JPH08221571A - Method and device for recognizing image - Google Patents

Abstract

PURPOSE: To accurately and automatically recognize the contours and partition lines (such as the walls of a construction drawing or the ruled lines of forms) of a drawing or a table. CONSTITUTION: Image data having the contours or partition lines of the drawing or table are fetched from an image reading part 2 or a communication control part, horizontal and vertical dot number measurement arrangement data are prepared by measuring the number of black or white dots in the horizontal and vertical directions of those image data while using a dot number measurement arrangement data preparing part 6 and based on those respective dot number measurement arrangement data, the contours and partition lines of the image are recognized by a contour/partition line recognizing part 7. Further, based on those recognized contours and partition lines, a recognizing object area is thinly divided by a remapping part 8 and it is repeated to recognize the contours and partition lines by performing processing again for each limited area until the new contour or partition line can not be recognized any more.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION This invention relates to construction drawings and other drawings widely used in the construction industry such as building and equipment industries, and
The present invention relates to a method for recognizing an image having contours and dividing lines (dividing lines) such as various forms, various tables such as statistical tables and data tables, and a recognition apparatus therefor.

[0002]

2. Description of the Related Art In recent years, using CAD systems, etc., construction drawings including houses, buildings, etc., construction drawings including water pipes, gas pipes, piping drawings for power / communication cables, etc. can be easily created, and the data thereof can be obtained. It has been practiced to memorize and store the information when design changes, extension and remodeling are performed. However, the data of the construction drawings are not compatible with each other due to the created system, and cannot be used due to the passage of time or the change of the contractor. In addition, when an extension or renovation of a house is performed, it is often the case that there are only old construction drawings drawn on paper, and it is necessary to redraw the entire floor plan including the portions that do not change the floor plans and the like.

Therefore, it has been attempted to read a construction drawing drawn on paper and recognize and store it as data that can be processed by a computer, but there is no special method or device for that purpose, and the construction drawing is scanned by an image scanner. Then, the image data is read into a personal computer or the like, and line segment recognition or pattern recognition is performed using a general figure recognition function.

Further, even if the figure recognition function is upgraded to a higher level by using a high-performance figure editor as an auxiliary, and even if the automatic recognition function is somewhat incomplete, basic lines such as straight lines and arcs are converted by raster-vector conversion. There is also a system that can automatically recognize the line diagram. It has also been attempted to recognize a ruled line of a form or the like by distinguishing it from a character or the like by recognizing continuous line segments by such a graphic recognition technique.

[0005]

However, such a conventional drawing recognition apparatus or image recognition apparatus requires a high degree of operation knowledge and is limited in usage to those who are accustomed to using a personal computer or a specialized operator. However, it has never been easy to use for those involved in the industry who frequently use construction drawings or clerks who frequently use forms and the like.

In the conventional recognition method or recognition device, basic diagrams such as straight lines and arcs can be automatically recognized, but from drawings and tables in which various line segments, characters and symbols are described, There is a problem in that it is difficult to accurately extract only the contour and the dividing line (partitions such as walls in the case of construction drawings), and it takes a lot of time and labor to correct the recognition result.

Further, the construction drawings of houses and buildings used in the construction industry have a thickness which is strictly the same even if each line segment on the drawing appears to the human eye as a continuous line of the same size. The trajectory is fluctuating as well. Even if it should be a continuous line, it may be broken in some places. These imperfections are caused not only at the time of drawing, but also as a result of aging of paper and errors at the time of reading. Therefore, for example, if the limit of the thickness that can be recognized as a line segment is too thin, it is recognized that the line segment is broken even with a slight blur.
In addition, a thick line segment may be recognized as a rectangle.

Not only the quality of the original drawing, but also many drawings with low contrast (the boundary between black and white is not clear) such as the so-called blueprint drawing using photosensitive paper,
Furthermore, since the detailed points that are characteristic of the blueprint appear on the surface, it is difficult to perform highly accurate automatic recognition with the conventional drawing recognition device, and it takes a lot of trouble to correct it, so it is hardly practical. It was

By the way, in the construction drawing of a house (architectural drawing), the wall that divides the floor plays a central role in the drawing, and by recognizing which part in the drawing is the wall, the floor plan of the construction drawing can be roughly estimated. I can understand. Therefore, recognizing the position and length of the wall in the drawing is the most important matter in recognizing the construction drawing. Further, in a table such as a form, it is important to recognize the ruled line by separating it from characters and symbols, but it is difficult to recognize the short dividing line in the table.

