JPH05108823A - Contour extracting method and device therefor - Google Patents

Abstract

PURPOSE:To provide the extracting method of two contour points set between two picture elements combined in an oblique direction, based on a rule at the time of extracting the contour points of an 8-combined picture element area. CONSTITUTION:This method is provided with the process for setting merely one point each for constituting a contour line between a noticed picture element and a neighboring picture element at each noticed picture element, based on the state of the neighboring element, and deciding the connecting direction of the points constituting the contour line, and the process for judging the connecting state of the point constituting the contour line with an another point constituting the contour line. The position of the noticed picture element is set on picture data in the order of a luster scanning, and the process is executed at each noticed picture element based on the state of the neighboring picture element, so that the contour point can be extracted. The picture element in the neighborhood of the position of a picture element (e) is processed (step S0001). The picture element in the neighborhood of the position of a picture element C is processed (step S0002). The picture element in the neighborhood of the position of a picture element A is processed (step S0003). The picture element in the neighborhood of the position of a picture element B is processed (step S0004). Then, the processing is returned to a case 00.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for extracting contour lines of a binary image stored in raster scanning order.

[0002]

2. Description of the Related Art Regarding this type of apparatus, the applicant has already applied for a patent application Japanese Patent Application No. 2-281958 (hereinafter referred to as a prior patent). In this prior application patent, "a step of determining a connecting direction of a point of a contour line based on a state of a neighboring pixel by setting a predetermined position as a point forming a contour line depending on a state of a target pixel and its neighboring pixel, A step of determining a connection state between a point forming the contour line and another point of the contour line; updating the position of the target pixel on the image data in a raster scanning order, and based on the state of neighboring pixels for each target pixel. The above steps are executed to extract contour points.

In the above arrangement, the state of the pixel of interest and its neighboring pixels in the image data is held, this pixel of interest is taken out in raster scanning order, and the horizontal and vertical directions are obtained based on the states of the pixel of interest and its neighboring pixels. The pixel array vector in the direction is detected. The connection state of these pixel array vectors is discriminated, and the contour of image data is extracted based on the discriminated connection state of the pixel array vectors. ] Is proposed. The method is
Since all the contour lines in the image can be extracted only in the order of one raster scan and the image memory for storing the entire image data is not required, the memory capacity can be reduced. .. In addition, the contour is extracted for each edge of the pixel instead of the center position of the pixel of the input image.
This is a contour line extraction method that has a significant width even for a thin line of pixel width. Further, it is introduced that not only the contour line of the connected pixel area connected in four directions of pixels in the original image can be extracted, but also the pixel area connected in eight directions can be extracted.

[0004]

However, in the embodiment introduced in the above-mentioned prior application, a description will be given in the case of extracting the contour lines of the connected pixel regions connected in eight directions (hereinafter referred to as "8 connected"). In the disclosed method, extraction of a contour line between pixels diagonally connected to a target pixel is complicated.

