JPH08263654A - Picture connection component extraction device - Google Patents

Abstract

PURPOSE: To reduce a work area (memory capacity) which is required in an overlap means by obtaining a judgement result on a circumscribed rectangular area selected through the use of a processing result against the other circumscribed rectangular area which is previously processed. CONSTITUTION: A connection component circumscribed rectangle extraction part 101, an overlap connection component extraction part 102, an overlap connection component ordering part 103, a connection component division part 104 and a judgement part 105 are provided. Furthermore, the device is provided with a minute division area size judgement part 106, a label picture generation part 107, an outline tracking part 108, a minute division area judgement part 109 and a minute division area storage part 110. When the other circumscribed rectangular area which is overlapped with the selected circumscribed rectangular area and which is previously processed is present, a result diversion means obtains the judged result on the circumscribed rectangular area selected by using the judged result obtained in the processing against the other external rectangular area. Furthermore, a connection component corresponding to the selected circumscribed rectangular area is specified on the judged result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image connected component extraction apparatus used for so-called image cutting, which extracts a group of images from image information during image processing.

[0002]

2. Description of the Related Art A so-called labeling method for extracting an image connected region is important as a means for analyzing an image and extracting a solid portion from the image. For example, in the handwritten character recognition OCR, a labeling method is used in a preprocessing unit of a unit that cuts out a character image from a plurality of character images. Furthermore, this labeling method is widely applied to image processing application systems for inspection and monitoring.

Conventionally, the applicant of the present application has proposed the following image connected component extracting device. In this image connected region extracting apparatus, for example, a connected component A1 composed of pixels 1 and 2 (black dots) connected as shown in FIG. 15B from a binary image as shown in FIG. , A2 circumscribed rectangles are extracted as shown in FIG. This rectangle is, for example, as shown in FIG. 15D, the coordinates of the upper right vertex (x1, y1) and the coordinates of the lower left vertex (x2, y).
2). Then, the circumscribed rectangle corresponding to the extracted connected component A1 of the pixel 1 and the connected component A of the pixel 2
It is determined whether the circumscribed rectangles corresponding to 2 overlap. For connected components corresponding to non-overlapping rectangles, the coordinates of the upper right vertex and the lower left vertex are stored. On the other hand, the overlapping connected components A1 and A2 are further subdivided into rectangular regions that do not overlap with other connected components.
Then, the connected component (image 1
Or it is determined whether or not it belongs to 2). Then, the respective connected components A1 and A2 are expressed by calculation of sum, difference, etc. of the subdivided rectangular areas.

For example, in FIG. 15 (c), the connected component A1 of the pixel 1 uses the maximum rectangular area 1-1 and the rectangular area 1-2 contained therein, and (1-1)-(1-2) , The connected component A2 of the pixel 2 is obtained by using the largest rectangular area 2-1 and the rectangular areas 2-2 and 2-3 included therein. )-(2-2) + (2-3), which is expressed by the sum and difference of each rectangular area.

In the basic labeling method, the connected component of pixel 1 and the connected component of pixel 2 are labeled with different labels (1
Or 2) (label image; FIG. 15).
(See (b)) For example, if the number of characters to be recognized in each page increases, the number of labels also increases, and the amount of information expressing each connected component increases (in the case of 256 characters, each dot is represented by 8-bit information). Will be represented).

On the other hand, if the connected components are represented by the circumscribing rectangles of the dots (pixels) connecting the connected components, the amount of information representing each connected component can be reduced.

[0007]

In the above processing, a rectangular area obtained by subdividing a portion where the circumscribed rectangular areas corresponding to the connected components A1 and A2 are subdivided is included in either of the original connected components A1 or A2. Whether or not the label image is generated is determined based on the result of generating the label image. For example, in the case of the example shown in FIG. 15, whether or not the rectangular area 2-3 belongs to the connected component A2 is determined by creating a label image of the connected component A2 and determining that the label image of the rectangular area 2-3 is the connected component A2. It is determined based on whether or not it matches the label image.

As described above, in order to subdivide the portion where the connected regions overlap and determine which connected component each of the subdivided rectangular regions belongs to, a conventional method of creating a label image of the connected component Then, when the connected region is large and other connected regions are included in the connected region, the number of label images increases. As a result, the amount of information handled in the determination processing increases, and a large work area (memory area) is required to make this determination.

Therefore, an object of the present invention is to provide an image connected component extracting apparatus which can reduce a work area as much as possible when extracting connected components from an image.

[0010]

In order to solve the above-mentioned problems, according to the present invention, as described in claim 1, a target image is divided into connected components composed of connected pixels, and a circumscribed rectangle thereof is divided. A circumscribing rectangle extracting means for extracting a region, an overlapping circumscribing rectangle extracting means for extracting overlapping ones of the circumscribing rectangular areas extracted by the circumscribing rectangle extracting means, and a circumscribing rectangular area extracted by the overlapping circumscribing rectangle extracting means Select a circumscribed rectangular area to be processed in a predetermined order from, and divide the selected circumscribed rectangular area into rectangular areas, and select which connected component corresponds to the circumscribed rectangular area where the separated rectangular areas overlap. An image connected component extraction that has an overlapping portion processing unit that determines whether it should belong and that specifies a connected component corresponding to the selected circumscribed rectangular area based on the determination result. In this case, the overlapping portion processing means determines the selected circumscribed rectangular area using the determination result obtained by the processing for the already processed other circumscribed rectangular area overlapping the selected circumscribed rectangular area. A result diversion means for obtaining results was provided.

