JP2013149210A - Image processing program, image processing method and image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To acquire a proper partial image corresponding to an area designated from a prescribed image.SOLUTION: An area of a partial image acquired from a prescribed image is designated, the prescribed image is recognized as one or a plurality of block areas on the basis of an analysis parameter analyzing an image layout, the recognized one or plurality of block areas are compared with the designated area to perform a first block determination for classifying the block areas into any one of a first block area included in the designated area, a second block area which is not included in the designated area and a third block area which is an undecided area. After the analysis parameter is changed for the block areas determined as the third block area and the block areas are re-recognized, a second block determination for classifying the block areas into the first block area or the second block area is performed. The partial image is generated on the basis of a set of the first block area obtained in the first block determination and the second block determination.

Description

  The present invention relates to an image processing program, an image processing method, and an image processing apparatus for acquiring an appropriate partial image corresponding to a specified region in a predetermined image.

  Conventionally, when a part of a paper document is to be cut out and stored, a method is known in which a part to be stored is cut and imaged, as in a conventional newspaper cutout. Conventionally, a method is known in which an area to be saved is surrounded by a marker pen line and only a portion surrounded by a scan image is cut out without actually cutting the paper (see, for example, Patent Document 1). . Note that all of the above-described conventional techniques have damaged the original document, and the operation has been complicated.

  Therefore, as another method, a method of scanning the entire document and designating a region on the screen to generate a partial image can be considered. For example, conventionally, a method is known in which images obtained by cutting out partial regions of a document are arranged and rearranged on a blank mount region provided on a screen. Conventionally, there is a known method for performing layout analysis to divide a document into areas such as text, diagrams, and titles, and determining whether or not to save an image in each area according to a scoring rule determined for each area. Yes. Furthermore, conventionally, a method for correcting the contour of a region designated by the user from the result of the layout analysis described above is known (see, for example, Patent Documents 2 to 4).

JP 2006-340103 A JP 61-15280 A JP 09-091450 A JP 2001-274969 A

  By the way, in the conventional method described above, coordinate input means such as a mouse is used to designate a region of a partial image to be cut out from the entire image, and the designated partial region is limited to a rectangle (quadrangle). For this reason, if the partial image is not rectangular, unnecessary images that are not to be saved and are included in the designated area are stored.

  On the other hand, in the conventional method, there is a method of drawing a curve along the outer frame of the partial image area with a mouse or the like. However, since it is difficult to specify the area accurately, the operation is complicated and inaccurate. Met.

  In addition, in the conventional method using the result of layout analysis, an area can be specified only in units of layout elements such as text and charts, so that the area desired by the user cannot be selected appropriately. Here, the above-mentioned problem will be described specifically. FIG. 1 is a diagram for explaining the problems of the prior art. 1A to 1C show document images such as newspaper articles, as an example. “AAA” means an arrangement of document data such as a title and a body composed of letters, numbers, etc., and includes an area in which diagrams and the like are arranged in addition to the document (the same applies to the following description). To do).

  In the prior art, for example, when there is an area to be selected as shown by a dotted line portion in FIG. 1A with respect to a document image as shown in FIG. 1, as shown in FIG. Assume that a correct block area is obtained. In such a case, a desired area can be obtained by combining the block areas (1) to (5). On the other hand, when an incorrect block area as shown in FIG. 1C is obtained in the layout analysis, for example, even if the block areas (a) to (c) are combined, the desired area is not sufficient. The block areas (a) to (d) are redundant. Therefore, in the conventional method, if the layout analysis is incorrect, a correct partial image cannot be obtained. Furthermore, since the area selected by the layout analysis is not fed back to the user, there is a problem that even if an inappropriate area is selected, the area is not noticed.

  In view of such a problem, the disclosed technique aims to acquire an appropriate partial image corresponding to a specified region from a predetermined image.

  The image processing program according to an aspect of the disclosure specifies a region of a partial image acquired from a predetermined image, recognizes the predetermined image as one or a plurality of block regions based on an analysis parameter for analyzing an image layout, The recognized one or more block areas are compared with the designated area, and a first block area included in the designated area, a second block area not included in the designated area, and an unspecified First block determination to be classified into any of the third block areas that are fixed areas is performed, the analysis parameter is changed for the block area determined to be the third block area, and the block After re-recognizing the area, a second block determination for classifying the first block area or the second block area is performed, and the first block determination is performed. And generating a partial image based on the first set of block area obtained by the second block determination to execute the process to the computer.

  According to the disclosed technique, an appropriate partial image corresponding to a specified region can be acquired from a predetermined image.

It is a figure for demonstrating the problem of a prior art. It is a figure which shows an example of schematic structure of an image processing apparatus. It is a figure which shows an example of the hardware constitutions of the image processing apparatus in this embodiment. It is a figure which shows the function structure of the block area determination means in 1st Example. It is a figure which shows an example of the layout analysis result by a separator. It is a figure which shows the specific example of a fine density parameter. It is a figure which shows an example of extraction of a blank separator. It is a figure for demonstrating the difference in operation | movement by a layout analysis parameter. It is a figure which shows the block area classification example in 1st Example. It is a flowchart which shows an example of the image process in 1st Example. It is a figure which shows an example of the layout analysis by the group integration in 1st Example. It is a figure which shows an example of the screen which prompts re-input of a user designation area. It is a figure for demonstrating the specific example of the partial image obtained by 1st Example. It is a figure which shows the function structure of the block area determination means in 2nd Example. It is a flowchart which shows an example of the image process in 2nd Example. It is a figure which shows the function structure of the block area determination means in 3rd Example. It is a flowchart which shows an example of the image process in 3rd Example.

  Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

<Image processing apparatus: schematic configuration example>
A schematic configuration example of the image processing apparatus according to the present embodiment will be described with reference to the drawings. FIG. 2 is a diagram illustrating an example of a schematic configuration of the image processing apparatus. The image processing apparatus 10 illustrated in FIG. 1 includes an image input unit 11, an area designation unit 12, a block area determination unit 13, and a partial image generation unit 14.

  The image input unit 11 inputs a target image for acquiring a partial image. Specifically, document data such as newspapers, magazines, and Internet homepages, and image data such as photographs are input.

  The area designating unit 12 obtains information on the area (user designated area) of the partial image that is designated by the user or the like from the image obtained by the image input unit 11 (hereinafter referred to as “whole image” as necessary). To do. The area designation may be set in advance, or may be designated by the user as necessary. The area designating unit 12 can designate one or a plurality of partial image areas for one image.

  The block area determination unit 13 recognizes the entire image obtained by the image input unit 11 as one or a plurality of block areas including a set of predetermined documents and the like by layout analysis. The block area determination unit 13 compares the block areas obtained by the layout analysis with the user-specified areas described above, and classifies them into, for example, first to third block areas based on preset threshold values. To do. Specifically, the block area determination unit 13 includes, for each block area obtained by layout analysis, a first block area included in the area specified by the user and a second block not included in the area specified by the user. Classify into block areas. Furthermore, the block area determination unit 13 classifies the undefined area that cannot be classified into the first and second block areas described above as the third block area.

  The block area determination unit 13 analyzes the image layout and changes the layout analysis parameter for recognizing the image as a predetermined block area, performs the layout analysis again, and reclassifies the block area. Thereafter, the block area determination unit 13 determines the first to third block areas described above for each reclassified block area. In addition, the specific process of the block area determination means 13 mentioned above is mentioned later.

  The partial image generation means 14 generates a partial image from the set of block areas included in the first block area by the block area determination means 13 and outputs the generated partial image. Thereby, in this embodiment, the suitable partial image corresponding to the area | region designated from the predetermined image can be acquired.

<Image processing apparatus: hardware configuration example>
FIG. 3 is a diagram illustrating an example of a hardware configuration of the image processing apparatus according to the present embodiment. In addition, the image processing apparatus 10 includes a control unit 21, a communication unit 22, a main control unit 23, an auxiliary storage unit 24, and a drive device 25. In the present embodiment, the control unit 21, the communication unit 22, the main storage unit 23, the auxiliary storage unit 24, and the drive device 25 are connected in a state where data and signals can be transmitted and received through a common bus.

  In the example of FIG. 3, the external device connected to the image processing apparatus 10 includes an image input unit 31, a coordinate input unit 32, a display unit 33, and a recording medium 34.

  Based on a control program such as an operating system (OS) and an execution program stored in the main storage unit 23, the control unit 21 performs various operations, data input / output with each hardware component, etc. Each process in this embodiment can be realized by controlling the process. Various information necessary during the execution of the program can be acquired from the auxiliary storage unit 24, and an execution result or the like can be stored.

  Specifically, the control unit 21 inputs image information from the image input unit 31 and input coordinate information from the coordinate input unit 32, and performs the entire image processing apparatus 10 in order to perform image processing in the present embodiment. Control. Further, the control unit 21 performs control for causing the display unit 23 to display various screens for executing image processing in the present embodiment. Further, the control unit 21 reads various information and programs necessary for the present embodiment from the recording medium 34 via the drive device 25, and stores information such as results and progress of execution of processing in the present embodiment on the recording medium 34. Control for writing, etc. The control unit 21 includes, for example, a central processing unit (CPU).

  The communication unit 22 uses a communication network typified by the Internet or a local area network (LAN) to connect to an external device and perform data transmission / reception. The communication unit 22 can also connect to an external device and acquire document image information such as newspapers and magazines.

  In addition, the communication unit 22 can acquire the image processing program, input information, and the like in the present embodiment from an external terminal through a communication network, and can transmit information such as the execution result and progress of the image processing to the external device.

  The main storage unit 23 stores an execution program and the like. Note that the main storage unit 23 includes a memory such as a read only memory (ROM) or a random access memory (RAM). In the present embodiment, for example, the entire image to be processed is stored in the main storage unit 23 and image processing or the like is performed.

  The auxiliary storage unit 24 is a storage unit such as a hard disk, for example. The auxiliary storage unit 24 stores an execution program, a control program provided in a computer, and the like, and can perform input / output as necessary. The auxiliary storage unit 24 can also store an entire image before image processing and a partial image as a processing result. Further, the auxiliary storage unit 24 can partially store an image that cannot be stored in the main storage unit 23 when the entire image is large data.

  The drive device 25 can set a recording medium 34 on which an execution program such as image processing in the present embodiment is recorded. In addition, the drive device 25 reads the execution program included in the recording medium 34 under the control of the control unit 21, and outputs it to the auxiliary storage unit 24 for installation. The drive device 25 can also write various data to the recording medium 34 via the drive device 25 under the control of the control unit 21.

  The image input unit 31 is a scanner device, a digital camera (imaging means), or the like. Specifically, the image input unit 31 inputs an entire image for generating a partial image. That is, the configuration of FIG. 3 performs image processing on the entire image such as a document image acquired by the image input unit 31 such as a scanner, but is not limited to this. The whole image or the like may be acquired from the above.

  The coordinate input unit 32 inputs the position of a user-designated area (trimming area) designated by the user or the like as a preset coordinate value on the image in order to generate a partial image from the entire image. As the coordinate input unit 32, for example, a personal computer (PC) includes a mouse and a pen, and a mobile information terminal such as a mobile phone and a smartphone includes a touch panel specified by a user's finger. It is not limited to this.

  The display unit 33 displays an input image for realizing the image processing in the present embodiment, a user specified area, various setting information, an execution result, an execution progress, and the like. These display contents are controlled by the control unit 21. The display unit 33 includes, for example, a display or a monitor. In the present embodiment, the user can specify a target area for an entire image such as a document image displayed on the display unit 33.