That is, in the above-mentioned prior application, when the target pixel is a black pixel, it is determined whether the eight surrounding pixels of the target pixel are white pixels or black pixels, and the processing content is determined for each pattern in each state. is doing. When extracting the contour line of the connected pixel area in the vertical direction and the horizontal direction of the pixel of interest, it is sufficient to define the contour point only when the surrounding pixels are white pixels.
To extract the outline of the diagonally connected pixel area,
It is necessary to define contour points even when both the pixel of interest and the pixels in the diagonal direction are black pixels. From this, when extracting the contour line of the diagonally connected pixel region, the position between the pixels existing between the two pixels connected in the diagonal direction becomes the target pixel position for each of the two pixels. It will be scrutinized twice in both cases. Figure 2
Shows an example. In the figure, A and B represent two pixels connected in an oblique direction, and C is a position between pixels where a contour point exists. For this reason, it is necessary to eliminate redundancy in the examination at each point of time and ensure consistency. The
In the example of the prior application, the procedure for extracting the contour points of the diagonally connected pixel area is not set so that it can be considered independently of the states of other surrounding pixels, and the extraction rule for each part is difficult to modularize. It was set. Taking the case of FIG. 2 as an example, as shown in FIG. 3, when the black pixel A is the target pixel, the start points of the two vertical vectors C ₁ and C ₂ are set as contour points, and C ₁ flows out. The destination is defined, the inflow source is defined as an undetermined contour vector, and C ₂ is defined as a contour vector in which neither the outflow destination nor the inflow source is determined. On the other hand, when the black pixel B is the target pixel, as shown in FIGS. 4 and 5, depending on whether the pixel position indicated by D is a white pixel (FIG. 4) or a black pixel (FIG. 5), the content of the processing is different. In these cases, no new contour vector is defined, and only the connection state of the contour vector defined in FIG. 3 is updated. When the pixel position indicated by D is a white pixel, the inflow source of C ₁ and the outflow destination of C ₂ are determined.
When the pixel position indicated by D is a black pixel, only the inflow source of C ₁ is determined. And so on. For this reason, there has been dissatisfaction in that the programming complexity is generated and the processing speed does not increase. In addition, it has been a factor that causes an increase in the circuit scale at the time of hardware implementation.

[0006]

The present invention has been made in view of the circumstances in the above-mentioned prior application example, and the rule for extracting the contour points of the 8-connected pixel areas in the above-mentioned prior application example is oblique. When extracting a pixel area connected in the direction, two contour points set between two black pixels connected in the diagonal direction are defined one by one at each time point when the two black pixels are the target pixel. By adopting the method described above, the extraction rule of each part can be easily modularized so that it can operate independently of the states of other surrounding pixels.

[0007]

[Example] Case 0 in another example of the previous application
0, case 01, case 02, case 03, case 0
4, case 06, case 08, case 09, case 01
2 is processed by the following method. Hereinafter, each process will be described with reference to the flowcharts shown in FIGS.

Each of the above cases corresponds to the case in which there is a possibility of existence of diagonally connected pixel areas when the contour points of the 8-connected pixel areas of the prior application are extracted. , By considering all of the above cases, all cases of diagonal connection are covered.

The present invention basically provides a method of extracting contour points relating to 8-connected regions by the method described in the above-mentioned prior patent.
Since the main purpose is to extract more efficient contour points by making improvements, the embodiments are also described based on the description in the above-mentioned prior patent, and only the improved parts are described in the format described here. go. Further, except for the contour point extraction portion for the 8-connected region, the method described in the prior patent is directly followed. Therefore, except for the newly added drawings, the drawings used in the prior patent will be referred to as they are.

FIG. 6 is a flowchart for processing the case (case XX) shown in FIG. 60 of the prior application. In step S0001, the pixel matrix 601 of FIG.

[0011]

[Outer 1] The processing in the vicinity of the pixel at the position is performed, and the content thereof is shown in the flowchart of FIG. this

[0012]

[Outside 2] The position of corresponds to the position of the pixel D in FIG. 1 of the prior application.

In the flowchart of FIG. 7, in step S2001, (2i-1, 2j-1) is registered in the horizontal vector registration table shown in FIG. 23 as the starting point of the horizontal vector. In step S2002, it is assumed that the vector of the outflow destination of this horizontal vector is undecided, it is registered in the table 321 shown in FIG. 32 of the prior application, and the counter 320 is updated. Step S200
In 3, it is determined whether the position of the pixel D is a black pixel (f (D) = 1) or not (f (D) = 0), and a black pixel, that is, f (D)
If = 1, the process proceeds to step S2006. If not, that is, when the white pixel f (D) = 0, step S2.
Proceed to 004. In step S2004, the inflow source vector of the horizontal vector is also determined to be undetermined, and is registered in the table 311 shown in FIG. 31 of the prior application, and the counter 310 is updated. In step S2005, the horizontal vector counter 230 is incremented by 1, and the process returns to the original process. In step S2006, a vertical vector flowing into the horizontal vector defined in step S2001 is searched. The search procedure conforms to the method described in FIG. 30 of the prior application. However, in step S52 in the figure, is the x coordinate of the starting point of the vertical vector indicated by the item number (k-1) (2i + 1)? ", There is a place"-(2i-1)
~]. When the process of step S2006 ends, the process proceeds to step S2005. Thus, the processing of case e in step S0001 of FIG. 6 is completed. Here, as shown in FIGS. 8 and 9, one horizontal vector between the pixel of interest and the pixel position D is extracted. (Here, 0 represents the starting point of the horizontal vector, the arrow from the circle indicates the direction in which the horizontal vector is facing (that is, left or right), and the arrow entering the circle indicates
It represents the direction (that is, up or down) of the vertical vector injected at the starting point of the horizontal vector. The broken line arrow means that the connection relation is undetermined, and the solid line arrow means that the connection relation is a vector to be determined at that time.