Further, in the image connected component extracting apparatus, when each circumscribing rectangular area is selected, as many other circumscribing rectangular areas that have already been processed exist as much as possible. As described above, the overlapping portion processing means has the area selecting means for selecting the circumscribed rectangular areas to be processed from the overlapping circumscribed rectangular areas in the ascending order of area.

Further, when a circumscribing rectangular area having a large area is to be processed, the circumscribing rectangular area is contained as much as possible in order to include the circumscribing rectangular area. The overlapping portion processing means has a means for preferentially selecting a circumscribed rectangular area included in another circumscribed rectangular area.

In order to solve the above problems, according to the present invention, as described in claim 7, the circumscribed rectangular area is extracted by dividing the target image into connected components composed of connected pixels. A circumscribing rectangle extraction means, an overlapping circumscribing rectangle extracting means for extracting overlapping ones of the circumscribing rectangle areas extracted by the circumscribing rectangle extracting means, and a predetermined order from the circumscribing rectangle area extracted by the overlapping circumscribing rectangle extracting means Select the circumscribed rectangular area to be processed in step 1, divide the selected circumscribed rectangular area into rectangular areas, and determine which connected component corresponding to the circumscribed rectangular area where the separated rectangular areas overlap. In the image connected component extraction device, which has an overlapping portion processing means for making a judgment, and which specifies a connected component corresponding to the selected circumscribing rectangular area based on the judgment result. The overlapping portion processing means performs contour tracing means for performing contour tracing starting from a predetermined pixel in the cut rectangular area, and based on the result of the contour tracing, the rectangular area is the selected circumscribed rectangle. And a determination unit that obtains a determination result as to whether or not it belongs to the connected component corresponding to the region.

Further, in the image connected component extracting apparatus described in claim 7, from the viewpoint that it is possible to easily determine which connected component the rectangular area belongs to, The determining means has means for determining whether or not the rectangle specified by the maximum value and the minimum value of the coordinates traced by the contour coincides with the circumscribed rectangle of the connected component. A rectangular area including the starting pixel for contour tracing is determined to be included in the connected component.

Further, in order to solve the above-mentioned problems, according to the present invention, as described in claim 11, the circumscribed rectangular area is extracted by dividing the target image into connected components composed of connected pixels. A circumscribing rectangle extraction means, an overlapping circumscribing rectangle extracting means for extracting overlapping ones of the circumscribing rectangle areas extracted by the circumscribing rectangle extracting means, and a predetermined order from the circumscribing rectangle area extracted by the overlapping circumscribing rectangle extracting means Select the circumscribed rectangular area to be processed in step 1, divide the selected circumscribed rectangular area into rectangular areas, and determine which connected component corresponding to the circumscribed rectangular area where the separated rectangular areas overlap. An image connected component extracting apparatus having an overlapping portion processing unit for making a determination, and for specifying a connected component corresponding to the selected circumscribed rectangular area based on the determination result. The overlapping portion processing means determines whether the size of the selected circumscribed rectangular area is larger than a predetermined value, and the size determination means determines the size of the selected circumscribed rectangular area to be predetermined. When it is determined that the rectangular area is larger than the value, a contour tracking unit that performs contour tracking starting from a predetermined pixel in the cut rectangular area, and the rectangular area is selected based on the result of the contour tracking. A determination unit that obtains a determination result as to whether or not it belongs to a connected component corresponding to a rectangular region, and the size determination unit determines that the size of the selected circumscribed rectangular region is not larger than a predetermined value. A label image creating means for creating a label image corresponding to a connected component included in the circumscribed rectangular area,
Based on the created label image, it has a means for determining which connected component the cut rectangular area belongs to.

[0016]

In the image connected component extracting apparatus according to claim 1, when there is another processed circumscribed rectangular area which overlaps the selected circumscribed rectangular area, the result diversion means performs processing for the other circumscribed rectangular area. A determination result for the selected circumscribed rectangular area is obtained using the determination result obtained as described above. Then, based on the determination result, the connected component corresponding to the selected circumscribed rectangular area is specified.

In the image connected component extraction device according to the present invention, the contour tracing means performs contour tracing starting from a predetermined pixel of the cut rectangular area, and based on the result, the judging means makes the rectangular area. Determines whether or not belongs to the connected component corresponding to the selected circumscribing rectangular area. Then, based on the determination result, the connected component corresponding to the selected circumscribed rectangular area is specified.

In the image connected component extraction device according to the eleventh aspect, when the size determination means determines that the area of the selected circumscribing rectangular area is larger than a predetermined value, the above-described contour tracing is performed to determine the area. Is done. In addition, when the size determining means determines that the area of the selected circumscribed rectangular area is not larger than the predetermined value, a label image corresponding to the connected component included in the selected circumscribed rectangular area is created, and its level image is generated. The area is determined based on the.