The present invention has been made in view of the above points, and is image image data obtained by reading an image having a contour and a dividing line in a drawing, a table, or the like, or encoded image data obtained by encoding the run length thereof. Outlines and division lines (partition lines) that are important information for knowing the drawings and tables
The purpose is to be able to recognize automatically with high accuracy. Thereby, for example, the position and length of the wall for knowing the floor plan and the like in the construction drawing can be recognized, and ultimately, the automatic integration and estimation can be performed from the recognition result. Alternatively, it is also an object to be able to accurately grasp the position and length of a ruled line composed of horizontal and vertical straight lines on a form or the like.

[0011]

In order to achieve the above object, the present invention provides the following image recognition method and image recognition apparatus. The image recognition method according to the present invention takes in image image data obtained by reading an image having contours and dividing lines such as drawings and tables, measures the number of black or white dots in the horizontal direction and the vertical direction of the image image data, and measures the horizontal direction. And the dot number measurement array data in the vertical direction are created, and the contour and the dividing line of the image are recognized based on the created dot number measurement array data in both directions.

Further, after recognizing the contours and dividing lines of the image based on the horizontal and vertical direction dot number measurement array data created as described above, the range limited based on the recognized contours and dividing lines The number of black or white dots in the horizontal and vertical directions of the image image data is measured for each time, and the dot number measurement array data in the horizontal and vertical directions is created again, based on the created dot number measurement array data in both directions. The recognition of the contour and the dividing line of the image within each of the limited ranges may be repeated until the new contour or the dividing line cannot be recognized.

Further, the image recognition apparatus according to the present invention comprises image data input means for inputting image image data obtained by reading an image having contours and dividing lines of drawings, tables, etc., and horizontal image data input by the means. Dot number measurement array data creating means for creating horizontal and vertical dot number measurement array data by measuring the number of black or white dots in the vertical and vertical directions, and the dot number measurement array data in both directions created by the means. Based on the above, there is a contour / dividing line recognition means for recognizing the contour and the dividing line of the image. The image data input means may be an image reading means for reading an image having a contour and a dividing line such as a drawing or a table drawn on paper and inputting the image data thereof.

Alternatively, the image data input means is a means for inputting coded image data obtained by coding the run length of image data obtained by reading an image having contours and demarcation lines such as drawings and tables, and measuring the number of dots. The array data creating means creates the dot number measurement array data in the horizontal and vertical directions by measuring the number of black or white dots in the horizontal and vertical directions of the original image image data from the input coded image data. May be. The image data input means in that case may be an image data reception means for receiving and inputting the encoded image data by communication.

Furthermore, in these image recognition devices,
The range of the image data is limited based on the contour and the dividing line recognized by the contour / dividing line recognizing means, and the number of dots in the horizontal and vertical directions is again determined by the dot number measuring array data creating means for each range. Creating the measurement array data, and causing the contour / dividing line recognition means to recognize the contours and the dividing lines of the image within the respective limited ranges based on the created bidirectional dot number measurement array data, a new contour or It is advisable to provide a means for repeating the dividing line until it cannot be recognized.

The contour / partition line recognizing means may be means for recognizing the peak value portion as a contour or a partition line by detecting the peak value of the dot number measurement array data. In that case, it is preferable that the contour / partition line recognition means has a means for analyzing whether or not there is a peak value of the dot number measurement array data for each of a plurality of types of predetermined intervals.

It is preferable that these image recognition devices are provided with display means for displaying the recognition result obtained by the contour / partition line recognition means. The display means preferably has means for displaying the image data inputted by the image data input means at the same time as the recognition result. Further, the means for displaying the image at the same time may be a means for displaying the image data by superimposing it on the recognition result.

Alternatively, display selection means for changing the display content may be provided so as to selectively display the image image data input by the image data input means and the recognition result. Further, it is desirable to provide a recognition result confirmation means for confirming the recognition result by the operator and confirming the recognition result.

