JP2878194B2 - Partial erasure and partial detection method of image data - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for partially erasing and partially detecting image data, and more particularly to a method for partially erasing only specified lines on binary image data input by a scanner and detecting only specified lines. To convert the data into vector data.

[0002]

2. Description of the Related Art A scanner reads a drawing such as a design drawing or a map or a character typeface written on paper as image data composed of a two-dimensional array of white pixels and black pixels, and compresses and stores the data. When vectorizing and saving as graphic data or outline font data, image data read by a scanner is displayed on the screen as preprocessing,
It is necessary to modify the image data interactively on the display screen.

In this case, it is often necessary to erase only one of the two intersecting line segments and leave an overlapping portion, but a black pixel is converted to a white pixel by designating an area or a pixel on the screen. However, the conventional pixel conversion method has a disadvantage that the correction process requires much labor and time. Note that a line segment means a straight line of a finite length with a thickness or a gentle curve. As a countermeasure against this problem, Japanese Patent Laid-Open Publication No. Hei 5-46690 discloses that a rectangular area including a line segment to be erased is designated, image data in the designated area is vectorized to form graphic data, There has been proposed a method of erasing all image data and then restoring graphic data excluding a line segment to be erased to image data.

If the image data read by the scanner is composed of a large number of intersecting line segments, and only a small number of the line segments need to be retained, all of them are processed or retained by a normal pixel conversion method. There is no suitable method other than processing a portion other than the line segment by a normal pixel conversion method and then erasing an unnecessary line segment crossing the remaining line segment by the method described in JP-A-5-46690.

[0005]

SUMMARY OF THE INVENTION The above-mentioned Japanese Patent Laid-Open No. 5-4
In the method described in Japanese Patent Application Publication No. 6690, image data in a specified rectangular area including a line segment to be erased is subjected to vectorization processing, and all of the line data are once converted into graphic data. The disadvantage is that the process takes too long, such as when there are many line segments that intersect with the line segment to be erased, or when the line segment to be erased is oblique and there are many other line segments in the specified area. is there.

Conventional methods for detecting a vector from binary image data include a method of thinning the run-length data from the outside until the line width becomes 1, and a method of extracting a contour line connecting end points of the run-length data. After performing the processing, there is a centering process using a center line between two substantially parallel contour lines as a center line. In both cases, a vector detection process is performed collectively for a fixed area of binary image data. I do.
In the case of thinning processing, it is necessary to search the same area many times. In the case of skeletonization, three-stage processing of outline detection, outline vector detection, and skeleton vector detection of a binary image is performed.
Vector detection is generally a time-consuming process.

Therefore, when only a small number of line segments are to be left from image data read by a scanner, all image data is converted into graphic data, and then all image data is erased before restoring only the graphic data of the remaining line data. This method is not considered to be an appropriate method because of the time-consuming disadvantage.

A first object of the present invention is to provide a method of partially erasing image data in which only specified line segments on binary image data are partially erased at high speed.

A second object of the present invention is to provide a method of detecting a specified line segment on binary image data at a high speed and performing a vectorization process on the image data.

[0010]

According to a first aspect of the present invention, there is provided a method for partially erasing image data, comprising the steps of obtaining a line segment from a positional relationship between a start point and an end point designated on a line segment to be erased on a screen displaying image data. A direction determination process for determining a tracking direction and a search direction of a black pixel row, and searching for a black pixel row in a predetermined area centered on the start point and parallel to the line segment tracking direction to set the start point setting environment condition Starting point processing for erasing the line segment to be erased in the predetermined area in a method corresponding to and detecting a black pixel column of the line to be erased on a boundary on the line segment tracking direction side of the predetermined area. The process of searching for a black pixel column having substantially the same run length connected in the direction orthogonal to the search direction of the black pixel column based on the already detected black pixel column and erasing the already detected black pixel column is repeated. Erase at the same time as detection while tracking the target line segment Sequential and tracking erasing process, the successive black pixel row connected while accumulating information of black pixels columns cancel erasing black pixel columns when detecting a rapid increase run length of black pixels rows in the course of the track erasing process Continue searching for black pixels
The run length of the row suddenly decreases in the opposite state
Run length almost equal to before the sharp increase in the direction of the line segment to be erased
Or a run line that is approximately equal to
NGs black pixel column is detected and not increased again.
If it is confirmed that the continuation is continued, the intersection processing for collectively processing the erasure of the erasure target line segment and the restoration of the intersection line in the intersection or the branch point section based on the accumulated information of the black pixel column, and the sequential tracking erasure processing or the intersection When the end point is detected during the processing, an end point process for erasing and restoring the line to be erased from a fixed area including the end point is executed by a method corresponding to the set environmental condition of the end point. By specifying a start point and an end point of a line segment to be erased on a display image, only the specified line segment is erased at high speed without erasing a crossing or branching line segment.