[0019]

Embodiments of the present invention will be described below with reference to the drawings. The image connected component extraction device of the present invention can be applied to, for example, a system for recognizing images such as characters.
FIG. 1 shows a basic configuration of a character recognition system to which an image connected component extraction device is applied.

In FIG. 1, this character recognition system is
CPU (central processing unit) 1 that performs various processes
0, a program memory 11 storing a program executed by the CPU 10, an image memory 12 storing image data in a bitmap format, and a work memory 13 used for processing the image data. The character recognition system further includes a scanner 1 that optically reads an image.
4, a memory 15 for temporarily storing information read by the scanner 14, a dictionary file 16 for storing characteristics of each character image, and a display unit 17 as an output device.
And a printer 18 and an interface circuit 19 for these output devices. The above-mentioned CPU 10, program memory 11, image memory 12, work memory 13,
The memory 15, the dictionary file 16 and the interface circuit 19 are connected to each other via a bus 20.

In the character recognition system as described above, the image data read by the scanner 14 is temporarily stored in the memory 15, and the image data is expanded in the image memory 12 in the bitmap format. And the image memory 1
Image connected components are extracted from the binary image data copied from 2 to the work memory 13. A character image is cut out from the read image based on the result, and the characteristics of the cut out character image and the dictionary file 1
The characteristic data stored in 6 is compared. Then, the character is recognized based on the comparison result. The recognition result is output to the display unit 17 or the printer 18.

In the above system, the image connected component extracting apparatus according to the present invention is realized as a function of the CPU 10 which performs processing in accordance with a program stored in the program memory 11, and its basic configuration is, for example, as shown in FIG. As shown. In FIG. 2, the image connected component extracting apparatus includes a connected component circumscribing rectangle extracting unit 101, an overlapping connected component extracting unit 102, and an overlapping connected component ordering unit 10.
3, a connected component subdivision unit 104 and a determination unit 105. The image connected component extraction apparatus further includes a subdivision area size determination unit 106, a label image generation unit 107, a contour tracking unit 108, a subdivision area determination unit 109, and a subdivision area storage unit 110.

The connected component circumscribed rectangle extraction unit 101 calculates the circumscribed rectangle of the connected component from the binary image developed on the work memory 13. Specifically, as shown in FIG. 15D, coordinate data of the upper left corner and lower right corner of the circumscribed rectangle are calculated. The overlap connection extraction unit 102 extracts, of the connection component circumscribed rectangles extracted by the connection component circumscribed rectangle extraction unit 101, ones that overlap each other. The circumscribing rectangles that do not overlap store the coordinate data of the upper left vertex and the lower right vertex for specifying the rectangles. The overlapping connected component ordering unit 103 calculates the area of the circumscribed rectangle of each connected component, and sorts the connected components in ascending order of area. The connected component subdivision unit 104 searches for an overlapping connected component from a connected component having a small area, and subdivides the overlapped connected component into rectangular regions so as not to overlap other regions. The determination unit 105 determines whether each subdivision area has already been processed. For the subdivision area for which the determination unit 105 has determined that the processing has already been completed, the processing result stored in the subdivision area storage unit 110 is stored in the subdivision storage unit 110 as a new processing result. (). On the other hand, for the subdivision area that the determination unit 105 has determined that the processing has not been completed, data regarding the area is provided to the subdivision area size determination unit 106 ().

The subdivision area size determination unit 106 calculates the size of the provided subdivision area. If it is determined that the work memory 12 has sufficient capacity even if the label image of the subdivided area is generated based on the size, the label image generation unit 107 generates the label image of the subdivided area. Then, the subdivision area determination unit 109 determines whether the label image (child label) of the subdivision area and the label image (parent label) of the connected component to be processed match. Based on the result, the connected component of the parent label is expressed by the sum and / or difference of the subdivided areas.

On the other hand, when the subdivision area size judgment unit 106 judges that the capacity of the work memory 12 is insufficient to create the label image of the subdivision area, the contour tracing is started from the pixel at the outermost portion of the subdivision area. Is done. The circumscribed rectangle of the label image corresponding to the subdivision area is obtained by calculating the maximum value and the minimum value of the tracking coordinates in contour tracking. Then, the subdivision area determination unit 109 checks whether or not the target parent label and the child label of the subdivision area match. Based on the result, the target connected component is represented by the sum and / or difference of the subdivided regions.