[0019]

[Function] The outlines of walls and skeletons, and the ruled lines such as tables, which are the most important for knowing floor plans in construction drawings, generally have a black density for positive images and a white density for negative images. Especially high. Further, most of the line segments that make up the construction drawings and forms are drawn in one of two orthogonal directions (horizontal direction or vertical direction).

According to the image recognition method of the present invention, the number of black or white dots in the horizontal direction and the vertical direction is measured from the image image data obtained by reading the image having the contour and the dividing line of the drawing or the table, and the horizontal direction is obtained. And the vertical dot number measurement array data (histogram) is created, and the contours and dividing lines of the image are recognized based on the dot number measurement array data in both directions, so even for drawings and forms with poor image quality,
The contour and the dividing line can be recognized with high accuracy.

Further, based on the contour and the dividing line recognized by this method, the range to be recognized is limited, and the number of black or white dots in the horizontal and vertical directions of the image image data is measured for each range. To create the horizontal and vertical direction dot number measurement array data again, and to recognize the contours and demarcation lines of the image within each of the above-mentioned limited ranges based on the dot number measurement array data in both directions, to create a new contour or If you repeat it until you cannot recognize the dividing line, you can surely recognize even the partition line such as a short wall or ruled line inside the contour, and you may misrecognize the contour or partition line that does not actually exist. Also disappears.

Further, according to the image recognition apparatus of the present invention, the above-mentioned image recognition method can be easily implemented, and the contours and partition lines such as walls and ruled lines of the drawings and forms with relatively poor image quality can be easily implemented. Can be automatically and accurately obtained and the recognition data can be obtained, and if the data is loaded into a personal computer or the like and used, the drawings and tables can be easily and quickly changed and created.

The encoded image data obtained by encoding the run length of the image image data obtained by reading the image is input, and the number of black or white dots in the horizontal and vertical directions of the original image image data is measured from the encoded image data. It is also possible to generate horizontal and vertical dot number measurement array data, and the data amount of the image data can be compressed thereby, so that the capacity of the memory for storing it can be reduced. Further, it becomes easy to receive image data from the outside by a facsimile function or the like and perform recognition processing.

Furthermore, by detecting the peak value of the dot number measurement array data and recognizing the peak value portion as the contour or the dividing line, it is possible to particularly easily recognize the wall or ruled line. In that case, it is analyzed whether or not there is a peak value of the dot number measurement array data for each of a plurality of types of predetermined intervals, preferably for each interval corresponding to the minimum interval of dividing lines such as walls and ruled lines in the image to be recognized. As a result, the recognition accuracy of the dividing line can be improved.

By displaying the result recognized by the contour / dividing line recognition means on the display means, the recognition result can be confirmed. In that case, by displaying the image image data such as the drawing or the form input by the image data inputting means at the same time as the recognition result, it is possible to find the erroneously recognized portion or the unrecognized portion. At this time, if the image image data and the recognition result are displayed in an overlapping manner, it becomes easier to find and correct the erroneously recognized portion or the unrecognized portion. Alternatively, it is also possible to selectively display the input image image data and the recognition result, and compare the displayed contents. Then, the operator can confirm and confirm the recognition result.

[0026]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The embodiments of the present invention will be specifically described below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of an example of an image recognition apparatus for carrying out an image recognition method according to the present invention, in which a hardware configuration and a software processing function by a microcomputer are shown in a mixed manner.

This apparatus comprises an overall control unit 1, an image reading unit 2, a communication control unit 3, a memory 4, an automatic skew correction unit 5,
Dot number measurement array data creation unit 6, contour / dividing line recognition unit 7, remapping control unit 8, display unit 9, operation input unit 1
0, an external storage device 11, a printing device 12, and a bus 13 connecting them. It should be noted that the interface unit required between each of these units (or devices) and the bus 12 is not shown.

The overall control unit 1 is a microcomputer (CP) for controlling the operation and function of the entire image recognition apparatus.
It consists of U, ROM, RAM, etc.
U ”), and each function of the automatic skew correction unit 5, the dot number measurement array data creation unit 6, the contour / dividing line recognition unit 7, and the remapping control unit 8 according to the present invention is also performed by the CPU. Can be realized by the software processing of.