  The recording medium 34 is a portable recording medium such as a Universal Serial Bus (USB) memory or a CD-ROM. In the recording medium 34, for example, various information such as an execution program installed in a computer main body such as the image processing apparatus 10 is recorded. The recording medium 34 is set in the drive device 25, and various information recorded in advance is read out, and an execution result and an execution progress are written. The recording medium 34 can also record an entire image before image processing and a partial image as a processing result. These recordings are performed via the drive device 25 under the control of the control unit 21.

  Note that the image input unit 31 and the coordinate input unit 32 illustrated in FIG. 3 may be included in the image processing apparatus 10 as the image input unit 11 and the region designation unit 12 illustrated in FIG. 2 described above.

  In the present embodiment, the image processing in the present embodiment can be executed with the hardware configuration described above. Further, by installing the program, image processing can be easily realized by a general-purpose personal computer or the like.

<Block area determination means: first embodiment>
Next, a first embodiment of the block area determination means 13 will be described with reference to the drawings. FIG. 4 is a diagram showing a functional configuration of the block area determination means in the first embodiment. 4 has fine density parameter setting means 41, layout analysis means 42, block area classification means 43, block area determination means 44, and repetition control means 45.

  The fine density parameter setting unit 41 sets the fine density parameter as a layout analysis parameter for recognizing the entire image obtained from the image input unit 11 as one or a plurality of block areas, for example. Specifically, the fine density parameter setting means 41 sets the width and length as the conditions as the fine density parameter in order to track the block area delimiter using a blank portion having a predetermined width and a predetermined length in advance. . A specific example of the fine density parameter will be described later.

  The layout analysis unit 42 performs layout analysis on the entire image to be processed based on the parameters set by the fine density parameter setting unit 41, and recognizes the entire image as one or a plurality of block regions from the analysis result. . Note that the layout analysis unit 42 may perform layout analysis on the entire image. The layout analysis unit 42 may perform a predetermined analysis on the outside of the user specified region with reference to the coordinates of the user specified region specified by the region specifying unit 12 described above. Layout analysis may be performed on a region including a range. As a result, layout analysis can be quickly performed when layout analysis is performed only on a predetermined range outside the user-specified area.

  The block area classification unit 43 compares each block area of the analysis result obtained by the layout analysis unit 42 with the user-specified area and classifies it into any one of the above-described first to third block areas. (First block determination). Here, for comparison between the block area and the user-specified area, for example, the position of the block area (for example, coordinates) obtained at the time of layout analysis and the position of the user-specified area set at the time of area specification (for example, coordinates) And compare. The classification into the first to third block areas is performed by determining how much the block area is included in the user designated area from the above-described position, for example.

  The block area determining unit 44 outputs the block areas determined by the block area classifying unit 43 to the partial image generating unit 14 together with the respective classification results.

  When the third block area exists as a result of the classification by the block area classification unit 43, the iterative control unit 45 changes the value of the fine density parameter and performs the layout analysis again only for the third block area. . Further, the repeat control unit 45 causes the block area obtained by the second layout analysis result to be reclassified into the first to third block areas (second block determination) by the block area classifying unit 43. Repeat control.

  That is, in the first embodiment, while changing the value of the fine density parameter as the layout analysis parameter, re-determination (re-classification) for the third block region is repeatedly performed until, for example, a predetermined end condition is satisfied, It is classified into either the first block area or the second block area.

  The predetermined termination condition includes, for example, until a predetermined number of times (for example, an upper limit value) is reached, or until the third block area disappears, but is not limited thereto. When the predetermined number of times described above is reached, the first block area, the second block area, and the third block area at that time may be output, and at that time, only the first block area set is output. May be output. Further, when the third block exists, a process of notifying the user of an error message, a message requesting the most input of the user designated area, or the like may be performed.

<Fine density parameters and specific examples of region classification based on fine density parameters>
Here, a fine density parameter and a specific example of region classification based on the fine density parameter will be described. FIG. 5 is a diagram illustrating an example of a layout analysis result by the separator. FIG. 6 is a diagram showing a specific example of the fine density parameter. FIG. 7 is a diagram illustrating an example of extraction of a blank separator. 5A to 5C show document images such as newspaper articles, as an example.

  For example, the layout analysis unit 42 described above detects a ruled line from the image by a ruled line cutout process included in the entire image, and further divides a block by a blank area detection process for the input image shown in FIG. Detect areas. The ruled line and the blank area described above function as a separator that separates the areas, and for example, as shown in FIG. 5B, a ruled line separator (black rectangle) and a blank separator (white rectangle) can be acquired. Therefore, in the first embodiment, an image is divided by these separators, and a block area as shown by a thick line frame in FIG. 5C is recognized by enclosing an area where pixels other than the background exist in each divided area. can do.

  Here, for example, in the process of extracting a blank rectangle, a blank portion having a predetermined width or more and a predetermined length or more is traced. When the predetermined width or predetermined length is a constant value or an inappropriate value, the result of layout analysis changes greatly. Therefore, in the first embodiment, a value called a fine density parameter is used as a layout analysis parameter, and a value such as a predetermined width is changed. In the first embodiment, only the predetermined width is changed to a different value, and the predetermined length is a constant value. However, the present invention is not limited to this. For example, the predetermined width and the predetermined length are changed to different values. Alternatively, the predetermined width may be a constant value, and only the predetermined length may be changed to a different value.

  Specifically, as shown in FIG. 6, when the level of the fine density parameter is increased from 1 to 5, the predetermined width of the blank rectangle is increased from 30 pixels to 10 pixels in a predetermined number of pixels (for example, 5 pixels). Use a setting that decreases. The type of level is not limited to that shown in FIG. Further, the predetermined number of pixels is not limited to five pixels, and may not be a predetermined number of pixels, and the predetermined width can be arbitrarily changed. In the first embodiment, as described above, it is possible to divide an appropriate block region by giving a range to the fine density parameter level.