The data of the connected component expressed by the sum and / or the difference of the obtained subdivided areas is stored in the subdivided area storage unit 110. The subdivision area storage unit 110
For example, it is set on the work memory 13. In the image connected component extraction apparatus as described above, when processing the image shown in FIG. 15A, the image 1 having an area smaller than the circumscribed rectangle corresponding to the connected component A2 of the pixel 2 shown in FIG. The circumscribed rectangle corresponding to the connected component A1 is processed first. Therefore, the overlapping region 2-2 of the connected components A1 and A2 is processed when the connected component A1 is targeted. As a result, when the connected component A2 is targeted, the subdivided area 2-2 uses the previous processing result (belongs to the connected area A1) as it is, and in particular, this subdivided area 2-2 belongs to the connected component A2. It is not necessary to create a label image to determine the size. In order to further reduce the processing time, the connected component A
When 2 is the processing target, the overlapping area of the connected components A1 and A2 can be expressed as follows using the subdivided areas 1-1 and 1-2 representing the connected component A1.

A2 = + (2-1)-(1-1) + (1-2) In the processing example of the image shown in FIG. 15, the connected component A1
, The region 1-in the largest circumscribing rectangle 1-1.
Since only 2 is subdivided and the area 1-2 has a single label, it is only necessary to calculate the circumscribed rectangle of the image without generating a label image.

When processing a complicated image as shown in FIG. 3A, the connected components A1 and A2 of the pixel 1 and the pixel 2 are obtained as shown in FIG. 3B. Also in this case, in order to determine which connected component the overlapping portion of both connected components A1 and A2 belongs to, it is only necessary to create a label image for the smaller connected component A1. When determining whether the subdivided regions 1-3 belong to the connected component A1, the label image of the connected component A1 is generated. Furthermore, the area of the connected component A1 is large, and the connected component A1
When the capacity of the work memory 13 is insufficient to create the label image of, the contour tracking is performed. In the contour tracking processing, for example, the contour is traced starting from the pixels on the rectangular boundary of the subdivided areas 1-3. As a result, it is confirmed that the subdivided regions 1-3 belong to the connected component A1.

Further, the image connected component extracting apparatus according to the embodiment of the present invention will be specifically described. This image connected component extraction apparatus executes processing according to the procedures shown in FIGS. 4, 5, 6, and 7. This processing flow consists of recursive operations of two subroutines SUBJECT1 and SUBJECT2.

In FIG. 4, the binary image data expanded in the image memory 12 is copied to the work memory 13, and as shown in FIG. 8, labeling of the entire binary image in the work memory 13 is performed (block 501). ). In this labeling process, the circumscribed rectangle of the connected pixels (connected components) is calculated. In the example shown in FIG. 8, m areas (circumscribing rectangles) are labeled, and each circumscribing rectangle has coordinates (x1i, y1i) of its upper left vertex and coordinates (x2) of its lower right vertex.
i, y2i) (i = 1, 2, ..., M). A concrete method of this processing is, for example, Japanese Patent Application No. 5-2.
3755 proposed by the applicant. The number of labels (MAXLABEL = m number of circumscribed rectangles) obtained as a result of this labeling is stored in a predetermined area of the work memory 13, and coordinate data for specifying each label 1 to m (circumscribed rectangles 1 to m) As shown in FIG.
3 is stored in the circumscribed rectangle storage area.

When the extraction of each label is completed, it is checked whether or not there is an overlap between the labels based on the coordinate data stored in the circumscribed rectangle storage area (block 50).
2). The result is stored in the circumscribed rectangle overlapping table as shown in FIG. In this example, labels 3 and 4
Overlap and labels 5 and 6 overlap.

Thereafter, coordinate data specifying each label other than the above-mentioned overlapping label is stored in the non-overlapping part circumscribed rectangle storage area set in the work memory 13 as shown in FIG. 11 (block 503). In this example, labels 1, 2, 7, ..., M are stored as non-overlapping labels (circumscribing rectangles).

Referring to the circumscribing rectangle overlapping table shown in FIG. 10, overlapping labels are extracted, and each overlapping label is identified based on the coordinate data (the circumscribing rectangle storage area in FIG. 9) that identifies the overlapping label. Calculate the area of the label. Then, the overlapping labels (a) are sorted in ascending order of area and stored in a table as shown in FIG. 12 (block 504).

With reference to the table shown in FIG. 12, the label (b) having the smallest area among the overlapping labels is selected (block 505). In the case of this example, first, the label 3 having the smallest area is selected. Then, it is determined whether or not all the labels (a) sorted in the block 504 have been processed (block 506), and all the labels overlapping the label (b) selected in the block 505 have been processed. It is determined whether or not (block 507). And if those processes are not completed, the SU
The processing of BJECT1 is executed.

The processing of SUBJECT1 is executed according to the procedure shown in FIG. In FIG. 5, block 505
Labeling of the label (circumscribing rectangular area) (b) selected in (1) is performed (block 510). In this labeling, the circumscribed rectangle of the pixels connecting the images in the selected label is calculated in the same manner as the processing of block 501. For example, in the case of the image shown in FIG. 3A, the label 1 having the smaller area is selected from the labels (circumscribing rectangular areas) 1 and 2,
The selected label 1 is 1-1 in block 510.
And labeled 1-2. The label 1-1 is the same as the parent label 1.

