JP3761923B2 - Image processing apparatus and method - Google Patents

Description

[0001]
[Industrial application fields]
The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus and method for determining or correcting attributes of input image data and passing them to a lower-level process.
[0002]
[Prior art]
Generally, in a character recognition device, a character area in a document image input from an image scanner or the like is extracted, and whether or not the attribute for each area (judgment of vertical writing or horizontal writing) is incorrect, or the operator After the attribute for each area is specified, the character recognition process is performed.
[0003]
[Problems to be solved by the invention]
However, if one attribute is specified for each area, when there are many areas, the work becomes complicated and very troublesome.
[0004]
[Means for Solving the Problems]
The present invention has been made in view of such problems, and an object of the present invention is to provide an image processing apparatus and method capable of setting the designation of the attribute of each area in an image by a simple operation.
[0005]
In order to solve this problem, for example, an image processing apparatus of the present invention has the following configuration. That is,
Area dividing means for dividing the image into a plurality of areas;
Attribute determination means for determining an attribute for each of the divided areas;
Display means for controlling the display of the image and the symbol by associating the symbol indicating the determined attribute for each of the divided areas;
A selection means for selecting a plurality of areas in response to a selection instruction from the user from the displayed image;
When there is an instruction to change the attribute for the symbol associated with any of the selected plurality of areas, the attribute of the area associated with the symbol that has received the instruction to change the attribute is changed, Attribute changing means for changing a plurality of areas selected by the selecting means to the same attribute as the attribute of the area whose attribute has been changed.
[0006]
Further, according to a preferred embodiment of the present invention, a frame display means for displaying a frame for each region in the input image,
Attribute information display means for displaying attribute information for each area,
Preferably, the designation means designates a desired frame, and the attribute designation means designates an attribute by designating one attribute information among a plurality of alteration designations.
[0007]
As a result, attributes can be specified in visual frame units, and operations can be performed without having special knowledge.
[0008]
In this case, further, when the attribute is specified by the attribute specifying unit, the attribute information displayed in the plurality of frames specified by the specifying unit is updated according to the specified attribute. It is desirable to provide updating means. As a result, the changed attribute can be confirmed.
[0009]
The attribute information displayed by the attribute information display means is preferably a symbol pattern. As a result, the attribute can be displayed to the operator with a small display area, while the attribute is displayed with characters or the like.
[0010]
Further, it is preferable that the specifying means includes specifying all areas at once. As a result, when an image having a single attribute is input, all attributes can be specified at once.
[0011]
Moreover, it is desirable that the lower-order process is a character recognition process. This makes it possible to perform correct recognition processing or to perform efficient character recognition.
[0012]
In addition, it is desirable that the attribute includes type information of vertical writing and horizontal writing of a character image. As a result, the recognition order for character recognition is recognized as the reading direction of the character string.
[0013]
【Example】
Embodiments according to the present invention will be described below in detail with reference to the accompanying drawings.
[0014]
<Description of Character Recognition Device (FIG. 1)>
FIG. 1 is a block diagram illustrating a schematic configuration of a character recognition apparatus according to an embodiment. In FIG. 1, reference numeral 1 denotes a character recognition device, and 2 denotes an input unit for reading and inputting a document image, which is composed of, for example, a scanner and inputs document image data as digital data. Reference numeral 3 denotes a central processing unit (hereinafter referred to as a CPU) for arithmetic processing that controls the entire apparatus 1. Reference numeral 4 denotes a ROM, which stores the control program and various data shown in the flowchart of FIG. A memory (RAM) 5 is used as a work area of the CPU 3 and stores an area for storing document image data input from the input unit 2 and a position and size of a character pattern cut out by the preprocessing unit 7. This is also used for storing the area to be recognized, and the place where the recognition is performed in the image, that is, the block frame coordinate information. An attribute change unit 6 is used to select a region and change an attribute by clicking a block symbol (details will be described later) by an operation using a pointing device (not shown). A pre-processing unit 7 cuts out a one-character pattern of the document image stored in the memory 5. A recognition unit 8 extracts a geometric feature from the pattern preprocessed by the preprocessing unit 7 and performs character recognition of the document image by comparing with a standard pattern stored in the recognition dictionary 9 in advance. . A display unit 10 includes an image display unit 14 for displaying an input document image, a recognition result display unit 15, a block frame display unit 16, and an attribute display unit 17. A system bus 11 includes a data bus from the CPU 3, an address bus, a control signal bus, and the like. Reference numeral 12 denotes an interface unit that performs interface control with an external output device such as a printer 13.