Note that, as described above, in the method of dividing a region using a separator and recognizing the divided region as each block region, it is necessary to determine how much white space is regarded as a separator. In that case, for example, as a blank separator, as shown in FIG. 7, a maximum blank rectangle is obtained from a set R _i (i = 1,..., N) of a plurality of blank rectangles, and the size of the blank rectangle is determined. The process is terminated when a predetermined value (parameter) is reached. Thereby, a blank separator can be extracted appropriately. That is, when the above method is used, a smaller layout element can be obtained by reducing the pixel size parameter for extracting the separator.

  FIG. 8 is a diagram for explaining a difference in operation depending on layout analysis parameters. As described above, in the method of extracting a block rectangle by extracting a blank rectangle from an image, as shown in FIG. 8A, when the fine density parameter is small, the title (right A column) and the text are divided. Divided. On the other hand, as shown in FIG. 8B, when the fine density parameter is large, a narrow space between lines is extracted, so that the block region is divided more finely. As described above, in the first embodiment, a large block area can be divided more finely by changing the fine density parameter.

<Example of classification by block area classification means>
Here, a classification example of the first to third block areas by the block area classification means 43 described above will be described with reference to the drawings. FIG. 9 is a diagram illustrating a classification example of block areas in the first embodiment.

  In the example shown in FIG. 9, how the block area 52 is classified by comparing the user-specified area 51 specified by the user and the block area 52 obtained by the layout analysis with respect to the entire image 50. explain. In FIG. 9, the block area classifying unit 43 first obtains an overlapping area using the position of the user designated area 51 and the position of the block area 52 to be determined. The block area classification 43 has two threshold values set in advance.

  Next, the block area classification unit 43 compares the obtained overlap area with the first threshold value, and if the ratio of the overlap area to the entire block area is equal to or greater than the first threshold value (for example, 85%), the block area 52 is determined as the first block area. The block area classification 43 classifies the block area 52 as the second block area when the ratio of the overlapping area to the entire block area is equal to or less than a second threshold (for example, 10%). Furthermore, when the first and second threshold values do not belong to the first block area or the second block area, the block area 52 is classified as an undetermined area as the third block area. The first threshold value and the second threshold value described above can be changed as appropriate according to the contents of the entire image, the block size by the layout analysis, the number of blocks, and the like. As described above, the block area classification unit 43 can classify the block area based on the degree of overlap with the user-specified area.

<Image Processing Procedure in First Embodiment>
Here, the image processing procedure in the first embodiment will be described with reference to a flowchart. Note that the image processing shown below is processing that is executed in cooperation by installing, for example, an image processing program (software) in hardware resources as shown in FIG.

  FIG. 10 is a flowchart illustrating an example of image processing in the first embodiment. In the example shown in FIG. 10, in the image processing, an image to be cut out of a partial image is input (S01), and a user designated area in which a portion to be cut out is designated by a user or the like with respect to the image input in the processing of S01. Input (S02).

  Next, in the image processing, the entire image is set as a processing target area (S03), and layout analysis parameters are set to initial values (S04). Here, the layout analysis parameter means, for example, the fine density parameter described above, but is not limited thereto. The initial value means, for example, a value with the smallest fine density parameter level (the predetermined width (number of pixels) is the largest), but is not limited to this.

  Next, in the image processing, the layout is analyzed based on the layout analysis parameter obtained in the process of S04 (S05), and one or a plurality of block areas obtained as an analysis result are any of the first to third described above. (S06) (first block determination). In the process of S06, for example, based on position information (for example, coordinate information set on the image, etc.) between each block area and the user-specified area obtained in S02, the user-specified area is included. Classification is made into a first block area, a second block area not included in the user-specified area, and a third block area (undetermined area) not belonging to the first and second block areas. In the image processing, the first block area and the second block area obtained by the process of S06 are determined (S07).

  Next, the image processing determines whether there is a third block area (undefined area) that does not belong to the first block area or the second block area (S08). Here, in the image processing, when the third block area exists (YES in S08), it is determined whether or not the number of repetitions reaches a preset upper limit value (S09). In the image processing, when the number of repetitions has not reached the upper limit (NO in S09), the third block area is set as a processing target area (S10). In the image processing, the layout analysis parameter is updated to a value with a high fine density (that is, a parameter having a small predetermined width (number of pixels)) (S11), and the process returns to the process of S05. Reclassification is performed (second block determination).

  Further, in the image processing, if the third block area does not exist (NO in S08), or if the number of repetitions has reached the upper limit (YES in S09), the first obtained at that time point A partial image is generated from the set of block areas (S12), and the generated partial image is output (S13).

  In other words, in the first embodiment, a block area whose boundary line does not coincide with the area specified by the user (that is, the third block area) is divided more finely by changing the fine density parameter. A partial image can be acquired.

  In the first embodiment described above, the degree of subdivision of the layout analysis can be controlled by resetting the parameters. For this reason, for layout elements (block areas) that do not match the user-specified area, an appropriate partial image can be acquired by changing this parameter to a more subdivided direction and re-executing the layout analysis. .

  In the above-described example, the example in which the predetermined width (the number of pixels) is decreased stepwise has been described as a method for using the fine density parameter. However, the present invention is not limited to this. Specifically, in the first embodiment, for example, the predetermined width (number of pixels) is first reduced and then increased stepwise so that layout analysis is performed by sequentially integrating the small block regions obtained first. It can be carried out.

  FIG. 11 is a diagram illustrating an example of layout analysis by group integration in the first embodiment. As shown in FIG. 11, the layout analysis unit 42 can analyze a layout by setting a small region such as a character and integrating these regions to generate a large region. For example, in the example of FIG. 11, the character is divided into character units (FIG. 11A), the predetermined width (number of pixels) is increased stepwise, and adjacent characters within a predetermined distance are integrated as rows (FIG. 11). (B)) Further, the same processing is performed for a plurality of rows within a predetermined distance to integrate the regions. Thereby, in the first embodiment, a block area as shown in FIG. 11C can be generated.

  When using the method shown in FIG. 11, for example, whether or not adjacent characters and lines are to be integrated is determined using a predetermined parameter or the like. Specifically, for example, it is assumed that a standard such as “integration when the interval between characters is smaller than 50% of the average value of each character size” is set. In this case, layout analysis is performed based on the layout analysis parameter “smaller than 50% of the average value of each character size”. In the above example, if the value of 50% is increased, a larger layout element can be obtained.

  Similarly, in the case of integrating the rows, for example, “the rows are integrated when the interval between the rows is smaller than 70% of the average value of the row heights”, and the like. Parameters can be set.

<Specific example of re-input instruction for user specified area>
In the first embodiment described above, there is a case where an undetermined region (third block region) exists even if the layout analysis parameter is changed to a different value and repeated a predetermined number of times. In such a case, in the first embodiment, it is possible to perform processing for notifying the user of a message requesting re-input of the user designated area.

  Here, FIG. 12 is a diagram illustrating an example of a screen that prompts re-input of the user-specified area. 12A shows a display example on a display (large screen) such as a PC, and FIG. 12B shows a display example on a small screen such as a portable terminal.

  For example, if the third block area exists even after repeating a predetermined number of times, a screen requesting the user to re-enter the area is displayed on the display unit 33.

  For example, in the case of a large screen 60 such as a PC display shown in FIG. 12A, the original image is left only in the third block area, and the other areas are filled as defined areas (FIG. 12A- ( 2)), request the user to input again. In this case, the request message 61 to the user is preferably displayed, for example, outside the entire image 62 displayed in the screen 60 so that the third block area is not hidden (FIG. 12A). (3)), however, the display position and contents are not limited to this. Further, in the case of a small screen 63 such as a cellular phone shown in FIG. 12B, for example, a block area that has been confirmed is not displayed and only a block area 64 for re-inputting the area is displayed so that coordinates can be input in detail. It can be enlarged. The request message 65 to the user is preferably displayed with a short sentence at the bottom of the screen 63 (FIGS. 12B to 12), but the display position and contents are not limited to this. .

  As described above, by re-inputting the user-specified area, it is possible to eliminate the area that is not confirmed and generate a more appropriate partial image. Note that the request messages 61 and 65 described above may be output by voice or the like.

<Specific Example of Partial Image Obtained by First Example>
Here, a specific example of the partial image obtained by the first embodiment will be described with reference to the drawings. FIG. 13 is a diagram for explaining a specific example of the partial image obtained by the first embodiment.

  In the first embodiment, for example, the user designates a user designation area 71 as shown in FIG. In the first embodiment, it is assumed that block regions 72-1 to 72-5 as shown in FIG. 13B are obtained by layout analysis. Next, in the first embodiment, the block areas 72-1 to 72-3 (shaded area) inside the user-specified area 71 and the block area 72-5 (white area) outside are classified. To do.

  At this time, the block area 72-4 located in the lower right of FIG. 13C cannot be determined whether it is inside or outside the user-specified area, and becomes an undetermined area (third block area). Yes. Therefore, in the first embodiment, the layout analysis is executed again only for this undefined area. In this case, layout analysis is executed by changing layout analysis parameters so that the block area is subdivided.

  As a result, in the first embodiment, as shown in FIG. 13D, the block area 72-4 is further finely divided (block areas 72-4a and 72-4b), and finally, as shown in FIG. As described above, all the regions are classified as either inside or outside the user-specified region. Therefore, in the first embodiment, a set of the inner block areas 72-1 to 72-4a (shaded areas in FIG. 13E) is generated as a partial image, as shown in FIG. 13F. A desired partial image designated by the user can be obtained.

<Block area determination means: second embodiment>
Next, a second embodiment of the block area determination means described above will be described with reference to the drawings. In the second embodiment, a plurality of block area determination processes are performed from the beginning using a plurality of preset layout analysis parameters, and the final result is directly acquired.

  FIG. 14 is a diagram showing a functional configuration of the block area determination means in the second embodiment. Note that the schematic configuration of the image processing apparatus having the block area determination unit in the second embodiment is substantially the same as that in FIG. 2 described above, and thus the description thereof is omitted here. In the following description, components that perform substantially the same operations as those in the first embodiment described above are denoted by the same reference numerals, and detailed description thereof is omitted here.

  14 includes a repeat control means 81-1, 81-2, a fine density parameter setting means 41, a layout analysis means 42, a block area classification means 43, a block area determination means 44, And an overlapping block deleting means 82. The block area determination unit 80 of the second embodiment is replaced with the block area determination unit 13 in the image processing apparatus 10 shown in FIG.

  In the second embodiment, two repetitive control means 81-1 and 81-2 are provided. The repetitive control means 81-1 performs repetitive control so that the layout analysis for the entire image is performed under all conditions (levels) of preset fine density parameters. Therefore, as the layout analysis result output by the layout analysis means 42, a plurality of analysis results performed using each fine density parameter are output.

  The block area classification unit 43 performs block area classification on the plurality of analysis results output as described above. Further, the block area determination unit 44 determines any one of the first to third block areas from the above-described block areas.

  Note that the overlapping block deletion unit 82 deletes a block area that overlaps with a block area that has already been determined to be the first block area or the second block area by another fine density parameter.

  With the above-described configuration, unlike the first embodiment, the second embodiment performs layout analysis from the beginning with a plurality of fine density parameters (layout analysis parameters) (for example, parameter ID = 1 to N). The areas 1 to 3 are acquired directly.

  Therefore, in the second embodiment, for example, the first to third block areas described above are obtained for each block area obtained using a fine density parameter having a small level (a predetermined width (number of pixels)) at first. Classify into: Thereafter, in the second embodiment, each block area obtained using the following parameter level is classified into the above-described first to third block areas. That is, in the second embodiment, for example, for each block region obtained using all the parameter levels while increasing the parameter level stepwise (decreasing the predetermined width (number of pixels) stepwise) It classify | categorizes into the 1st-3rd block area | region mentioned above.

  Next, in the second embodiment, the first block area and the second block area obtained at the level where the fine density parameter is the smallest (the largest predetermined width) are determined. In the second embodiment, the block is determined using the classification result for the block area obtained at the next parameter level. At this time, in the second embodiment, in the classification result of the next level, the block area is deleted for the block area that overlaps the block area in which the first block area or the second block area is already determined.

  In other words, in the second embodiment, those that overlap with the confirmed block are deleted, and this process is repeated as many times as the number of parameters, whereby the first block area is determined with high accuracy and an appropriate partial image is acquired. be able to. In the second embodiment, even when the number of parameters is repeated, if the third block area exists, the user may be requested to re-input the area as described above, for example.

<Image Processing Procedure in Second Embodiment>
Here, the image processing procedure in the second embodiment will be described with reference to a flowchart. Note that the image processing shown below is processing that is executed in cooperation by installing, for example, an image processing program (software) in hardware resources as shown in FIG.

  FIG. 15 is a flowchart illustrating an example of image processing in the second embodiment. In the example shown in FIG. 15, in the image processing, an image to be cut out of a partial image is input (S 21), and a user specified area in which a portion to be cut out is specified by a user or the like with respect to the image input in the processing of S 21. Input (S22).

  Next, in the image processing, the entire input image is set as a processing target area (S23), and the layout analysis parameter is set to an initial value (ID = 1) (S24). Here, as the initial value in the second embodiment, for example, the above-described fine density parameter having the smallest level (the largest predetermined width (number of pixels)) is set as ID = 1, but is not limited thereto. Is not to be done.

  Next, in the image processing, the layout is analyzed based on the layout analysis parameter obtained in the processing of S24, and the result is stored in a buffer or the like (S25). The buffer may be provided in the layout analysis unit 42 or the like, and may have a buffer storage unit.

  Next, in the image processing, it is determined whether or not analysis is performed with all parameters (ID == N?) (S26). When analysis is not performed with all parameters (NO in S26), the next parameter ( ID = ID + 1) is set (S27) and the process returns to S25. Further, in the image processing, when analysis is performed with all parameters (YES in S26), the image processing is classified into one or a plurality of block regions from each analysis result (S28). In the process of S28, the first block area, the second block area, and the third block area described above based on the position information of each block area and the user-specified area obtained in S22. (Undetermined area).

  Next, in the image processing, the layout analysis parameter is set to an initial value (ID = 1) (S29), and the classification of the first and second block areas classified by the current parameter ID is determined (S30). Next, the image processing deletes a block that overlaps the determined block area (S31). Note that when the process of S31 is performed for the first time, there is no overlapping block because the process has not been repeated yet.

  Next, it is determined whether or not the image processing has been analyzed with all parameters (ID == N?) (S32). If the analysis has not been performed with all parameters (NO in S32), the next parameter (ID = ID + 1) is set (S33), and the process returns to S30.

  Further, when the image processing is analyzed with all parameters (YES in S32), it is determined whether or not the third block area exists (S34). When the third block exists (YES in S34), the image processing outputs a message requesting the user to re-enter the area (S35).

  In the image processing, if the third block area does not exist (NO in S34), a partial image is generated from the set of first block areas (S36), and the generated partial image is output (S37).

  That is, in the second embodiment, the block area can be appropriately classified while changing the fineness of the block area, and an appropriate partial area is extracted based on the analysis result using a plurality of fine density parameters. Can do. In addition, about the process of S35 mentioned above, it waits until a user designation | designated area | region is input again, and may return to the process of S23, after being input again.

<Block area determination means: third embodiment>
Next, a third embodiment of the block area determination means described above will be described with reference to the drawings. For example, if the third block area, which is an undetermined area, is, for example, a figure, table, image, or the like (hereinafter collectively referred to as “chart”), it is not further divided even if the parameter is changed . Therefore, in the third embodiment, when the third block area is a chart, it is forcibly classified into the first block area or the second block area.

  FIG. 16 is a diagram showing a functional configuration of the block area determination means in the third embodiment. Note that the schematic configuration of the image processing apparatus having the block area determination unit in the third embodiment is substantially the same as that in FIG. 2 described above, and thus the description thereof is omitted here. In the following description, components that perform substantially the same operations as those in the first embodiment described above are denoted by the same reference numerals, and detailed description thereof is omitted here.

  16 includes a fine density parameter setting means 41, a layout analysis means 42, a block area classification means 43, a block area determination means 44, a repetition control means 45, and a chart block area classification means 91. And have. The block area determination unit 90 of the third embodiment is replaced with the block area determination unit 13 in the image processing apparatus 10 shown in FIG.

  The chart / block area classification unit 91 determines the first block area if the overlapping area (within the enclosed area) with the user-specified area is 50% or more with respect to the block area including the chart, for example. Further, the chart block region classification unit 91 sets the second block region if the overlapping area with the user designated region is less than 50%, for example, with respect to the block region including the diagram.

