extraction unit 901 included in inkjet printing apparatus 1 according to the first embodiment of the present invention. Fig. 2(a) is a diagram showing image data D10 of the first page of the image data, and Fig. 2(b) is a diagram showing image data E20 of the second page of the image data. Fig. 3 is an explanatory diagram that schematically illustrates the operations of analysis unit 902 and determination unit 903 included in inkjet printing apparatus 1 according to the first embodiment of the present invention.

[0029] Image data D10 shown in FIG. 2(a) is an analysis page manuscript to be analyzed, and this image data D10 includes objects D11 to D14. Image data E20 shown in FIG. 2(b) is a reference page manuscript to be referenced, and this image data E20 includes objects E21 to E24. Note that in the case of image data read by the scanner unit 101, or in the case of a print job generated by the client terminal 14 that is not divided into objects but is only image data, the objects are extracted and divided. Furthermore, objects containing characters are subjected to OCR processing and converted into text data. In the case of a print job generated by the client terminal 14 that has been divided into objects in advance, only necessary processing may be performed, and unnecessary processing may be omitted.

[0030] The extraction means 901 extracts objects D11 to D14 from the image data D10, and also extracts objects E21 to E24 from the image data E20.

[0031] The analysis means 902 sets an image of one page of the document as a reference page and the other pages as analysis pages, and analyzes each object on the reference page and each object on the analysis page. Specifically, the determination means 92 calculates the content difference, coordinate difference, and size difference between the object on the reference page and the object on the analysis page.

[0032] The determination means 903 determines whether each object on the reference page is a necessary part or an unnecessary part according to the result of the analysis means 902. Specifically, the determination means 903 determines that the two objects are a header or a footer when the calculated content dissimilarity, coordinate difference, and size difference are all smaller than threshold values. Appropriate values ​​are set in advance for each threshold value to determine whether the object is a header or a footer.

[0033] In the examples shown in FIGS. 2(a) and 2(b), the object D11 of the image data D10 and the object E21 of the image data E20 are image objects. When comparing them, since they are the same logo image, the determination means 903 determines that the content difference degree calculated by these pattern matching processes is smaller than the threshold value. Also, the determination means 903 determines that the difference in the position coordinates and object size of these objects is also smaller than the threshold value. Thus, since all of the content difference degree, coordinate difference, and size difference are smaller than the threshold value, the determination means 92 classifies both the object D11 and the object E21 as header objects.

[0034] Also, the object D12 of the image data D10 and the object E22 of the image data E20 are text objects. When comparing them, if the difference degree between the two character strings is calculated using the normalized Levenshtein distance, since they are the same character string, it becomes 0.00. Even if, for example, due to OCR misrecognition, the character "工" in "工業" is recognized as the katakana character "エ", the Levenshtein distance is 0.10, and it can still be suppressed to a small value. Therefore, the determination means 903 determines that the content difference degree is smaller than the threshold value. Also, the determination means 903 determines that the difference in the position coordinates and object size of these objects is also smaller than the threshold value. Thus, since all of the content difference degree, coordinate difference, and size difference are smaller than the threshold value, the determination means 903 classifies both the object D12 and the object D22 as header objects.

[0035] Object D13 of image data D10 and object E23 of image data E20 are text objects, and when these are compared, the dissimilarity between the two character strings calculated using the standardized Levenshtein distance is 0.94, which means that the character strings are clearly different, so the determination unit 903 determines that the content dissimilarity is greater than the threshold. The determination unit 903 also determines that the differences in the position coordinates and object size of these objects are also greater than the threshold. As a result, the determination unit 903 classifies both object D13 and object E23 as not being header objects.

[0036] Object D14 of image data D10 and object E24 of image data E20 are text objects, and when comparing these, the determination means 903 classifies both objects D14 and E24 as not being header objects, just like objects D13 and E23.

[0037] As shown in FIG. 3, the determining means 903 outputs object data D30 in which header / footer appropriateness information is added to each object.

[0038] This means that even if there is an error in the reading position during scanning or an OCR misjudgment, unnecessary parts such as headers and footers can be properly extracted and excluded.

[0039] In the example shown in Figure 3, the first page of the image data is used as the analysis page manuscript and the second page is used as the reference page manuscript to determine whether the header or footer is present. However, this is not limited to this. As shown in Figure 4, image data E20 may be obtained as the reference page manuscript from a past document stored in the external storage device 93, and image data D10 may be used as the analysis page manuscript for a new document, and the header or footer may be determined using the analysis means 902 and the determination means 903.

[0040] FIG. 5 is a flowchart showing the processing contents of the inkjet printing apparatus 1 according to the first embodiment of the present invention.

[0041] 5, in step S101, extraction means 901 extracts an object from image data read by the scanner unit 101 or image data acquired by the external interface unit 11 via the local area network LAN. If the extracted object is text, OCR processing is performed.

[0042] In step S103, the CPU 90 repeatedly executes steps S105 to S115 up to step S117 for all objects D11 to Dxx included in the analysis page document D.

[0043] In step S105, the CPU 90 repeatedly executes steps S107 to S113 up to step S115 for all objects E11 to Exx included in the reference page document E.

[0044] In step S107, the analysis unit 902 calculates the content difference a of each object between the analysis page manuscript D and the reference page manuscript E.

[0045] In step S109, the analysis unit 902 calculates the difference b between the coordinates of each object between the analysis page document D and the reference page document E.

[0046] In step S111, the analysis unit 902 calculates the difference c in size of each object between the analysis page document D and the reference page document E.

[0047] In step S113, the determining means 903 determines whether or not the content difference a, the coordinate difference b, and the size difference c are all smaller than the threshold values.

[0048] If it is determined that the content dissimilarity a, coordinate difference b, and size difference c are all smaller than the threshold (step S113; YES), in step S121, the determination means 903 adds header / footer appropriateness information (meta information) to each object D and outputs it.

[0049] (Addendum) The present application discloses the following inventions.

[0050] (Appendix 1) an extraction means for extracting an object from an image of each page of a document; an analysis means for analyzing an image of one page of a document as a reference page and other pages as analysis pages, and analyzing each object on the reference page and each object on the analysis page; a determining means for determining whether each object on the reference page is a necessary part or an unnecessary part according to the result of the analyzing means; An image processing device comprising:

[0051] This makes it possible to properly extract and remove unnecessary parts such as headers and footers even if there is an error in the reading position during scanning or an OCR misjudgment. By removing information such as document forms, headers, and footers, documents can be digitized, improving the reusability and searchability of information. [Explanation of symbols]

[0052] 1. Inkjet printing device 10. Control Unit 11 External interface section 14 Client Terminal 15 External interface section 19 Input section 20 Output section 21 External storage device 80 Operation Panel 901 Extraction means 902 Analysis means 903 Judgment means 93 External storage device 101 Scanner section 102 Printer section 110 Printer Driver

Claims

[Claim 1] an extraction means for extracting an object from an image of each page of a document; an analysis means for analyzing an image of one page of a document as a reference page and other pages as analysis pages, and analyzing each object on the reference page and each object on the analysis page; a determining means for determining whether each object on the reference page is a necessary part or an unnecessary part according to the result of the analyzing means; An image processing device comprising:

Images