[0015]
In the above configuration, a document image is input from the input unit 2 and stored in the memory 5. Then, a reduced image is generated on the memory 5 separately from the stored image data. The reduced image is, for example, an n × n pixel block of the original image as one reduced pixel, and when at least one black pixel exists in the n × n pixel block, the reduced pixel is determined as a black pixel. Is. When this process is performed, the result is the same as when the original image is 1 / n times in length and width, and the connected character images (dot images of character patterns) are connected to each other.
[0016]
The rectangle circumscribing the connected dot distribution is sequentially defined, and the region is divided. It should be noted that most attributes of the processing block, such as the size of the area obtained by the area division process and the ratio of black pixels included in the area (character or graphic, vertical writing or horizontal writing if character, etc.) Can be judged.
[0017]
In the following description, the region division process is described as being performed as described above. However, the region division process is not limited to this, and any process may be performed, and the attribute determination process is not limited to this.
[0018]
<Description of image window (FIG. 2)>
FIG. 2 is a diagram illustrating an example of an image window indicating a document image (stored in the memory 5) input to the character recognition device 1 of the embodiment, a region division result, and region attributes.
[0019]
In FIG. 2, reference numeral 21 denotes an image window showing a document image read by the input unit 2. Reference numeral 22 denotes a frame (a frame that clearly indicates the size of one area) as a result of area division, and shows the position of a paragraph in the text portion. Reference numerals 23 and 24 are block symbols indicating region attributes. The block symbol 23 faces the right side, indicating that the text in the block frame is recognized as horizontal writing. The number 1 in the middle indicates the order of character recognition. The block symbol 24 indicates that the corresponding area is vertically written.
[0020]
That is, FIG. 2 shows a state in which all the areas should be recognized as horizontal writing, but sentences in some block frames have been recognized as vertical writing.
[0021]
In such a situation, in the embodiment, misjudged attributes are corrected or designated collectively.
[0022]
Hereinafter, the character recognition processing in the embodiment will be described based on the flowchart of FIG. A control program for executing this process is stored in the ROM 4.
[0023]
First, in step S31, a document image set by the operator is read and input by the input unit 2 (scanner) and stored in the memory 5.
[0024]
Next, it progresses to step S32 and area division is performed. Extract text paragraphs, diagrams, tables, ruled lines, etc. by area division.
[0025]
In step S33, the attribute of the area is determined, the result is stored in the memory 5, and the attribute display unit 17 displays the attribute in the form of a block symbol.
[0026]
In step S34, the region attributes are corrected using a mouse or the like. In step S35, character recognition is performed according to the corrected attribute. Finally, the process proceeds to step S36, and the recognition result is displayed.
[0027]
FIG. 4 is a flowchart showing the corresponding image display processing in step S34 of FIG.
[0028]
First, in step S41, a plurality of block frames are selected. In the embodiment, for example, as shown in FIG. 5, the pointing device is operated to display the “Edit” pull-down menu in the menu bar displayed at the top of the display screen, and “Select all block frames”. By specifying, the attributes of all the block frames in the image window can be set as edit targets. As shown in FIG. 6, the selected block frame is marked with a gray square at the four corners to be distinguished from the unselected block frame. Here, all the block frames are selected.
[0029]
In step S42, when one block symbol is designated (clicked) with the pointing device with the mouse, all the selected attributes are corrected at once according to the change. For example, as shown in FIG. 6, if the third, fifth, and sixth block frames in the recognition order are erroneously determined in the set direction and are vertically written, any of the third, fifth, and sixth block symbols is Click to change the block attribute, here the group direction, from vertical writing to horizontal writing. All the blockwork attributes selected at that time are written horizontally. In other words, the third and fifth attributes can be corrected at once by clicking the block symbol in the sixth block frame. Since the other block frames are originally certified as horizontal writing, no changes are made.
[0030]
When the above designation is performed, in step S43, the display of the block symbols of all the changed block frames is changed. The result is shown in FIG. Here, all block symbols turn to the right, indicating that they are written horizontally.
[0031]
As a result, it is possible to change or modify the attributes of information in a plurality of block frames with few operations.
[0032]
After that, by specifying “Recognition” in the menu bar, the attribute information is used for lower-level character recognition processing along with the image information of the block frame selected at that time (with square marks at the four corners). Then, recognition processing is performed. In the character recognition process, the character images in the block frame are received in order from the smallest block number selected at that time, and the character recognition process is sequentially performed according to the determined direction (attribute).