Next, the process proceeds to step S0002, and the processing in the vicinity of the pixel at the position C shown in FIG. The contents are shown in the flowchart of FIG. In step S2101, (2i-1, 2j + 1) is registered as a vertical vector start point in the vertical vector registration table shown in FIG. 24 of the prior application. In step S2102, assuming that the vector of the inflow source of this vertical vector is undecided, it is registered in the table 331 shown in FIG. 33 of the prior application, and the counter 330 is updated. Step S210
In 3, it is determined whether the position of the pixel C is a black pixel (f (C) = 1) or not (f (C) = 0), and the black pixel, that is, f
If (C) = 1, the process proceeds to step S2104. If not, that is, when the white pixel f (C) = 0, the process proceeds to step S2106. In step S2104, it is assumed that the vector of the outflow destination of the vertical vector is undecided, and it is registered in the table 341 shown in FIG. In step S2105, the vertical vector counter 240 is incremented by 1, and the process returns to the original process.
In step S2106, the horizontal vector from which the vertical vector defined in step S210 flows is searched. The search procedure conforms to the method described in FIG. 35 of the prior application. When the process of step S2106 is completed, the process proceeds to step S2005. Thus, the processing of case C in step S0002 in FIG. 6 is completed. here,
As shown in FIGS. 11 and 12, one of the vertical vectors between the pixel of interest and the pixel position C is extracted. (Here, the triangle indicates the starting point of the vertical vector, the arrow coming out of the triangle indicates the direction in which the vertical vector is facing (that is, up or down), and the arrow entering the triangle indicates the starting point of the vertical vector. The direction of the incoming horizontal vector (that is, left or right) is shown.The broken line arrow means that the connection relation is undecided, and the solid line arrow is
It means that the connection relation is a vector that is fixed at that time.

Next, the process proceeds to step S0003, and the processing in the vicinity of the pixel at the position A shown in FIG. The contents are shown in the flowchart of FIG. In step S2201, (2i + 1,2j-1) is registered as a vertical vector start point in the vertical vector registration table shown in FIG. 24 of the prior application. In step S2202, the horizontal vector flowing into the vertical vector defined in step S2201 is searched. The search procedure conforms to the method described in FIG. 45 of the prior application.
In step S2203, it is determined whether the position of the pixel A is a black pixel (f (A) = 1) or not (f (A) = 0), and if the pixel is a black pixel, that is, f (A) = 1, step S2206. If not, that is, if the white pixel f (A) = 0, the process proceeds to step S2204. In step S2204, the outflow destination vector of the vertical vector is determined to be undetermined, and is registered in the table 341 shown in FIG. 34 of the prior application, and the counter 340 is updated. Step S220
At 5, the vertical vector counter 240 is incremented by 1 and the process returns to the original process. In step S2206, step S2
A horizontal vector from which the vertical vector defined in 201 flows out is searched. The search procedure conforms to the method described in FIG. 35 of the prior application. However, in step S62 of the figure, is the x coordinate of the starting point of the horizontal vector of the table 311 indicated by "item number (k-1) (2i-1)?" ], A certain place operates as "~ (2i + 1) ~". When the process of step S2206 is completed, step S2
Go to 205. Thus, the process of case a in step S0003 in FIG. 6 is completed. Here, as shown in FIGS. 14 and 15, one of the vertical vectors between the pixel of interest and the pixel position A is extracted.