According to a second aspect of the present invention, there is provided a method of detecting a portion of image data.
A black pixel column is converted into a white pixel column for erasing a line segment to be erased in each of the start point processing, the sequential tracking erasure processing, the intersection processing, and the end point processing in the partial erasure method of image data according to claim 1. In this case, position information on both ends of each black pixel column is stored and stored, and at the end of the processing of the end point processing, the information stored and stored in each processing is combined to create contour data.

According to a third aspect of the present invention, there is provided a method of detecting a portion of image data.
3. The method according to claim 2, wherein the contour data is subjected to a vectorization process to generate a contour vector.

[0013]

Next, embodiments of the present invention will be described with reference to the drawings.

In the method of partially erasing image data according to the present invention, the start direction and the end point of a line segment to be erased are designated on a display image, so that the line segment tracing direction (line segment direction) and line segment tracing Tracing that determines the direction of the search process for tracing, traces the line segment based on the connection information of the continuous black pixel row (referred to as black run) in one direction in the orthogonal direction, and simultaneously performs detection and erasure. The erasing process is a basic operation in which an intersection process for preventing a line segment intersecting a target line segment from being erased and a terminal process for a start point and an end point are added. In principle, one continuous process is performed. Thus, only the specified line segment can be erased at high speed.

FIG. 1 is a flowchart showing a basic processing procedure of an embodiment of a method for partially erasing image data according to the present invention. Hereinafter, this will be described in detail with reference to FIG.

First, a start point and an end point of a line segment to be deleted are designated on the screen by a mouse or the like (step S1). From the positional relationship between the start point and the end point, determine the direction in which the line segment is to be traced from the start point (initial line segment tracking direction), and search for a black run in the vertical direction to track the line segment, or The search direction of the black run is determined whether the black run is searched and the line segment is tracked (step S2).

When the start point and the processing direction are determined, start point processing is performed (step S3). The start point processing determines whether the start point is set at the end of the target line segment, is set at the intersection with another line segment, or is set at the middle of the line segment, and This is a process for performing a desirable erasure corresponding to the above. The process is performed by searching an area having a constant width in the black run search processing direction extending on both sides along the initial line segment tracking direction with the start point as the center. Details of the start point processing will be described later.

When the start point process is completed, the process proceeds to a sequential tracking and erasing process in which a black run is searched, a line segment is traced based on the connection information in the orthogonal direction, and detection and erasure are simultaneously performed (step S).
4). In this sequential tracking erasure process, as shown in FIG. 2A, a black run on line A already detected as a processing target line segment is set as a “detected run” and the line segment tracking direction is taken into consideration. A range of a predetermined width (for example, 3 to 4 times the run length) on the next line B is searched to detect a connected black run connected to the “detected run” in the connection search direction. When the connected black run is detected, as shown in FIG. 2B, the “detected run” on line A is replaced with a white run (white pixel column) and erased, and the detected connected black on line B is deleted. With the run as the “detected run”, a process of searching for a range of a predetermined width on the next line C and similarly detecting a connected black run is repeated.

When the run length of the connected black run is extremely longer (for example, 1.5 times or more) than the run length of the “detected run”, the line of the black run is regarded as the start line of the intersection. (Step S5), the intersection processing is started (Step S6). In the intersection processing, detection of the connected black run is continued without erasing the “detected run”, and when the run length of the connected black run changes suddenly and a run length close to the value before the start of the intersection processing is detected, the intersection is detected. (Step S8), and the erasure of the line to be erased and the restoration of the intersection between the start line and the end line of the intersection are collectively processed (step S8).

If an end point is detected during the intersection processing (step S7), the intersection processing is interrupted and the process proceeds to the end point processing (step S10). The end point processing is, like the start point processing, the end point set at the end of the target line segment, set at the intersection with another line segment, or set in the middle of the line segment. This is a process for judging whether or not the image data is deleted, and performing erasure corresponding to each of the images, and is performed by searching for a fixed area in the line segment tracking direction including the end point. The details are
The details of the start point processing will be described later.

When a black run having an end point is detected during the sequential tracking and erasing process (step S9),
The process proceeds to step S10 to perform an end process. Otherwise, the process returns to step S4 and the tracking erasure process is sequentially repeated.

With the above-described series of processing, the search range is limited only by specifying the start point and the end point of the line segment to be erased on the display image, and based on the connection information in the orthogonal direction of the black run. Detection and erasure are performed simultaneously while tracing the line segment, and only the line segment specified by one continuous process is erased at high speed without erasing the intersection line segment by intersection processing and start point and end point processing. Can be.

FIG. 3 is an explanatory diagram of a specific example of the intersection processing, and shows a case where one of two straight lines intersecting at an inclination of 45 degrees is deleted. A 45-degree slant line is a case where the angle between the tracking direction of the line to be erased and the processing direction of the black run search is the smallest, and is a representative example of the intersection processing that is the most complicated. Hereinafter, the intersection processing will be specifically described with reference to FIG.

Of the two intersecting straight lines shown in FIG.
A straight line M descending to the right from the upper left to the lower right is erased, and a straight line N descending to the left is intersected with the straight line M. The erasing of the straight line M is sequentially performed from the black run of the first column by successive pursuit erasure processing. As shown in FIG. 3B, when a connected black run in the fourth column connected to the black run in the third column is searched for, the run length is rapidly increased, so that the black run in the third column is not deleted. Enter intersection processing. Upon entering the intersection processing, the operation of searching for a connected black run is continued while accumulating information on the detected black run, and the run length of the connected black run sharply decreases in a state almost opposite to that at the time of rapid increase, and the straight line M to be deleted is deleted. When a connected black run with a run length almost equal to the original width is detected in the tracking direction of the crossing, the detection of the connected black run is stopped and the information of the accumulated black run is used to process the intersection processing section from the start line to the end line. Of the straight line M and the restoration process of the straight line N are performed. FIG.
(C) shows a state after the erasing process of the straight line M is performed, in which the straight line M excluding the end line is erased and the straight line N is divided. FIG. 3D shows a state where the crossing straight line N has been restored based on the accumulated information. When the restoration of the crossed straight line N is completed, the tracking erasure process is sequentially restarted from the last end line of the intersection processing section. If the end point is detected during the intersection processing, the process proceeds to the end point processing while inheriting the accumulated information up to that point.

FIGS. 4A and 4B are explanatory diagrams of the branch point processing as a special case of the intersection processing. FIGS. 4A and 4B show a case where the straight line to be erased is the same as the example of FIG. The case
FIG. 4C shows a case where the erasing target straight line is a vertical straight line and the horizontal straight line is branched in a T-shape. As can be easily understood from FIG. 4, in the case of the branch point processing, only the erasing processing needs to be performed in the intersection processing section, and the recovery processing does not need to be performed. In the case of FIG. 4B, the run length of the connected black run is reduced and the run length of the connected line is the same as the run length (4 dots) of the start line after the intersection processing is started by detecting a rapid increase in the run length. And it is necessary to determine the end line of the branch point processing (intersection processing) by confirming that the line continues without increasing again. This is for the purpose of distinguishing from the hollow portion generated at the center of the intersection in the intersection processing described with reference to FIG.