[0033]
In the above description, the example in which the attributes in all the block frames are changed at a time is shown. However, the desired block frames may be sequentially selected. For example, if you use a mouse or the like, add the block frame to be processed by clicking the desired block frame with the left button while pressing the right button (or pressing a predetermined key on the keyboard) To choose. By doing so, it becomes possible to select only a desired block frame (a plurality of block frames).
[0034]
In the above description, only two states of vertical writing and horizontal writing have been described as attributes in the block frame. That is, by simply designating a block frame symbol that was erroneously recognized as vertical writing, it was uniquely corrected to horizontal writing. However, when there are three or more attribute types, the changed attribute cannot be uniquely specified. Therefore, when there are three or more candidate attributes, when a block frame (plural) to be changed is selected, a menu window for inquiring about the attribute is displayed, and one of the menus or A plurality may be selected and changed to the corresponding attribute.
[0035]
For example, considering a case where a manuscript is a mixture of text (characters), photographs, tables, figures, etc., the symbol when it is determined to be a character area is a pentagon (has portrait and landscape orientations) as described above, In the case of a photograph, it is assumed to be displayed as a circle in the case of a circle, a table or a figure. At this time, if some areas are recognized as photographs even though they are characters, when the symbol of the block frame (in this case, a circle) is specified, either a character or a table graphic will be used as an attribute. A menu window is displayed to specify whether it is “character”.
[0036]
In particular, even if it is a character, even if it is recognized as a photo, a table, or a figure, the remaining attribute is only “character”. At this time, simply change the attribute of the block frame to be changed. When specified, an attribute as "character" can be automatically given.
[0037]
The attributes are not limited to the above characters, photos, and figures. For example, in the case of characters, there are horizontal writing and vertical writing as described above, but other types such as Japanese and English can also be included. Note that any method may be adopted for the rough determination process of the attribute.
[0038]
Needless to say, the recognition result can be stored and held as text data, edited, and printed.
[0039]
As described above, according to the present embodiment, it is possible to realize a character recognition device that is easy to use for a user, in which a plurality of regions exist in a document and correction can be easily performed even when there are many attribute corrections.
[0040]
In particular, by making it possible to select all block frames at once, a document with a single attribute (for example, a document with only horizontal writing) can be selected with a very small number of operations. Can be changed, modified or specified in a batch.
[0041]
In addition, as a low-level process for images in block frames with attributes such as tables (including ruled lines) and graphics, or images in blocks with attributes such as photographs, the processing at the time of saving as a file This can be applied to the type of compression processing, the type of scaling when scaling.
[0042]
<Second embodiment>
In the first embodiment, the attribute is changed by using the mouse in step S42 in FIG. 4, but anything can be used as long as the attribute can be changed, such as using a key or a menu. As described above, also in the second embodiment, the same operations and effects as those in the first embodiment can be obtained.
[0043]
<Third embodiment>
In the first embodiment, the block symbol indicating the attribute of the block has been described as being a pentagon or the like, but any shape, position, color, size, etc. may be used. However, it is desirable that they can be distinguished from each other as viewed from the operator.
[0044]
As described above, also in the third embodiment, the same operations and effects as those in the first embodiment can be obtained.
[0045]
As described above, according to the present embodiment, even when there are a plurality of fine areas in the document and there are many attribute corrections, the correction can be easily performed. For this reason, when the document original is read, recognized and corrected, there is an effect that it is easy to use and the operability of character recognition is improved.
[0046]
In the embodiments, the document image to be processed is described as being input from an image scanner or the like. However, the present invention can also be applied to a storage medium (for example, a floppy disk) storing an image or a case where the image is received via a line. The present invention is not limited by the above example.
[0047]
Furthermore, in the embodiment, it goes without saying that it can be achieved by supplying a program to the apparatus or system from the outside even if it is adapted to a system composed of a single apparatus or a plurality of devices. . Therefore, it is possible to easily adapt to a storage medium in which a program for realizing the present invention is incorporated.
[0048]
【The invention's effect】
As described above, according to the present invention, the designation of attributes of each area in the input image can be set by a simple operation.
[0049]
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration of a character recognition apparatus according to a first embodiment.
FIG. 2 is a diagram illustrating a display unit and a menu in the character recognition device according to the first embodiment.
FIG. 3 is a flowchart showing processing of the character recognition device according to the first embodiment;
FIG. 4 is a flowchart showing details of step S34 in the first embodiment;
FIG. 5 is a diagram illustrating a display unit and a menu in the character recognition device according to the first embodiment.
FIG. 6 is a diagram illustrating a display unit and a menu in the character recognition device according to the first embodiment.
FIG. 7 is a diagram illustrating a display unit and a menu in the character recognition device according to the first embodiment.
[Explanation of symbols]
1 Character recognition device 2 Input unit 3 CPU
4 ROM
DESCRIPTION OF SYMBOLS 5 Memory 6 Attribute change part 7 Preprocessing part 8 Recognition part 9 Recognition dictionary 10 Display part 11 System bus 12 Interface part 13 Printer 14 Image display part 15 Recognition result display part 16 Block frame display part 17 Attribute display part 21 Image window 22 Block frame 23, 24 Block symbol

Claims (14)

Area dividing means for dividing the image into a plurality of areas;
Attribute determination means for determining an attribute for each of the divided areas;
Display means for controlling the display of the image and the symbol by associating the symbol indicating the determined attribute for each of the divided areas;
A selection means for selecting a plurality of areas in response to a selection instruction from the user from the displayed image;
When there is an instruction to change the attribute for the symbol associated with any of the selected plurality of areas, the attribute of the area associated with the symbol that has received the instruction to change the attribute is changed, An image processing apparatus comprising: attribute changing means for changing a plurality of areas selected by the selecting means to the same attributes as the attributes of the changed areas. The display means controls to display a frame indicating the range of the divided area, and further, when the area is selected by the selection means, the selected area can be identified. The image processing apparatus according to claim 1, wherein control is performed so that a display form of a frame indicating a range of the selected area is changed and displayed. The image processing apparatus according to claim 1, wherein the selection unit can select all regions at once. The image processing apparatus according to claim 1, wherein each region of the image and the attribute changed by the attribute changing unit are passed to a character recognition process. The image processing apparatus according to claim 1, wherein the attribute includes information indicating a type of whether the character image is vertical writing or horizontal writing. The image processing apparatus according to claim 1, wherein the attribute includes information indicating a type of a character or a graphic. The attribute changing means controls to display a menu of attribute options when the user gives an instruction to change the attribute for a symbol associated with any of the plurality of areas selected by the selecting means. The user selects a desired attribute from the menu, changes the attribute of the area associated with the symbol for which the attribute change instruction has been received, and changes the plurality of areas selected by the selection unit to the menu-selected attribute. The image processing apparatus according to claim 1, wherein the attribute is changed to the same attribute as the attribute of the changed area. A region dividing step for dividing the image into a plurality of regions;
An attribute determination step for determining an attribute for each of the divided areas;
A display step of controlling to display the image and the symbol by associating a symbol indicating the determined attribute for each of the divided areas;
A selection step of selecting a plurality of regions in accordance with a selection instruction from the user from among the displayed images;
When there is an instruction to change the attribute for the symbol associated with any of the selected plurality of areas, the attribute of the area associated with the symbol that has received the instruction to change the attribute is changed, An image processing method comprising: an attribute change step of changing a plurality of areas selected in the selection step to the same attribute as the attribute of the attribute-changed area. In the display step, control is performed so that a frame indicating the range of the divided area is displayed, and when the area is selected in the selection step, the selected area is identified so that the selected area can be identified. 9. The image processing method according to claim 8, wherein control is performed such that the display form of a frame indicating the range of the selected area is changed and displayed. The image processing method according to claim 8, wherein in the selection step, it is possible to select all regions at once. 9. The image processing method according to claim 8, wherein each region of the image and the attribute changed in the attribute changing step are passed to a character recognition process. The image processing method according to claim 8, wherein the attribute includes information indicating a type of whether the character image is vertical writing or horizontal writing. The image processing method according to claim 8, wherein the attribute includes information indicating a type of character or graphic. In the attribute change step, control is performed to display a menu of attribute options when an instruction to change the attribute is given from the user to a symbol associated with any of the plurality of areas selected in the selection step. The user selects a desired attribute from the menu, changes the attribute of the area associated with the symbol for which the attribute change instruction has been received, and changes the plurality of areas selected in the selection step to the menu-selected attribute. The image processing method according to claim 8, wherein the attribute is changed to the same attribute as the attribute of the changed area.

Images