Next, the process proceeds to step S0004, and the processing in the vicinity of the pixel at the position B shown in FIG. The contents are shown in the flowchart of FIG. In step S2301, (2i + 1,2j + 1) is registered in the horizontal vector registration table shown in FIG. 23 as the starting point of the horizontal vector. In step S2302,
A vertical vector in which the horizontal vector defined in step S2301 flows out is searched for. The search procedure conforms to the method described in FIG. 46 of the prior application. In step S2303, the position of the pixel B is a black pixel (f
Whether (B) = 1) or not (f (B) = 0) is determined, and in the case of a black pixel, that is, f (B) = 1, the process proceeds to step S2304; otherwise, that is, a white pixel f (B). If = 0, the process advances to step S2306. In step S2304,
Assuming that the vector of the inflow source of the horizontal vector is undecided, it is registered in the table 311 shown in FIG. 31 of the prior application, and the counter 310 is updated. In step S2305, the horizontal vector counter 230 is incremented by 1, and the process returns to the original process. In step S2306, the vertical vector at the inflow source of the horizontal vector defined in step S2301 is searched. The search procedure conforms to the method described in FIG. 30 of the prior application. When the process of step S2306 ends, the process advances to step S2305. Thus, step S000 in FIG.
The process of case b of 4 is finished. This is the 6th of the prior application
The processing in the case shown in FIG. 0 (case XX) is terminated.

FIG. 19 is a flowchart for processing the case (case 01) shown in FIG. 61 of the prior application.

First, in step S0101, FIG.

[0019]

[Outside 3] The processing near the position of is performed. The content of the processing is exactly the same as that of the case c shown in FIG. Next, in Step S0102, the prior application Japanese Patent Application No. 61 in FIG.

[0020]

[Outside 4] The processing near the position of is performed. The content of the processing is exactly the same as that of the case b shown in FIG. Thus, the processing in the case shown in FIG. 61 of the prior application (case 01) is completed.

FIG. 20 is a flowchart for processing the case (case 02) shown in FIG. 62 of the prior application.

First, in step S0201, FIG.

[0023]

[Outside 5] The processing near the position of is performed. The contents of the processing are exactly the same as the case e shown in FIG. Next, in step S0202, the prior application Japanese Patent Application No. 62 shown in FIG.

[0024]

[Outside 6] The processing near the position of is performed. The content of the processing is exactly the same as that of the case c shown in FIG. Thus, the processing in the case shown in FIG. 61 of the prior application (case 02) is completed.

FIG. 21 is a flow chart for processing the case (case 03) shown in FIG. 63 of the prior application.

First, in step S0301, FIG. 63 of the prior application is filed.

[0027]

[Outside 7] The processing near the position of is performed. The content of the processing is exactly the same as that of the case c shown in FIG. Next, in Step S0302, the prior application Japanese Patent Application No. 63 shown in FIG.

[0028]

[Outside 8] The processing near the position of is performed. The contents of the processing are exactly the same as the case A shown in FIG. 38 of the prior application. Thus, the case shown in FIG. 63 of the prior application (Case 03)
Ends the process.

FIG. 22 is a flow chart for processing the case (case 04) shown in FIG. 64 of the prior application.

First, in step S0401, FIG. 64 of the prior application is filed.

[0031]

[Outside 9] The processing near the position of is performed. The contents of the processing are exactly the same as the case e shown in FIG. Then step S
Proceeding to 0402,

[0032]

[Outside 10] The processing near the position of is performed. The contents of the processing are exactly the same as in the case a shown in FIG. Thus, the processing in the case shown in FIG. 64 of the prior application (case 04) is completed.

FIG. 23 is a flowchart for processing the case (case 06) shown in FIG. 65 of the prior application.