One label (d) to be processed is selected from the labels (c) obtained as described above (block 51).
1). For example, in the case of the example shown in FIG. 3, the label 1-2 is selected from the labeled regions 1-1 and 1-2. Here, the label (c) obtained in block 510
(For example, 1-1, 1-2) whether or not all the processing has been completed (block 512), the processing of the label (d) (for example, 1-2) selected in block 511 has already been completed. It is determined whether or not (block 513).
Then, when those processes are not completed, whether or not there is already a label image of the label (b) (for example, A1 (1-1)) obtained in the block 505 to be processed is determined. A determination is made (block 515). Here, if the label image has not been generated yet, the label (d) selected in block 511 is further labeled (block 516). For example, in the case of the example shown in FIG.
The selected label 1-2 is labeled 1-2 and 1-3.

It is judged whether or not the condition that the label (e) obtained in this block 516 is not single and there is an unprocessed label in it (block 517). ), If such a condition is satisfied, it is further determined whether the area of the label (b) selected in block 505 is larger than the specified area (block 519). This is performed, for example, depending on whether or not the memory amount required for creating the label image of the label is a predetermined amount (for example, 64 kbytes) or less. If it is determined that the area is smaller than the designated area, the label image of the label (b) selected in block 505 is created (block 521). Since the area of the label (b) is smaller than the designated area, the label image can be expanded on the work memory 13.

When the label image of the label (b) selected in the block 505 is created in this way, thereafter, the processing is executed according to the procedure shown in FIG. In FIG. 6, coordinate data identifying the rectangular area of the label (d) extracted in the block 511 is described in the overlapping portion circumscribed rectangular storage area of the label (b) in the table as shown in FIG. 13 (block 522). ). Then, a-(minus) operator is given to the label (d). For example, in the case of the example shown in FIG. 3, the coordinate data identifying the rectangular area of label 1-2 extracted in block 511 is described in the overlapping portion circumscribed rectangular storage area of label 1-1.
Then, an operator-is given to this label 1-2.

Thereafter, it is determined whether or not the rectangular area of the label (d) processed as described above includes the label (b) selected in the block 505 (block 52).
3). In the case of this example, since the label (d) is completely included in the label (b), it is determined that the rectangular area of the label (b) is included in the rectangular area of the label (d). Then, the label (d) selected in block 511 is further labeled (block 52).
4). For example, for the example image of FIG. 3, block 511
The label 1-2 selected in 1 is further labeled to 1-2 and 1-3.

From the labels (f) thus obtained, one label (g) is selected (block 525). For example, for the example image shown in FIG. 3, labels 1-3 are selected from labels 1-2 and 1-3 obtained at block 514. After that, it is determined whether or not all the labels obtained in block 524 have been processed (block 526), and the label (g) selected in block 525 is determined in block 505.
It is determined whether or not it includes the label image of the label (b) selected in. The label (g) selected in this block 525 is the label (b) selected in block 505.
, The coordinate data specifying the rectangular area of the label (g) is described in the overlapping portion circumscribed rectangular storage area of the label (b) of the table as shown in FIG. 13 (block 528). Then, a + (plus) operator is added to the label (g). For example, in the case of the example of the image shown in FIG. 3, the labels 1-3 extracted in block 525 are displayed.
The coordinate data for specifying the rectangular area is described in the overlapping part circumscribed rectangular storage area of the label 1-1. Then, the operator + is added to the labels 1-3.

Since some labels (g) are not included in the label (b), there are cases where (g) corresponds to (b) and the process of SUBJECT1 is recursively executed. As a result of the above processing, the label specified as the overlapping label is expressed using the label (coordinate data) described in the overlapping portion circumscribed rectangle storage area of the label in the table of FIG. 13 and its operator. It
For example, in the case of the example of the image shown in FIG. 3, the connected component A1 of the pixel 1 is represented by the label 1-1 indicating the entire circumscribing rectangle and the labels 1-2 and 1-3 included therein. 1)-(1-2) + (1-3) is expressed as (1).

If the label (g) is not a part of the label image of the label (b) in the block 527 in the above processing, the next label (g) is selected from the labels (f) obtained in the block 524. (Block 52
5) Then, the same processing as described above is performed on the selected label. Further, when all the processing for the label obtained in block 524 is completed, or the rectangular area of the label (b) selected in block 505 is added to the rectangular area of the label (d) selected in block 511. If it is determined that a part of is not included, the process is SUBJE
Return to block 511 of CT1. And block 51
The next label is selected from the labels obtained at 0, and the same processing as described above is executed thereafter. In the case of the example of the image shown in FIG. 3, after the processing of the label 1-2 is finished, there is no label to be further processed in the label obtained in block 510 (block 512), so SUBUJ
The processing of ECT1 ends.