  That is, when the block area classification unit 43 determines that the block area contains a chart, the chart block area classification unit 91 assigns the block area to the first block area using a predetermined threshold value. Alternatively, it is forcibly classified into the second block area. The chart / block area classifying unit 91 outputs information on the classified block areas to the block area classifying unit 43. Thereby, the block area classification means 43, if the third block area corresponds to the chart, based on the classification result obtained by the chart block area classification means 91, the first block area or the second block It can be forcibly classified into regions. Whether or not the third block area is a chart is determined based on the position information, for example, when the position information of the chart is included in an image input in advance by the image input means 11 or the like. Although it can be determined, it is not limited to this. For example, the block area classification unit 43 determines the presence / absence of a chart in the block area by performing pattern matching with a preset chart pattern, image analysis for searching for a human face, or the like on the input image. can do.

  In other words, in the third embodiment, when a chart is included in the block area, it is possible to quickly classify whether or not the block area belongs to the user designated area. Also in the third embodiment, as in the above-described embodiment, when the third block area exists, the layout analysis parameter is changed to a different value and the block area is re-determined (reclassified). Therefore, an appropriate partial area can be extracted.

<Image Processing Procedure in Third Embodiment>
Here, the image processing procedure in the third embodiment will be described with reference to a flowchart. Note that the image processing shown below is processing that is executed in cooperation by installing, for example, an image processing program (software) in hardware resources as shown in FIG.

  FIG. 17 is a flowchart illustrating an example of image processing in the third embodiment. In the example shown in FIG. 17, in the image processing, an image to be cut out of a partial image is input (S41), and a user specified area in which a portion to be cut out is specified by a user or the like with respect to the image input in the processing of S41. Input (S42).

  Next, in the image processing, the entire image is set as a predetermined target area (S43), and layout analysis parameters are set to initial values (S44). Next, in the image processing, the layout is analyzed based on the layout analysis parameter obtained in the process of S44 (S45), and one or a plurality of block areas obtained as an analysis result are any of the first to third described above. These block areas are classified (S46).

  Here, in the image processing, if the block area including the chart is classified into the third block area, it is classified into the first block area or the second block area using the threshold value described above (S47). The first block area and the second block area are determined (S48).

  Next, the image processing determines whether there is a third block area (undefined area) that does not belong to the first block area or the second block area (S49). In the image processing, if the third block area exists (YES in S49), it is determined whether or not the number of repetitions has reached the upper limit (S50). In the image processing, when the number of repetitions has not reached the upper limit (NO in S40), the third block area is set as a process target area (S51). In the image processing, the layout analysis parameter is updated to a value with a high fine density (that is, a parameter with a small number of pixels) (S52), and the process returns to the process of S45 to perform layout analysis and block area reclassification again. Do.

  Further, in the image processing, when the number of repetitions reaches the upper limit value in the processing of S50 (YES in S50), when there is a third block (YES in S34), the user is requested to re-enter the area. Message is output (S53).

  Further, in the image processing, when the third block area does not exist in the process of S49 (NO in S49), a partial image is generated from the set of the first block areas at that time (S54). A partial image is output (S55).

  That is, in the third embodiment, when a chart is included in the third block area, it can be quickly classified into the first block area or the second block area, so that a partial image is easily generated. be able to. In addition, about the process of S53 mentioned above, it waits until a user designation | designated area is input again from a user, and after re-inputting, you may return to the process of S43.

  Here, the first to third embodiments described above can be applied in combination. In each of the embodiments described above, one user designated area is set for one image. However, the present invention is not limited to this, and a plurality of user designated areas may be set for one image. . In that case, the above-described processing is performed for each user-specified area, and each partial image is output.

  As described above, according to each embodiment, it is possible to acquire an appropriate partial image corresponding to a specified region from a predetermined image. Further, in each of the above-described embodiments, for example, a partial image in which a desired area is appropriately cut out can be easily generated without specifying the area to be trimmed (partially cut out) accurately. In each of the embodiments described above, even if there is an error in the layout analysis to be executed, a more appropriate result can be obtained by changing the layout analysis parameter to a different value and re-executing the layout analysis. Therefore, in each of the above-described embodiments, it is possible to generate a partial image of a more appropriate region as compared with the related art.

  Each embodiment has been described in detail above, but is not limited to a specific embodiment, and various modifications and changes can be made within the scope described in the claims.

In addition, the following additional remarks are disclosed regarding the above Example.
(Appendix 1)
Specify a partial image area to be acquired from a given image,
Based on the analysis parameter for analyzing the image layout, the predetermined image is recognized as one or a plurality of block areas, the recognized one or a plurality of block areas are compared with the specified area, and the specified area is specified. The first block determination is performed to classify the block into any one of a first block area included in the area, a second block area not included in the specified area, and a third block area that is an indeterminate area. ,
A second block that is classified into the first block area or the second block area after changing the analysis parameter for the block area determined to be the third block area and re-recognizing the block area. Block judgment of
An image processing program for causing a computer to execute a process of generating a partial image based on a set of first block areas obtained by the first block determination and the second block determination.
(Appendix 2)
The image processing program according to appendix 1, wherein the second block determination is repeated until a predetermined end condition is satisfied.
(Appendix 3)
Specify a partial image area to be acquired from a given image,
Recognizing the predetermined image as one or a plurality of block areas using a plurality of analysis parameters for analyzing an image layout, comparing the recognized one or a plurality of block areas with the designated area, and Block determination to be classified into any one of the first block area included in the specified area, the second block area not included in the specified area, and the third block area that is an indeterminate area for each analysis parameter To
When determining the first block area, the second block area, and the third block area using the block determination result for each analysis parameter, the first block area or the second block area has already been determined. An image processing program for causing a computer to execute a process of deleting a block area overlapping with a block area determined as a block area.
(Appendix 4)
The image processing program according to any one of appendices 1 to 3, wherein a fine density parameter for controlling how finely the block area is recognized is set as the analysis parameter.
(Appendix 5)
The image processing program according to appendix 4, wherein the number of pixels of a blank rectangular area between the block areas is set as the fine density parameter.
(Appendix 6)
One of the first block area, the second block area, and the third block area by comparing a ratio of the designated area included in the entire block area with a preset threshold value. The image processing program according to any one of appendices 1 to 5, wherein the image processing program is classified into:
(Appendix 7)
When the third block area is a block area including a chart, the third block area is classified into the first block area or the second block area according to the ratio of the designated area included in the entire block area. The image processing program according to any one of appendices 1 to 6, characterized by:
(Appendix 8)
The image processing according to any one of appendices 1 to 7, wherein when the third block area exists after the second block determination, a request for re-designating the partial image area is output. program.
(Appendix 9)
Specify a partial image area to be acquired from a given image,
Based on the analysis parameter for analyzing the image layout, the predetermined image is recognized as one or a plurality of block areas, the recognized one or a plurality of block areas are compared with the specified area, and the specified area is specified. The first block determination is performed to classify the block into any one of a first block area included in the area, a second block area not included in the specified area, and a third block area that is an indeterminate area. ,
A second block that is classified into the first block area or the second block area after changing the analysis parameter for the block area determined to be the third block area and re-recognizing the block area. Block judgment of
An image processing method comprising: processing for generating a partial image based on a set of first block areas obtained by the first block determination and the second block determination.
(Appendix 10)
Area designating means for designating an area of a partial image acquired from a predetermined image;
Based on the analysis parameter for analyzing the image layout, the predetermined image is recognized as one or a plurality of block regions, and the recognized one or a plurality of block regions are compared with the designated region obtained by the region designating means. Block area determination means for classifying the block area into any one of a first block area included in the specified area, a second block area not included in the specified area, and a third block area that is an undetermined area;
Partial image generation means for generating a partial image based on a set of first block areas obtained by the block area determination means,
The block area determination means changes the analysis parameter for the block area determined to be the third block area, and after re-recognizing the block area, the first block area or the second block area An image processing apparatus that is classified into block areas.

DESCRIPTION OF SYMBOLS 10 Image processing apparatus 11 Image input means 12 Area designation means 13, 80, 90 Block area determination means 13 Partial image generation means 21 Control part 22 Communication part 23 Main control part 24 Auxiliary storage part 25 Drive apparatus 31 Image input part 32 Coordinate input Unit 33 Display unit 34 Recording medium 41 Fine density parameter setting means 42, 81 Repetition control means 43 Block area classification means 44 Block area determination means 45 Repetition control means 50, 62, 70 Whole image 51, 71 User specified areas 52, 64, 72 Block area 60, 63 Screen 61, 65 Request message 82 Overlapping block deletion means 91 Chart block area classification means

Claims (8)

Specify a partial image area to be acquired from a given image,
Based on the analysis parameter for analyzing the image layout, the predetermined image is recognized as one or a plurality of block areas, the recognized one or a plurality of block areas are compared with the specified area, and the specified area is specified. The first block determination is performed to classify the block into any one of a first block area included in the area, a second block area not included in the specified area, and a third block area that is an indeterminate area. ,
A second block that is classified into the first block area or the second block area after changing the analysis parameter for the block area determined to be the third block area and re-recognizing the block area. Block judgment of
An image processing program for causing a computer to execute a process of generating a partial image based on a set of first block areas obtained by the first block determination and the second block determination.   The image processing program according to claim 1, wherein the second block determination is repeated until a predetermined end condition is satisfied. Specify a partial image area to be acquired from a given image,
Recognizing the predetermined image as one or a plurality of block areas using a plurality of analysis parameters for analyzing an image layout, comparing the recognized one or a plurality of block areas with the designated area, and Block determination to be classified into any one of the first block area included in the specified area, the second block area not included in the specified area, and the third block area that is an indeterminate area for each analysis parameter To
When determining the first block area, the second block area, and the third block area using the block determination result for each analysis parameter, the first block area or the second block area has already been determined. An image processing program for causing a computer to execute a process of deleting a block area overlapping with a block area determined as a block area.   One of the first block area, the second block area, and the third block area by comparing a ratio of the designated area included in the entire block area with a preset threshold value. The image processing program according to claim 1, wherein the image processing program is classified into:   When the third block area is a block area including a chart, the third block area is classified into the first block area or the second block area according to the ratio of the designated area included in the entire block area. The image processing program according to any one of claims 1 to 4, wherein:   6. The image according to claim 1, wherein when the third block area exists after the second block determination, a request to re-specify the area of the partial image is output. Processing program. Specify a partial image area to be acquired from a given image,
Based on the analysis parameter for analyzing the image layout, the predetermined image is recognized as one or a plurality of block areas, the recognized one or a plurality of block areas are compared with the specified area, and the specified area is specified. The first block determination is performed to classify the block into any one of a first block area included in the area, a second block area not included in the specified area, and a third block area that is an indeterminate area. ,
A second block that is classified into the first block area or the second block area after changing the analysis parameter for the block area determined to be the third block area and re-recognizing the block area. Block judgment of
An image processing method comprising: processing for generating a partial image based on a set of first block areas obtained by the first block determination and the second block determination. Area designating means for designating an area of a partial image acquired from a predetermined image;
Based on the analysis parameter for analyzing the image layout, the predetermined image is recognized as one or a plurality of block regions, and the recognized one or a plurality of block regions are compared with the designated region obtained by the region designating means. Block area determination means for classifying the block area into any one of a first block area included in the specified area, a second block area not included in the specified area, and a third block area that is an undetermined area;
Partial image generation means for generating a partial image based on a set of first block areas obtained by the block area determination means,
The block area determination means changes the analysis parameter for the block area determined to be the third block area, and after re-recognizing the block area, the first block area or the second block area An image processing apparatus that is classified into block areas.

Images