The image reading section 2 is an image data input means for scanning a set drawing, form, etc., reading the image, and inputting image image data. The image reading section 2 includes an image sensor such as a scan optical system and a CCD and its drive. This is a known image scanner including a circuit and the like. It also includes a circuit for binarizing the read image image data at a predetermined resolution into image data of white dots and black dots.

The communication control unit 9 receives the image data from the image reading unit 1 instead of receiving the image data from the image reading unit 1 and externally receives the image image data or the encoded image data in which the run length is encoded by the communication to input the image data. In addition to being a means, it is also possible to transmit the contour and dividing line data of the image recognized by this device to an external device. Specifically, it includes a FAX modem and a personal computer communication control means.

The memory 4 starts with the image image data read by the image reading unit 1, the image image data received by the communication control unit 3, or the encoded image data.
Image data skew-corrected by the automatic skew correction unit 5, dot number measurement array data created by the dot number measurement array data creation unit 6, and contour / separation line recognition unit 7
It is a memory such as a large-capacity RAM or a hard disk that stores the recognition result of the contour and the demarcation line recognized by, the image data remapped by the remapping controller 8, and the like.

The automatic skew correction section 5 is for adjusting the angle of the image data stored in the memory 4 so as to correct the horizontal and vertical line segment directions so as to match the horizontal and vertical reference directions of the apparatus. A known automatic skew correction technique can be used. The image data corrected by the automatic skew correction unit 5 is stored in the memory 4 again.

The dot number measurement array data creation unit 6 applies the automatic skew correction to the image image data or encoded image data stored in the memory 4 and the image data remapped by the remapping control unit described later. Then, the image data is limited to two directions, horizontal and vertical directions, and the number of black or white dots is measured (counted) in each dot width unit, and the result is the dot number measurement array data in the horizontal and vertical directions (histogram). ) And stores it in the memory 4.

When the read image is a positive image (the lightness of the image is lower than the lightness of the ground), the number of black dots is measured, and when the negative image is (the lightness of the image is higher than the lightness of the ground). Measures the number of white dots.

The contour / separation line recognition unit 7 recognizes the contour and separation line of the input image on the basis of the dot number measurement array data in the horizontal and vertical directions created by the dot number measurement array data creation unit 6, For example, in the case of an image of a construction drawing, it is a contour / dividing line recognition means for extracting wall data such as the position, length, thickness, and type of the wall as contours and dividing lines, the details of which will be described later. I will describe. If it is difficult or uncertain to recognize (extract) the contours and dividing lines with the referenced dot number measurement array data, the entire control of the request to reset the dot number measurement array data creation range in the horizontal and vertical directions of the image is performed. The data is sent to the unit 1, the creation range is changed and set, and the dot number measurement array data creation unit 6 is made to create the dot number measurement array data again.

The remapping control unit 8 limits the range of the image to be recognized by the part recognized as the contour or the partition line by the contour / partition line recognition unit 7, and considers other recognition information for that part. Then, mapping is performed again, and image data for causing the dot number measurement array data creation unit 6 to create the dot number measurement array data again for each limited range is created. Image data in which the reading noise in the reading unit 2 is removed is created. This data is also stored in the memory 4.

The display unit 9 receives image data such as a drawing which is input from the image reading unit 2 or the communication control unit 3 and skew-corrected by the skew correction unit 5, and the horizontal and horizontal lines created by the dot number measurement array data creation unit 4. It is for displaying vertical black or white dot number measurement array data, wall data recognized by the contour / separation line recognition unit 7, image data remapped by the remapping control unit 8, and the like. , CRTs and liquid crystal displays.

2 and 3 show examples of the display state of the screen 9a of the display unit 9, and FIG. 2 shows the outline and the dividing line of the construction drawing recognized by the contour / dividing line recognizing unit 7 (this In the example, it is a display example of image data (recognition result) obtained by remapping the data of the wall) by the remapping control unit 8. In this display example, double solid lines indicate both sides of the wall, and thin lines indicate the core line of the wall and its extension.