When the processing of SUBJECT1 is completed as described above, block 50 in the procedure shown in FIG.
Return to 5. Here, referring to the table shown in FIG. 12, the label (b) having the next smallest area is selected from the label (a) obtained in block 505. In the case of this example, label 4 is selected. Moreover, in the case of the example of the image shown in FIG.
Label 2-1 is selected. Then, all the processing for the label (a) obtained in the block 504 has not been completed yet, and the processing for all the labels overlapping the label (b) selected in the block 505 has ended. If not, as described above, SUBJECT
The process 1 is executed (block 509). For example, in the case of the example image shown in FIG. 3, since the labels obtained in block 504 are two labels 1 and 2, when label 2 is selected in block 505, the label is selected at that point. This means that the processing for all the overlapping labels (connected image area A1) has been completed.
Therefore, in this case, all the label numbers that overlap this label 2 are described in the overlapping part circumscribed rectangle storage area of the label 2 in the table as shown in FIG. 13 (block 508). Since the label 1 overlapping the label 2 has different label images, the operator- (minus) is given to the label 1 described in the storage area of the circumscribed rectangle of the overlapping portion of the label 2. As a result, label 2
The connected component A2 corresponding to is as follows: A2 = (2-1) -A1 = (2-1)-{(1-1)-(1-2) + (1-3)} (2) Expressed. In this case, in particular, SUBJECT
Even if the process 1 is not executed, the overlapping label can be described by using the result of the already processed label.

Even if it is determined in the processing of block 507 that the processing for all the labels is not completed, if the processing for some of the labels is completed, the processing of SUMJECT1 is performed. When the label included in the label for which the processing has been completed is selected as the label (d) in block 511, the processing shifts from block 513 to block 514. In this case as well, the label number identifying the already processed label (d) is directly written in the overlapping part circumscribed rectangle storage area of the label (b) to be processed.

Further, in the block 517 in the processing of SUBJECT1, if the condition that the label (e) obtained in the block 516 is not a single label and includes a label that has not been processed yet, Coordinate data for specifying the rectangular area of the label (d) obtained in the block 511 is described in the overlapping portion circumscribed rectangular storage area of the label (b) (block 518). In this case, since the label (d) is a label image different from the label (b), the operator- (minus) is given to the label (d).

If the label (b) to be processed is determined to be larger than the predetermined area in block 519 of the above-mentioned SUBJECT 1, the SUBJECT 2
The process of is executed. The process of SUBJECT2 is executed according to the procedure shown in FIG.

In FIG. 7, the label (d) selected in block 511 is processed in the same manner as block 518 (block 529). That is, the label (d)
The coordinate data for specifying the rectangular area is described in the overlapping area circumscribed rectangular storage area of the label (b). Label (d) is then labeled (block 530), similar to the processing in block 516. One label (i) is selected from the labels (h) thus obtained (block 531). Here, it is determined whether or not the processing has been completed for all the labels obtained in the block 530 (block 532). If the processing has been completed, the processing of SUBJECT2 ends.

On the other hand, if the processing has not been completed for all the labels, contour tracing of the label image of the label (i) selected in block 531 is performed (block 533). This contour tracking is based on known methods such as boundary line tracking (contour tracking) described in “Digital Image Processing II for Image Understanding II” by Shoichiro Torikiwa, PP.66-79 of Shokoido Co., Ltd. Is done. When the contour tracking of the label image of the label (i) is completed, whether or not the label (i) is a part of the label (b) that is the target of the processing is determined based on the result obtained by the contour tracking. Is determined (block 534), and if it is a part thereof, the coordinate data specifying the rectangular area of the label (i) is described in the overlapping part circumscribed rectangular storage area of the label (b) (block 535). .
Then, the operator + (plus) is given to the label (i). On the other hand, when the label (i) is not a part of the label (b), the next label (i) is selected from the labels (h) obtained in the block 530, and the same processing as above is performed.

In block 534, it is judged whether or not the label (i) is included in the parent label (b) based on the result of the contour tracing. A specific example will be described with reference to FIG. . In FIG. 14, the image 7 and the image 8 are distinguished for convenience, but the actual processing is performed on the binary image.

In FIG. 14A, the parent label (rectangular area) 8 is divided into rectangular areas 8-1, 8-2, 8-3, 8-4.
Composed of. In this case, when the divided rectangular area 8-3 is selected as the label (i) (block 531), the contour is started from the black pixel of the circumscribing rectangle, for example, as shown in FIG. Track in order from.
After the tracking, the maximum value of the tracked coordinate values (XMAX, YM
AX) and the minimum value (XMIN, YMIN). The maximum value (XMAX, YMAX) and the minimum value (XMIN, Y
Coordinate data (x18, MIN) specifying the parent label 8
y18) and (x28, y28) are determined. When these match, it is determined that the label 8-3 is included in the parent label 8. Pixels located at the boundary of the divided area 8-3 of interest and not belonging to other divided areas are selected as the starting pixels for the contour tracking.

Further, in the above contour tracing, when the target parent label 8 (circumscribing rectangular area) is exceeded in the course of the tracing, it is determined that the parent label 8 does not include the divided area including the pixels of the contour tracing. To be judged. When the above process is completed, label (i) is further labeled (block 53).
6). Then, one label (k) is selected from the labels (j) obtained by the labeling (block 537). Contour tracking of the selected label (k) is performed for all labels (j) obtained at block 536 (block 538). For each selected label (k), it is determined whether the label (k) is not part of the parent label (b) based on the results of its contour tracking (block 540). For the label (k) that is not a part of the parent label (b), SUBJECT2
Is processed recursively. This gives the label (k)
The labels within are processed further.