First, in step S0601, the process of FIG.

[0035]

[Outside 11] The processing near the position of is performed. The processing content is exactly the same as that of the source e shown in FIG. Then step S0
Proceed to 602 and proceed to FIG.

[0036]

[Outside 12] The processing near the position of is performed. The contents of the processing are exactly the same as the case B shown in FIG. 42 of the prior application. Thus, in the case shown in FIG. 65 of the prior application (case 06)
Ends the process.

FIG. 24 is a flow chart for processing the case (case 08) shown in FIG. 66 of the prior application.

First, in step S0801, FIG. 66 of the prior application is filed.

[0039]

[Outside 13] The processing near the position of is performed. The contents of the processing are exactly the same as in the case a shown in FIG. Next, in Step S0802, the prior application Japanese Patent Application No. 66 in FIG.

[0040]

[Outside 14] The processing near the position of is performed. The content of the processing is exactly the same as that of the case b shown in FIG. Thus, the processing in the case shown in FIG. 66 of the prior application (case 08) is completed.

FIG. 25 is a flowchart for processing the case (case 09) shown in FIG. 67 of the prior application.

First, in step S0901, FIG. 67 of the prior application is filed.

[0043]

[Outside 15] The processing near the position of is performed. The contents of the processing are exactly the same as case D shown in FIG. 48 of the prior application. Next, in step S0902, in FIG.

[0044]

[Outside 16] The processing near the position of is performed. The content of the processing is exactly the same as that of the case b shown in FIG. Thus, the processing in the case shown in FIG. 67 of the prior application (case 09) is completed.

FIG. 26 is a flow chart for processing the case (case 012) shown in FIG. 68 of the prior application.

Still in step S1201, FIG.

[0047]

[Outside 17] The processing near the position of is performed. The contents of the processing are exactly the same as the case C shown in FIG. 52 of the prior application. Next, in step S1202, the prior application Japanese Patent Application No. 68 shown in FIG.

[0048]

[Outside 18] The processing near the position of is performed. The contents of the processing are exactly the same as in the case a shown in FIG. Thus, the processing in the case shown in FIG. 68 of the prior application (case 012) is completed.

Above, description has been added regarding how to improve the above-mentioned prior application No. 00, No. 01, No. 02, No. 03, No. 04, No. 06, No. 08, No. 09, and No. 012. However, the other parts of the embodiment can be implemented by following the embodiment of the prior application. FIG. 1 shows the scanning method of the image and the relationship between the pixel of interest and its neighboring pixels. The description of the whole operation based on this drawing is in the embodiment of the prior application.

(Other Embodiments) Case 00, Case 03,
Case 06, case 09, and case 012 may be configured as case 00 ′, case 03 ′, case 06 ′, case 09 ′, and case 012 ′ shown in FIGS. 27 to 31, respectively.

That is, in each of the above cases, the case e is processed before the cases a and c. Case b is processed after case a and case c. It can be easily understood that the two conditions mentioned above must be satisfied.

FIG. 32 is the same as FIG. 5 of the prior application, and is a block diagram showing an example of a hardware configuration for carrying out the contour detection of this embodiment described above. The programs represented by all the above-mentioned flowcharts are stored in the memory 520 of FIG. In FIG. 32, reference numeral 501 is an input control (interface) unit for exchanging image data input via the signal line 500, and binary image data is sequentially input from the signal line 500 in a raster scanning format. 5
Reference numeral 02 denotes a latch, which stores the image data input from 501
Each pixel is sequentially updated and held in synchronization with a pixel synchronization clock (not shown). At the next pixel synchronization clock, the latch 502 inputs the next pixel data from the input control circuit 501. At this time, the image data already held is latched and held in the latch 503 in synchronization with the image clock. Similarly, the pixel data held in the latch 503 is latched by the latch 50 at the next pixel synchronization clock.
Held at 4.