FIG. 3 shows an example in which the image data (input image image data) of the construction drawing skew-corrected by the automatic skew correction unit 5 and the image data of the re-mapped recognition result are displayed at the same time. Show.
In this case, the skew-corrected input image image data is displayed in halftone (indicated by a stippled line in FIG. 3 for convenience of illustration) so that the two can be easily distinguished, and the remapped recognition result image is displayed. Display the data as a solid line. When the display unit 9 is a color display device, the distinguishability can be improved by changing the colors of both images and displaying them. For example, by displaying the skew-corrected input image image data in light blue and the re-mapped image data of the recognition result in orange or green, the operator can easily identify the recognition result portion.

Alternatively, the screen 9a of the display unit 9 may be divided and the skew-corrected input image image data and the remapped image data of the recognition result may be displayed in contrast to each divided screen. . Alternatively, the skew-corrected input image image data and the re-mapped image data of the recognition result may be selectively displayed on the same screen. In that case, a display selection means (key or the like) may be provided in the operation input unit 10 described later.

Further, the display unit 9 also displays a screen for the operator to confirm the recognition result. That is, the image data of the remapped recognition result is displayed, and a display such as "Are you sure with this recognition result? (YES / NO)" is displayed. As a result, the operator can confirm whether or not the wall is correctly recognized.

The operation input unit 10 is for inputting various operation instructions, function selection instructions, edit data, etc., and is a keyboard, a mouse, a touch panel, or the like. The operation input unit 10 also has a function as a display selection unit,
The display state of the display unit 9 can be changed to a display state desired by the operator. For example, by key operation, the skew-corrected input image image data and the remapped image data of the recognition result can be displayed in an overlapping manner, or only one of them can be selected and displayed.

Further, the operation input section 10 inputs a key or the like for inputting "YES" or "NO" information in response to the above-mentioned "is this recognition result OK? (YES / NO)" display. We also have means. Then, if "YES" is selected, the recognition process is ended, and if "NO" is selected, the process proceeds to the re-recognition process or the correction process. This allows the operator to confirm the content of the recognition result and confirm it.

The external storage device 11 stores the input image data, the black dot number measurement array data created by the dot number measurement control unit 6, the contour and partition line data recognized by the contour / partition line recognition unit 7, The storage device stores the image data of the recognition result remapped by the remapping controller 8 in a storage medium such as a floppy disk (FD) or a magneto-optical disk (OMD) that can be taken out to the outside. The printing device 12 is a printer or plotter that prints or draws the above various data on paper and outputs the data.

Here, the "contour" when the recognition target is a construction drawing (mainly an architectural drawing) and the definition of "skeleton" which is a dividing line (partition line) and "outer wall" and "inner wall" are listed. 1
And Table 2 and FIG. The "outline" is an "outer contour", and only a portion facing the outside of the outer peripheral wall as indicated by a circle in Table 1 (the line outside the double line shown in (a) of FIG. 4). 1) meaning the case), and cases 2 and 3 meaning the entire wall in contact with the outside (the part indicated by the thick line in FIG. 4B). "Skeleton" means all of the walls (both of the parts shown by thick lines in (b) and (c) of FIG. 4)
Cases 1 and 2 meaning the case and Case 3 meaning the wall excluding the outer contour (only the part indicated by the thick line in FIG. 4C)
There are cases. In these definitions, unless otherwise specified, the normal meaning of Case 1 is used.

[0046]

[Table 1]

[0047]

[Table 2]

The "outer wall" means only the portion facing the outside of the outer peripheral wall as shown in FIG.
Of the case 4 which means the part outside the double line shown in (a) of FIG. 4 and Case 5 which means the entire wall in contact with the outside (the part shown by the thick line in FIG. 4B). There are cases. The “inner wall” is a wall excluding the outer wall (wall = outer wall + inner wall) in both cases 4 and 5 as indicated by a circle in Table 2.
In the case of Case 4, it is the portion of the line inside the double line shown in FIG. 2 and the portion shown in bold in FIG. 4 (c), and in the case of Case 5, the portion shown in bold in FIG. 4 (c). is there. In these definitions as well, unless otherwise specified, the normal meaning of Case 5 is used.