As described above, according to the present embodiment, an image is distinguished into a plurality of labels (rectangular areas), and if the obtained labels have no overlap, coordinate data for specifying the rectangular area is stored. To be done. Then, the overlapping labels are processed in ascending order of area, and in particular, label images are created only for those having a smaller area, and the structure of the overlapping labels is determined based on the result. Therefore, the work area can be reduced. Also, when processing overlapping labels,
Since the result obtained in the previous processing is used (see blocks 508, 514, and 518), high-speed processing can be realized while saving the work area.

For a label larger than a predetermined area, the structure of the overlapping label is determined based on the result of contour tracing, and in this case also, the work area can be saved. Although the coordinates of the upper left corner and the lower right corner of the rectangle are used to specify the circumscribed rectangular area of each label as described above, the present invention is not limited to this, and various specific methods are possible. .

In the above embodiment, the label (d) to be processed is selected in the ascending order from the overlapping label (b), but the labels included in other labels are preferentially selected. You may make it select. In this case, when processing a relatively large label including another label, the processing for the included label is always completed, and in processing the relatively large label,
The processing result of the included label can be used.

Although the above embodiment is applied to the character recognition system, the image connected area extracting apparatus according to the present invention can also be applied to other image processing systems.

[0056]

As described above, according to the invention of claim 1, the determination result for the selected circumscribed rectangular area is obtained by using the processing result for the other circumscribed rectangular area that has already been processed. Since this is done, it is possible to reduce the work area (memory capacity) required by the overlapping portion processing means. Moreover, the processing time is also shortened.

According to the seventh aspect of the invention, since the determination result is obtained based on the processing result by the contour tracing means, the required work area is reduced as compared with the case where the label image is generated. It becomes possible to reduce. Furthermore,
According to the eleventh aspect of the invention, in particular, the contour tracking is performed only when the processing is performed on the large circumscribed rectangular area, and the label image is generated in the processing on the small circumscribed rectangular area. Therefore, it is possible to determine the area without requiring an unnecessary work area.

[Brief description of drawings]

FIG. 1 is a block diagram showing a basic configuration of a character recognition system.

FIG. 2 is a block diagram showing a basic configuration of an image connected component extracting apparatus according to an embodiment of the present invention.

FIG. 3 is a diagram showing a relationship between an image and a connected component.

FIG. 4 is a flowchart (part 1) showing a processing flow for extracting a connected region.

FIG. 5 is a flowchart (part 2) showing a processing flow for extracting a connected region.

FIG. 6 is a flowchart (part 3) showing a processing flow for extracting a connected region.

FIG. 7 is a flowchart (No. 4) showing a processing flow for extracting a connected region.

FIG. 8 is a diagram showing an example of a labeled image.

FIG. 9 is a diagram showing a storage area of coordinate data that specifies a circumscribed rectangular area of each label.

FIG. 10 is a table showing overlapping of circumscribed rectangular areas.

FIG. 11 is a table showing circumscribing rectangular areas that do not overlap.

FIG. 12 is a table showing each ordered label (circumscribing rectangular area).

FIG. 13 is a table showing an overlapping state of labels and an operator assigned to an overlapping portion.

FIG. 14 is a diagram showing the principle of determination of label matching by subdivided images and contour tracking.

FIG. 15 is a diagram showing a relationship among an image, a connected component, and a circumscribed rectangular area.

[Explanation of symbols]

 10 CPU 11 program memory 12 image memory 13 work memory 14 scanner 15 memory 16 dictionary file 17 display unit 18 printer 19 interface circuit 101 connected component circumscribed rectangle extraction unit 102 overlapping connected component extraction unit 103 overlapping connected component ordering unit 104 connected component subdivision Unit 105 determination unit 106 subdivision region size determination unit 107 label image generation unit 108 contour tracking unit 109 subdivision region determination unit 110 subdivision region storage unit

Claims (13)