FIG. 5 is an explanatory diagram of the start point processing. In the start point processing, whether the start point is set at the tip of the target line segment as shown in FIG. 5A or is set at the intersection with another line segment as shown in FIG. 5B This is a process for determining whether the line segment is set in the middle of the line segment as shown in FIG. For this reason, it extends on both sides along the initial line segment tracking direction (the direction connecting the start point and the end point as indicated by the arrow in FIG. 5) with the start point as the center, and the black run search processing direction (the horizontal direction in FIG. 5). Is performed by searching the initial search area SS having a constant width. As a result of this search,
In the case of FIG. 5 (a), all the hatched lines in the initial search area SS including the tip in the opposite direction from the start point are deleted,
The black run at the lower end of the area indicated by the black dot sequence is sequentially passed to the tracking and erasing process. Based on the run length of the transferred black run, a search range for searching for a connected black run in the sequential tracking and erasing process is determined. In the case of FIG. 5B, only the vertical line segment from the lower end of the intersecting horizontal straight line to the lower end of the area is erased, and the black run at the lower end of the area is sequentially passed to the tracking erasure processing. In the case of FIG. 5C, the portion from the start point to the lower end of the area is not erased, and the black run at the lower end of the area is delivered. In addition,
In the case of FIG. 5, the initial line segment tracking direction and the direction of the line segment to be erased at the start point are completely the same, but it is not necessary that they completely match, and the directions are almost the same. Should be fine.

FIG. 6 is an explanatory diagram of the end point processing. The end point processing is, like the start point processing, the end point set at the end of the target line segment, set at the intersection with another line segment, or set in the middle of the line segment. This is a process for judging whether the data is the same or not and performing erasure corresponding to each of the data. When the end point is detected in the course of the sequential tracking erasure process, the terminal search area ES extending a predetermined length from the end point in the line segment tracking direction is searched, and as shown in FIG. Is set, the entire line segment including the end point in the terminal search area ES is deleted. As shown in FIG. 6B, when the end point is set near the intersection and an intersection line is detected in the terminal search area ES, erasure is performed so as to leave the intersection line segment. If the line segment is set at the middle of the line segment as shown in FIG. 6C, only the black run of the line including the end point is deleted, and the process is performed while leaving the line segment in the terminal search area ES. finish.

When the end point is set in the intersection and the end point is detected during the intersection processing,
A process is performed in which the accumulated information accumulated in the intersection processing up to that point is inherited, and only the line segment to be erased is erased while leaving the crossing or branching straight line. FIG. 7 is an explanatory diagram of a processing example in this case. FIG. 7A shows a case where an end point is set in an intersection in the middle of erasing one of two intersecting diagonal lines described in FIG. FIG. 7B shows the processing of FIG.
The processing when the horizontal line branches in one direction from the middle of the vertical line shown in (c) is shown. In any case, after the processing similar to the intersection processing or the branch point processing, the restoration processing of the portion indicated by the mark ○ is performed.

As described in detail above, it is determined whether or not the start point is set at the start, middle, or intersection of the target line segment by the start point processing, and the erasing process corresponding to the set position is performed. Since the black run is extracted and the sequential tracking erasure process is started, the search width in the sequential tracking erasure process can be appropriately limited according to the thickness of the target line segment, and the line to be erased from the start point can be limited. Even if the direction does not completely match the direction connecting the start point and the end point, it is possible to easily enter the tracking and erasing process easily. In addition, the intersection processing automatically detects the intersection and the branch point during the sequential tracking erasure processing, and can erase only the line segment to be erased without erasing the intersection line segment. Further, appropriate end point processing is performed according to the situation, including the case where the end point is set in an intersection or a branch point. Therefore, according to the partial erasing method of the image data of the present embodiment, only the start point and the end point are specified on the line segment to be erased on the image read by the scanner, and only the specified line segment is sequentially repeated and erased. Can be erased at high speed.

The method of partially erasing image data has been described in detail above. By using this method, only a specific line segment can be partially extracted from image data at high speed and converted into vector data as required. is there. In other words, when erasing a line segment to be erased in the partial erasing method described above, the coordinate data of both ends of the black run to be erased is stored and accumulated. Minute contour data is obtained. This outline data may be stored as it is, but if the vectorization processing is performed to obtain an outline vector, the data amount can be reduced.
Only the information of the specified line segment is separated and extracted from the image data and stored, and if necessary, the partial detection of the image data reproduced in the image data can be easily performed by one operation. Further, it is also possible to further create a core line vector from the contour vector. When the contour data or the contour vector is created by detecting only the designated line segment from the image data, the contour image is reproduced and the original image in which the segment is designated (the designated segment is deleted. When this is superimposed and displayed, the processing target becomes clear, and operations such as when vectorization is performed by specifying a plurality of line segments from the same binary image data become easy.

[0031]

As described above, according to the method of partially erasing image data according to the present invention, if the start point and the end point are specified on the line segment to be erased on the image read by the scanner, the intersection is made. By automatically searching only the peripheral area of the specified line segment and repeating successive erasure without erasing other line segments, only the specified line segment can be erased at a high speed by one operation. it can. In addition, since it is based on sequential tracking erasure, not only a straight line but also a gentle curve is tracked without expanding the search area, and the terminal processing according to the situation is performed by searching around the start point and the end point. The degree of freedom in specifying the start point and end point is large, and the operation is easy.
Therefore, the operation is easy and the processing speed is high as compared with the conventional method, so that the effect that the operability is remarkably improved can be obtained.

The method for detecting a portion of image data according to the present invention extends the above-described method, acquires contour data of a line segment to be deleted, and further performs contour vectorization. Can be detected at a high speed. This has the effect of simplifying and speeding up processing such as when selecting and using a small number of line segments from binary image data read by a scanner.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a basic processing procedure of an embodiment of an image data partial erasing method according to the present invention.

FIG. 2 is an explanatory diagram of the basic operation of the sequential tracking erasure process in FIG. 1;

FIG. 3 is an explanatory diagram showing a specific example of an intersection process in FIG. 1;

FIG. 4 is an explanatory diagram of branch point processing in the intersection processing in FIG. 1;

FIG. 5 is an explanatory diagram of a start point process in FIG. 1;

FIG. 6 is an explanatory diagram of an end point process that occurs during the sequential tracking erasure process in FIG. 1;

FIG. 7 is an explanatory diagram of end point processing that occurs during the intersection processing in FIG. 1;

[Explanation of symbols]

 ES terminal search area SS initial search area

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G06T 1/00-11/80 G06K 9/18-9/58

Claims (3)

(57) [Claims] 1. A direction determination process for determining a line segment tracking direction and a black pixel column search direction from a positional relationship between a start point and an end point specified on a line segment to be erased on a screen displaying image data. And searching for a black pixel column in a predetermined area parallel to the line segment tracking direction centered on the start point and searching for the line segment to be erased in the predetermined area by a method corresponding to the set environmental condition of the start point. Starting point processing for performing erasure and detecting a black pixel array of the line to be erased on the boundary of the predetermined area on the line segment tracking direction side, and searching for a black pixel array based on the already detected black pixel array A process of searching for a black pixel array having substantially the same run length connected in a direction orthogonal to the direction and erasing the already detected black pixel array is repeated. Black during processing Continuing the search for black pixel row connected while accumulating the information of the black pixel row to cancel the erase black pixel rows when detecting a rapid increase run length of pixel rows, black pixel column
The run length of the
The run length is almost the same as before the sudden increase in the direction of the target line segment.
Approximately equal run-lens before detecting or rapidly increasing black pixel column
Detects the black pixel column of Gus and keeps it without increasing again
When the continuity is confirmed, intersection processing for collectively processing erasure of the erasure target line segment and restoration of the intersection line in the intersection or branch point section based on the accumulated information of the black pixel column, and the sequential tracking erasure processing or the intersection processing When the end point is detected during the process, an end point process for erasing and restoring the line to be erased from a certain area including the end point is executed by a method corresponding to the set environmental condition of the end point. By specifying a start point and an end point of a line segment to be deleted on a display image, only the specified line segment is deleted at a high speed without deleting a line segment that crosses or branches. Partial erasing method. 2. A black pixel column for erasing a line segment to be erased in each of start point processing, sequential tracking erasure processing, intersection processing, and end point processing in the partial erasure method of image data according to claim 1. When converting to a white pixel column, the position information at both ends of each black pixel column is stored and stored, and at the end of the end point process, the information stored and stored in each process is combined to create contour data. Method of detecting partial image data. 3. A contour vector is created by performing vectorization processing on the contour line data.
Partial detection method of the described image data.