Next, a procedure for recognizing a construction drawing (mainly a construction drawing of a house or a building) by the image recognition apparatus shown in FIG. 1 will be described with reference to the flow charts of FIGS. In these flow charts, each step is indicated by "S". Further, in this embodiment, it is assumed that the recognized construction drawing is a positive drawing. FIG. 5 is a flow chart showing a procedure for recognizing the position of the wall as the contour and partition line of the image.

First, in step 1, the image of the construction drawing set in the image reading unit is read and the binarized image image data is input. Alternatively, the communication control unit 3 may externally receive and input image data of the construction drawing (image image data or encoded image data obtained by encoding the run length).

In step 2, the contour / partition line recognition unit 7 based on the horizontal and vertical black dot number measurement array data created by the dot number measurement array data creation unit 6.
In order to improve the accuracy of the wall recognition processing by, the automatic skew correction of the input image data is performed. Since this automatic skew correction may be performed by some conventionally known methods, detailed description of its contents will be omitted.

Next, in step 3, the entire image data that has been automatically skew-corrected is examined. And
In step 4, the nest variable is 0 (initial value: means to target the entire image). “Nest variable” represents the stage of narrowing down the analysis range of construction drawings from the top down. The larger the value of the nested variable, the deeper the analysis (advance to the finer part).

In step 5, processing is performed to limit the area to be investigated at the wall position. Specifically, the instruction of the investigation area is shown immediately before this processing, and here, only preparation for the processing of the subsequent steps 6 to 9 (standardization of interface) is performed. The shape of each surveyed area is a rectangular diagram. Initially, the nest variable is 0, so the entire drawing is targeted for investigation.

In step 6, horizontal black dot number measurement array data is created. This is to count (measure) the number of black dots in the horizontal direction for each one-dot width in the vertical direction of the image data, and hold each count data.
Next, in step 7, a wall extraction (recognition) process is performed based on the horizontal black dot number measurement array data created in step 6. The processing procedure will be described later.

In step 8, as in step 6, vertical black dot number measurement array data is created. That is, the number of black dots in the vertical direction is counted (measured) for each 1-dot width in the horizontal direction of the image data, and each count data is held. In step 9, a wall extraction (recognition) process is performed based on the black dot number measurement array data in the vertical direction.

Then, in step 10, it is determined whether there is a wall candidate for dividing the area. If there is at least one wall candidate that divides the region in the horizontal or vertical direction, go to step 11 and go to step 13 if there is no such wall candidate. For example, it is determined whether or not there is a wall candidate Wd that divides the area in the area to be investigated (the area surrounded by the wall W) as shown in FIG.

In step 11, the nest variable is incremented by 1 and reset. This means entering a step of recognizing a wall from the current analysis region and limiting the analysis range to a small region within the current region newly partitioned using the wall. Then, in step 12, the region to be investigated is subdivided. Specifically, with respect to the core lines (center lines) of the wall candidates recognized in step 7 and step 9, for example, as shown in FIG. 8, the investigation target region is divided into regions Sa1 and Sa2 by the core lines indicated by the thin lines of the wall candidates W and Wd. Divide into two. Further, the survey target is positioned in the first subdivided region (for example, the upper left region).

The order of analysis in this newly subdivided area group does not require a special order, but, for example, it should be performed in the order of smaller x and y coordinate values of the area start position. And so on. FIG. 9 shows an example of the nest No. of each area subdivided by the wall candidate and the analysis processing order thereof, where the solid line is the core line (center line) of the wall candidate ,.
Indicates the nest No., and the small numbers 1 to 8 indicate the processing order.

Thereafter, the process returns to step 5 and the above-mentioned processing is repeated. In step 10, if there is no wall candidate for dividing the area, the process proceeds to step 13 and it is determined whether or not there is a remaining homogeneous nest area (area of the same nest No. in FIG. 9). If there is any remaining space, the investigation target is advanced to the area next to the homogeneous nest in step 15, and the process returns to step 5.

If there is no remaining homogeneous nest area, the process proceeds to step 14 to judge whether the nest variable is 0 or not. If the nest variable is not 0, the nest variable is decremented by 1 in step 16 and the process returns to the process of the nest region one level higher, and it is determined in step 13 whether there is a remaining nest region at that stage. If there is, step 15 advances the investigation target to the next nest area and returns to step 5.