[Claims] 1. A circumscribing rectangular area extracting unit that extracts a circumscribing rectangular area by dividing a target image into connected components composed of connected pixels and an circumscribing rectangular area extracted by the circumscribing rectangular area extracting unit overlaps each other. The circumscribing circumscribed rectangular area to be processed is selected in a predetermined order from the circumscribing circumscribing rectangle extracting means for extracting a matching object and the circumscribing rectangular area extracted by the circumscribing circumscribing rectangle extracting means. The region is divided, and an overlapping portion processing unit that determines which connected component corresponding to the circumscribed rectangular region in which the divided rectangular region overlaps is included, and based on the determination result, the selected In the image connected component extracting apparatus configured to identify the connected component corresponding to the circumscribed rectangular area, the overlapping portion processing means is configured to detect the connected component that has already overlapped with the selected circumscribed rectangular area. Image connected component extracting apparatus having the results diversion means for obtaining a judgment result for the selected circumscribed rectangular area using the determination result obtained by the processing for the physical and other circumscribed rectangular area. 2. The image connected component extracting apparatus according to claim 1, wherein the result diversion means includes a case where a circumscribed rectangular area of the processed connected component is present in the rectangular area obtained by the division. An image connected component extraction apparatus having means for using the determination result used to identify the processed connected component. 3. The image connected component extraction apparatus according to claim 1, wherein the result diversion means includes a rectangular area obtained by cutting and there is a circumscribed rectangular area of the already processed connected component. An image connected component extracting apparatus having means for using all the determination results for the processed circumscribing rectangle. 4. The image connected component extraction apparatus according to claim 1, wherein the result diversion means includes a rectangular area obtained by cutting, and there is a circumscribed rectangular area of the already processed connected component. An image connected component extraction device having means for using the result of the region of the overlapped portion among the processed determination results for the included circumscribed rectangle. 5. The image connected component extracting apparatus according to claim 1, wherein the overlapping portion processing means selects a circumscribing rectangular area to be processed from overlapping circumscribing rectangular areas in ascending order of area. An image connected component extracting apparatus having: 6. The image connected component extracting apparatus according to claim 1, wherein the overlapping portion processing means preferentially selects a circumscribed rectangular area included in another circumscribed rectangular area. An image connected component extracting apparatus having: 7. A circumscribing rectangle extracting means for dividing a circumscribing rectangular region by dividing a target image into connected components formed by concatenated pixels and an circumscribing rectangular region extracted by the circumscribing rectangle extracting means are overlapped. The circumscribing circumscribed rectangular area to be processed is selected in a predetermined order from the circumscribing circumscribing rectangle extracting means for extracting a matching object and the circumscribing rectangular area extracted by the circumscribing circumscribing rectangle extracting means. The region is divided, and an overlapping portion processing unit that determines which connected component corresponding to the circumscribed rectangular region in which the divided rectangular region overlaps is included, and based on the determination result, the selected In the image connected component extracting apparatus for identifying the connected component corresponding to the circumscribed rectangular area, the overlapping portion processing means is a predetermined pixel in the cut rectangular area. A contour tracing means for performing the starting contour tracing, and a judging means for obtaining a judgment result as to whether or not the rectangular area belongs to a connected component corresponding to the selected circumscribing rectangular area based on the result of the contour tracing. An image connected component extraction device having. 8. The image connected component extraction apparatus according to claim 7, wherein the determination means determines whether the rectangle specified by the maximum value and the minimum value of the coordinates of contour tracing matches the circumscribed rectangle of the connected component. When there is a match,
An image connected component extraction device configured to determine that a rectangular region including a starting pixel for contour tracing is included in the connected component. 9. The image connected component extracting apparatus according to claim 7, wherein the contour extracting means is an image located at a boundary of the rectangular region of interest as a starting pixel and is in another rectangular region. An image connected component extraction device adapted to select images that do not belong. 10. The image connected component extraction apparatus according to claim 7, wherein the determination unit determines a starting pixel for contour tracing when a circumscribed rectangular area selected during contour tracing is exceeded. An image connected component extraction apparatus having means for determining that the included rectangular area is not included in the connected components corresponding to the included circumscribed rectangular area. 11. A circumscribing rectangle extracting means for dividing a circumscribed rectangular area by dividing a target image into connected components composed of connected pixels and an circumscribing rectangular area extracted by the circumscribing rectangle extracting means overlap each other. The circumscribing circumscribed rectangular area to be processed is selected in a predetermined order from the circumscribing circumscribing rectangle extracting means for extracting a matching object and the circumscribing rectangular area extracted by the circumscribing circumscribing rectangle extracting means. The region is divided, and an overlapping portion processing unit that determines which connected component corresponding to the circumscribed rectangular region in which the divided rectangular region overlaps is included, and based on the determination result, the selected In the image connected component extraction apparatus configured to identify the connected component corresponding to the circumscribed rectangular area, the overlapping portion processing unit determines that the size of the selected circumscribed rectangular area is predetermined. Size determining means for determining whether or not the size is larger, and when the size determining means determines that the size of the selected circumscribed rectangular area is larger than a predetermined value, a predetermined pixel in the cut rectangular area Contour tracing means for starting contour tracing, and determination means for obtaining a determination result as to whether or not the rectangular area belongs to a connected component corresponding to the selected circumscribing rectangular area, based on the result of the contour tracking, Label image creation for creating a label image corresponding to a connected component included in the selected circumscribed rectangular area when the size determination unit determines that the size of the selected circumscribed rectangular area is not larger than a predetermined value An image connected component extracting apparatus having means and means for determining to which connected component the rectangular region cut out belongs based on the created label image. 12. The image connected component extraction apparatus according to claim 11, wherein the overlapping portion processing means further includes another processed circumscribed rectangular area that overlaps the selected circumscribed rectangular area. An image connected component extraction apparatus having result diversion means for obtaining a determination result for a selected circumscribed rectangular area by using a determination result obtained by processing for a circumscribed rectangular area. 13. The image connected component extracting apparatus according to claim 11, wherein the overlapping portion processing unit further selects a circumscribing rectangular area to be processed from the overlapping circumscribing rectangular areas in ascending order of area. An image connected region extracting apparatus having: