JP2008210327A - Character image output system and character image output method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the accuracy of a database and to perform high-speed registration of a character and image data. <P>SOLUTION: This character image output system comprises: a scanner 10 for imaging and reading first image data for character recognition and second image data for image storage from the images of forms on which the character is written; an image input device 20 for recognizing, from the first image data, the character data included in the first image data and outputting it, recognizing, from the second image data, the basic image property including the image width and height of the second image data and the form coordinate information of the forms, and outputting the basic image property, the form coordinate information, and the second image data; and an image processor 30 for performing the image processing of the second image data output from the image input device based on the basic image property and the form coordinate information and outputting the processed image data having undergone the image processing. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a character image output system and a character image output method for reading character data and image data described in a form and registering them in a database, and in particular, character data and image data can be registered in a database at high speed. The present invention relates to a character image output system and a character image output method.

  In general, an optical character recognition device has a function of reading image data of a form in which characters are described, and recognizing character data from the read image data. It is used as an output device for character image information for registering character data such as name / passport number, etc. written on the passport) and a face photograph and the entire page as image data in a database.

Note that the following patent documents are listed as documents describing the technology relating to the computer system for reading the character data and the image data.
JP 7-49541 A

  The technique described in the above-mentioned patent document describes a technique for reading image data of X-ray film in the medical field and character data of patient information included in the X-ray film. The point of registering in the database etc. at high speed was not taken into consideration.

  That is, a text image output system that reads both image and text data according to the prior art and outputs it to a database or the like includes, for example, a step of reading a large number of forms as image data by scanning, A batch process including a process of storing after performing image processing such as binarization and a character recognition process of recognizing characters using a character recognition technology from a large number of stored image data, and batch processing of all these processes Since the image data and character data are stored in the database after the completion of the process, the data to be registered in the database cannot be constructed until all the batch processes described above are completed, and the data registration time in the database is redundant. There was a problem to say.

  For example, a searcher who searches for any information using the database has a short time lag and a high probability that the search target will be hit (searched information), rather than searching for completed information including image data. In the conventional technology, the data cannot be registered in the database until the data including the image data for storage is completed. For this reason, the database cannot be registered. There was a problem that it was difficult to improve the accuracy of the.

  An object of the present invention is to eliminate the above-mentioned problems, and to provide a character image output system and a character image output method capable of registering image data and character data in a database at high speed.

  To achieve the above object, the present invention captures an image of a form or the like on which characters are described, recognizes character data from the image, cuts out the image data, and outputs the character data and the image data. A system that includes a scanner that reads first image data for character recognition and second image data for image storage from an image including characters of the form, and the first image data includes the first image data. Recognizing and registering character data in a database, recognizing basic image properties including image width height of the second image data and form coordinate information of the forms from the second image data; An image input device that outputs basic image properties, form coordinate information, and second image data, and second image data output from the image input device is used as the basic image property. And performs image processing on the basis of the form coordinate information, the first feature in that an image processing apparatus for outputting the processed image data subjected to the image processing.

  Further, the present invention provides the character image output system according to the first feature, wherein the image input device includes an output format, an output resolution, and skew correction information of the second image data from the second image data. Recognizing information, the image processing apparatus cuts out the cut-out image data from the second image data based on the basic image property and the form coordinate information, and determines the inclination of the cut-out image data based on the skew correction information. 2. The character image output system according to claim 1, wherein correction is a second feature, wherein the scanner reads image data of a form, and the image data read by the image input device Registration of character data recognized from the database, processing for outputting basic image properties, form coordinate information, and second image data; The image processing apparatus to a third feature be carried out continuously and processing for outputting the processed image data.

  Further, according to the present invention, a scanner captures an image of a form or the like on which characters are described, and the image input device and the image processing device recognize character data from the image and register it in a database and cut out the image data. A character image output method in a character image output system for outputting image data, wherein the scanner reads first image data for character recognition and second image data for image storage from an image including characters of the forms. A first step, wherein the image input device recognizes character data included in the first image data from the first image data and registers it in a database, and increases the image width of the second image data from the second image data. A basic image property including the size and form coordinate information of the form, and outputting a basic image property, form coordinate information, and second image data. And the image processing device performs image processing on the second image data output from the image input device based on the basic image property and the form coordinate information, and outputs processed image data subjected to the image processing. The fourth feature is to execute the third step.

  According to the present invention, in the character image output method having the characteristics described above, the second step recognizes editing information including an output format, output resolution, and skew correction information of the second image data from the second image data. The third step includes a step of cutting out the cut-out image data from the second image data based on the basic image property and the form coordinate information, and an inclination of the cut-out image data based on the skew correction information. And a character image output method according to any one of the above features, wherein the scanner according to the first step reads the image data of the forms, and the second step through the second step. A process in which the image input device in three steps outputs the recognized character data, basic image properties, form coordinate information, and second image data; And sixth aspect of the performing the processing image processing apparatus outputs the processed image data and character data continuously.

  In a character image output system and method according to the present invention, a scanner reads first image data for character recognition and second image data for image storage from an image including characters of a form, and the image input device reads the read image data. First, processing for recognizing and outputting character data and recognizing basic image properties and form coordinate information for image editing is performed, and then the image of the second image data is based on the basic image properties and form coordinate information. Since processing is performed and output, output of character data to a database or the like can be performed at a higher speed than batch processing.

Hereinafter, an embodiment of a character image output system to which a character image output method according to the present invention is applied will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing a configuration of a character image output system according to the present embodiment, FIG. 2 is a diagram showing two types of image data read by the character image output system according to the present embodiment, and FIG. 3 is spooled by the present embodiment. FIG. 4 is a diagram for explaining information, FIG. 4 is a diagram showing a flow of one image data input by the image input device of the present embodiment, and FIG. 5 is a flow of one image data input by the image processing device of the present embodiment. FIG.

<Configuration>
As shown in FIG. 1, the character image output system according to the present embodiment includes a multi-sensor scanner 10 that continuously captures and reads image data of forms, and spool data that will be described later from image data scanned by the multi-sensor scanner 10. And an image input device 20 that outputs the image data together with the image (image) data, and an image processing device 30 that receives the spool data and the image (image) data from the image input device 20 and performs image processing. . Note that the forms described in this specification are not limited to paper sheets but also include booklets such as passports.

  The multi-sensor scanner 10 reads the OCR image data 60 which is a monochrome image for character recognition shown in FIG. 2 and reads the image data 61 which is color image data. And an image reading unit for capturing an image. The OCR image data 60 and the image image data 61 are read in one scan.

  The image input device 20 inputs / outputs both image data read from the multi-sensor scanner 10 and temporarily stores both image data read from the input / output port 21 for processing. Memory 24, hard disk device 25 for storing image data and the like stored in memory 24, I / F (interface) port 23 for outputting data to image processing device 30, and each circuit The CPU 22 controls the above.

  The spool data is recognized based on the OCR image data 60 and the image image data 61, and as indicated by reference numeral 70 in FIG. 3, a header portion such as a reading number included in the OCR image data 60 and others. 74, basic image properties 75 indicating basic information of read image data such as image width / image height / resolution / bit depth / compression information, and upper left coordinates / upper right coordinates of the forms in the image data 60 and 61 / Form coordinate information 76 including lower left coordinates / lower right coordinates and output format of image data 61 (BMP, JPEG, GIF and other image data formats) / output resolution (DPI) / bit conversion information / border width / halftone dot Edit information 77 including removal information / skew correction information (information for correcting the inclination of the form) and the image data 61 for image storage described above. Constructed.

  The image processing device 30 temporarily stores the input spool data and image data, and an I / F (interface) port 31 to which the spool data and image data output from the image input device 20 are input. It comprises a memory 33 for processing, a hard disk device 34 for storing the processed image data, etc., and a CPU 32 for controlling the respective circuits.

<Operation>
Next, the operation of the character image output system according to the present embodiment will be described with reference to FIGS.
As shown in FIG. 4, the character image output system includes a step 40 in which the multi-sensor scanner 10 starts reading both the image data 60 and 61 of the form, and a CPU 22 of the image input device 20 from the scanner 10. Are input via the input / output port 21 and stored in the memory 24, and the four-point coordinates of the four corners of the form such as the form in the read image data 60 or 61 are stored. Step 42 for detection by comparison with the background, Step 43 for starting recognition of character data based on the image data surrounded by the four-point coordinates and output to a database (not shown), and step for creating the aforementioned spool data 44, and the created spool data and the image data 61 are transferred to the image processing apparatus 30 via the I / F port 23. And flop 45, through the step 46 to wait for the end of step 45 in step 44 to transfer threads of the aforementioned recognition thread, it executes the step 47 to end the process. As a result, the character image output system operates to register the character data that has undergone character recognition in the database, and output spool data including basic image properties and other information for image processing and image data for image storage. To do. The CPU 22 may compress the image data in order to increase the transfer speed during the data transfer in the step 45, and transfer the compressed image data.

  As described above, in the character image output system according to the present embodiment, the multi-sensor scanner 10 simultaneously scans the OCR image data 60 for character recognition and the image image data 61 which is image data for image storage from one form. The image input device 20 that has received the data 60 and 61 receives the spool data (basic image property / form coordinate information / edit information) recognized from the OCR image data 60 and the like, and the image image data 61 sequentially. It operates to output to 30.

  Next, in this character image output system, as shown in FIG. 5, the CPU 32 of the image processing apparatus 30 starts inputting spool data and image image data 61 sent from the image input apparatus 20 via the I / F port 31. Steps 50 to 51, step 52 for determining whether or not the read storage image data 61 is compressed, and decompressing (decompressing) the image data when it is determined in step 52 that the image data 61 is compressed. Step 53, and when it is determined in step 52 that the image data 61 is not compressed, the basic image property and coordinate information 76 (form coordinate information that is the four-point coordinates of the form category) included in the spool data from the image data 61 are obtained. Step 54 for cutting out image data of a form part (indicated by oblique lines) based on the cut-out image data Step 55 for performing skew correction based on skew correction information included in the editing information; Step 56 for performing other image processing such as resolution and change of the cut-out image data and color correction following the step 55; Step 57 for temporarily recording the processed image data on the hard disk device 34 and outputting the processed image data to a database system (not shown) and step 58 for deleting the spool data, and then ending the processing. By executing, it operates to output image data for image storage.

  As described above, in the character image output system according to the present embodiment, the image input device 20 inputs the OCR image data 60 for character recognition of forms and the image image data 61 which is image data for image storage, and the OCR image is input. Character data is extracted from the data 60 and the like and output to the database, spool data including image editing information (basic image property 75 / coordinate information 76 / editing information 77) is extracted, and the image processing apparatus 30 together with the image image data 61 is extracted. The image processing apparatus 30 that has received and received the image data of the image data to be stored using the coordinate information 76 including the coordinate values for cutting out the forms included in the received spool data and the editing information 77 for the image processing. It operates so that it is processed and output, and this process is continuously performed for each form, so that a large number of books can be obtained as in the prior art. Compared to batch processing, which first reads image data of the same type and then recognizes character data, it outputs character data to a database, etc. to execute character data recognition and image data storage for each image. Can be performed at high speed.

  In particular, in this embodiment, when data is created by batch processing according to the prior art, registration of character data for search in the database is delayed, whereas in this example, character data is stored for each form. Since it is recognized and registered in the database, the character data for search can be registered in the database earlier than in the prior art, and therefore the data to be searched can be registered in the database or the like earlier.

  That is, the character image output system and method according to the present embodiment includes a process in which the scanner 10 reads form image data, output to the database of character data recognized from the image data read by the image input device 20, and basic image properties. And the process of outputting the form coordinate information and the second image data, and the process of outputting the processed image data by the image processing apparatus 30 are continuously performed, so that the data to be searched can be quickly stored in a database or the like. You can register.

The figure which shows the structure of the character image output system by one Embodiment of this invention. The figure which shows two types of image data which the character image output system by this embodiment reads. The figure for demonstrating the information spooled by this embodiment. The figure which shows the processing flow by the image input device of this embodiment. The figure which shows the processing flow by the image processing apparatus of this embodiment.

Explanation of symbols

  10: Multi-sensor scanner, 20: Image input device, 23: I / F port, 24: Memory, 25: Hard disk device, 30: Image processing device, 31: I / F port, 33: Memory, 34: Hard disk device, 60: OCR image data, 61: Image image data, 74: Header part, 75: Basic image property, 76: Form coordinate information, 77: Editing information.

Claims (6)

A character image output system that captures an image of a form in which characters are described, recognizes character data from the image, cuts out the image data, and outputs the character data and the image data,
A scanner for reading first image data for character recognition and second image data for image storage from an image including characters of the forms;
Character data included in the first image data is recognized from the first image data, basic image properties including the image width height of the second image data from the second image data, and form coordinate information of the forms An image input device that outputs the recognized character data, basic image properties, form coordinate information, and second image data;
A character image comprising: an image processing device that performs image processing on the second image data output from the image input device based on the basic image property and form coordinate information, and outputs processed image data subjected to the image processing Output system.   The image input device recognizes editing information including an output format, output resolution, and skew correction information of the second image data from the second image data, and the image processing device recognizes the basic image property and the form. The character image output system according to claim 1, wherein cutout image data is cut out from the second image data based on coordinate information, and an inclination of the cutout image data is corrected based on the skew correction information.   A process in which the scanner reads form image data, a process in which the image input device recognizes the read image data, outputs basic image properties, form coordinate information, and second image data; The character image output system according to claim 1, wherein the image processing apparatus continuously performs a process of outputting the processed image data. A character image output system in which a scanner captures an image of a form in which characters are described, and the image input device and the image processing device recognize character data from the image, cut out the image data, and output the character data and image data A character image output method in
A first step in which the scanner reads first image data for character recognition and second image data for image storage from an image including characters of the forms;
The image input device recognizes and outputs character data included in the first image data from the first image data, and includes basic image properties including an image width height of the second image data from the second image data. Recognizing the form coordinate information of the form, and outputting a basic image property, form coordinate information, and second image data;
Third step in which the image processing device performs image processing on the second image data output from the image input device based on the basic image property and form coordinate information, and outputs processed image data on which the image processing has been performed. Character image output method to execute.   The second step includes a step of recognizing edit information including an output format, output resolution, and skew correction information of the second image data from the second image data, and the third step includes the basic image. 5. The character image according to claim 4, comprising: a step of cutting out cutout image data from the second image data based on properties and form coordinate information; and a step of correcting the inclination of the cutout image data based on the skew correction information. output method.   A process in which the scanner in the first step reads image data of a form, an output of the character data recognized by the image input device in the second to third steps, basic image properties, form coordinate information, and second image data. 6. The character image output method according to claim 4, wherein the process of outputting the image data and the process of outputting the processed image data by the image processing apparatus according to the third step are continuously performed.

Images