Reference numerals 505 and 503 denote FIFOs (first-in first-out memory) for holding pixel data for one raster. The FIFO 505 sequentially takes in the output of the latch 504 in synchronization with the pixel synchronization clock and outputs the data one raster before to the latch 507. Similarly, the FIFO 506 also takes in the output of the latch 509 and outputs the pixel data of one raster before to the latch 510. Latches 507, 508, 509 and latch 51
0, 511, 512 are both latches 502, 503,
Works exactly the same as 504.

In this way, the latches 502, 503,
The nine pixels stored in 504, 507, 508, 509, 510, 511 and 512 are 9 (3) shown in FIG.
× 3) Pixel data of a region including pixels is stored. That is, the data of these latches are respectively "B", "2", "C", "1", and

[0055]

[Outside 19] It corresponds to "3", "A", "0", and "D".

Reference numerals 513 and 514 are the input ports of the CPU 519, and the input port 513 is the latches 510 and 50.
The data of 2, 504 and 512, that is, the position data of “A”, “B”, “C” and “D” in FIG. 1 are input to the CPU 519. Similarly, the input port 514 has the data of the latches 511, 507, 503, 509, and 508, that is, "1", "2", "3",

[0057]

[Outside 20] Input as position data of.

Reference numeral 515 is a main scanning counter which indicates a pixel position in the main scanning direction, which is reset by a sub scanning synchronizing signal (not shown) and counts up by the pixel synchronizing signal.
A sub-scanning counter 516 indicates a pixel position in the sub-scanning direction, which is reset by a page sync signal (not shown) and counted up by the sub-scan sync signal. Reference numeral 517 is an input / output port for input / output control, and the input / output control circuit 50
1 holds a signal for instructing execution and suspension of pixel data input, a signal for notifying the CPU 519 of pixel data update from the input / output control circuit 501, and the like. Reference numeral 521 is an input / output control device for the hard disk 522. Input / output port 517, main scanning counter 515, sub-scanning counter 5
16, the input ports 513 and 514, the memory 520, and the disk I / 0521 are connected to the CPU 519 via the bus 518.

Thus, the CPU 519 updates the pixel data via the input / output control port 517, and can know the pixel position (i, j) of the target pixel via the main scanning counter 515 and the sub-scanning counter 516. it can. Also,
Through the input ports 513 and 514, the states of the pixel of interest and the pixels in the eight directions in the vicinity thereof can be known.

When the processing of the pixel of interest is completed in this way, the CPU 519 sends a 9
The pixel data stored in each latch is instructed to be updated, and at the same time, the pixel data update completion signal is reset. Upon receiving this update instruction, the input / output control circuit 501 clears the pixel data update instruction signal, updates the image data latched in the latch at the subsequent stage, and when this update ends, the input / output control port 517. Then, the update completion signal is output.

The CPU 519 monitors the input of the update completion signal from the input / output control port 517 after the output of the update instruction. When this update completion signal is input,
The processing relating to the pixel data newly stored in the nine latches is executed, and the same is repeated thereafter. In addition, the input control circuit 501, when finishing processing the last pixel of the image area as the pixel of interest, inputs / outputs the control port 51.
The end signal is output to 7.

[0062]

As described above, according to the rule for extracting the contour points of the 8-connected pixel areas in the above-mentioned prior application, two rules are set between two black pixels connected in an oblique direction. It is possible to configure the contour point so that it can operate independently of the states of other surrounding pixels by adopting the method of defining one point at each time point when the two black pixels become the target pixel. It is possible to easily configure the extraction means of each unit as a module. As a result, there is an effect that the circuit configuration scale can be reduced when the hardware is realized. In addition, since the program can be configured simply even during software imprint, the processing speed can be improved.

[Brief description of drawings]

FIG. 1 is a diagram showing a relationship between a scanning direction of an image, a pixel of interest, and its neighboring pixels.

FIG. 2 is a diagram showing an example of input for explaining a process in a prior application.

FIG. 3 is a diagram for explaining processing for the input of FIG. 2 in the prior application.

FIG. 4 is a diagram for explaining processing for the input of FIG. 2 in the prior application.

FIG. 5 is a diagram illustrating a process for the input of FIG. 2 in the prior application.

FIG. 6 is a flowchart showing the improved processing contents of Case 00 of the prior application of the present invention.

FIG. 7 is a flowchart showing processing of case e.

FIG. 8 is a diagram showing the processing contents of case e.

FIG. 9 is a diagram showing the processing contents of case e.

FIG. 10 is a flowchart showing the processing of case c.

FIG. 11 is a diagram showing the processing contents of case c.

FIG. 12 is a diagram showing the processing contents of case c.

FIG. 13 is a flowchart showing the processing of case a.

FIG. 14 is a diagram showing the processing contents of case a.

FIG. 15 is a diagram showing the processing contents of case a.

FIG. 16 is a flowchart showing the processing contents of case b.

FIG. 17 is a diagram showing the processing contents of case b.

FIG. 18 is a diagram showing the processing contents of case b.

FIG. 19 is a flowchart showing the improved processing contents of Case 01 of the prior application of the present invention.

FIG. 20 is a flowchart showing the improved processing contents of Case 02 of the prior application of the present invention.

FIG. 21 is a flow chart showing the improved processing contents of Case 03 of the prior application of the present invention.

FIG. 22 is a flow chart showing the improved processing contents of case 04 of the prior application of the present invention.

FIG. 23 is a flowchart showing the improved processing contents of Case 06 of the prior application of the present invention.

FIG. 24 is a flowchart showing the improved processing contents of Case 08 of the prior application of the present invention.

FIG. 25 is a flowchart showing the improved processing contents of Case 09 of the prior application of the present invention.

FIG. 26 is a flowchart showing the improved processing contents of Case 012 of the prior application of the present invention.

FIG. 27 shows case 00, case 03, and case 0, respectively.
6 is a flowchart showing another embodiment for Case 6, Case 09, and Case 012.

FIG. 28 shows case 00, case 03, and case 0, respectively.
6 is a flowchart showing another embodiment for Case 6, Case 09, and Case 012.

FIG. 29 shows Case 00, Case 03, and Case 0, respectively.
6 is a flowchart showing another embodiment for Case 6, Case 09, and Case 012.

FIG. 30 shows Case 00, Case 03, and Case 0, respectively.
6 is a flowchart showing another embodiment for Case 6, Case 09, and Case 012.

FIG. 31 shows Case 00, Case 03, and Case 0, respectively.
6 is a flowchart showing another embodiment for Case 6, Case 09, and Case 012.

FIG. 32 is a diagram showing a hardware configuration.

[Explanation of symbols]

 519 CPU 520 memory

Claims (5)

[Claims] 1. At most one point forming a contour line between the target pixel and each of the neighboring pixels is set for each target pixel based on the state of the neighboring pixel, and connection of the points forming the contour line is established. The method includes a step of determining a direction and a step of determining a connection state between a point forming the contour line and another point forming the contour line, and setting the position of the target pixel on the image data in raster scanning order. Then, the above-described process is executed for each pixel of interest based on the state of the neighboring pixels, and the contour points are extracted. 2. The contour extracting method according to claim 1, wherein a position of a point forming the contour line is an intermediate position between input pixels. 3. The contour extracting method according to claim 1, wherein the positions of the extracted contour points are positions between input pixels. 4. A holding unit that holds the state of a pixel of interest in image data and its neighboring pixels, a procedure for updating the pixel of interest and its neighboring pixels in raster scan order, and a state of the pixel of interest and its neighboring pixels. A detection means for detecting at most one horizontal or vertical contour vector between the pixel of interest and each of the neighboring pixels, and a discrimination means for discriminating the connection state between the contour vectors; A contour extracting device comprising an extracting means for extracting a contour of the image data based on the connection state determined by the means. 5. The contour extracting apparatus according to claim 4, wherein the neighboring pixels are eight neighboring pixels surrounding the target pixel.