If the nest variable is 0 in step 14,
Proceeding to step 17, the position and size of each wall are determined based on the recognition data of the wall candidate for each area obtained in steps 7 and 9, and the wall recognition data is stored in the memory 4. The storage method will be described later. Then, in step 18, the recognition data is displayed on the display unit 9 as necessary, printed by the printing device 12, or stored in the external storage device 11, and the process ends.

Next, the procedure of the wall extraction (recognition) process in steps 7 and 9 of FIG. 5 will be described with reference to FIG. 6. Prior to that, the construction drawing (house floor plan) which is a construction drawing will be described. 10 and 11 show specific examples of the horizontal and vertical black dot number measurement array data created from the image data and the entire area thereof. In these drawings, (a) is image data of architectural drawings, (b) is horizontal black dot number measurement array data, and (c) is vertical black dot number measurement array data. Furthermore, the vertical black dot number measurement array data shown in FIG. 10C is enlarged and shown in FIG.

In the architectural drawing of FIG. 10, the wall symbol is represented by both sides of the wall and the core line, and in the architectural drawing of FIG. 11, the wall symbol is filled in within the wall thickness (black).
Represented by. In this black dot number measurement array data, the width of the thinnest black line is 1 dot width, and the length of each black line is horizontal or vertical for each 1 dot width of the image data of the architectural drawing shown in (a). This corresponds to the count value of the number of black dots in the direction. As is clear from these figures, most (90% or more) of the line segments that make up the architectural drawing are drawn in the horizontal or vertical direction, and the density of black dots is particularly high in the wall part. . Therefore, a peak appears at the position where the wall exists in the horizontal and vertical black dot number measurement array data.

When the processing of the flow of FIG. 6 is started, first in step 21, a direction intersecting with the designated direction (vertical direction if designation of black dot number measurement array data in the horizontal direction is made, and vertical direction designation is made). Horizontally if present)
First, it is confirmed how many times the wall recognition processing loop can be performed at a predetermined interval corresponding to the reference unit length of the construction drawing. here,
In the case of a general house, this unit length is the minimum wall interval of half a meter or 1 meter.
cm).

For the black dot number measurement array data in the horizontal direction, the width size L is set to be slightly longer than the image width in the vertical direction, and for the black dot number measurement array data in the vertical direction, the image in the horizontal direction is set. A dimension slightly longer than the width is automatically set as the width size L. Further, the half size is set to h, and this value is determined by inputting the scale data and half length of the drawing in advance or by automatic calculation by calculation or the like. L / h is calculated from this width size L and half size and h is rounded up to the right of the decimal point as the number of times n the large loop of the wall recognition process is executed. The range indicated by h in FIG. 12 is the processing range in one large loop.

Next, at step 22, the count value i of the large loop counter is initialized (i ← 1).
Then, in step 23, the count value i of the number counter exceeds the number n of executions of the large loop that is possible (i> n).
Determine whether or not. If it exceeds, the analysis of the area is finished. If it does not exceed, it is positioned at the highest peak (indicated by P in FIG. 12) as the first analysis processing in the i-th half in step 24, and the point is set as xp. By performing the analysis processing every half period in this way, it is highly possible that the peak value therein is a part of the wall.

Next, in step 25, it is determined whether or not the height of the peak (maximum value), which is the first condition for the target of the wall, is equal to or greater than a half (h). As a result, when the height of the peak is less than half, it is determined that the wall is not recognized, and the process proceeds to step 34 to move to the next half analysis. When the peak height is more than half, the process goes to the next analysis step 26. In this step 26, the left and right sides (for example, 2 W width) are checked from the position of the highest peak, and the range of wall thickness (positions on both sides: W ₁ , W ₂ and thickness We = | W ₁ −W ₂ |)
Narrow down. Here, W means the thickness of the wall, and is used in the same value as Wmax, for example.

As this narrowing down method, the peak positions on both ends or one side of the highest peak position having a value of (maximum peak value-min) * rate + min or more are determined as positions W ₁ and W _{2 on} both sides of the wall.
Alternatively, there is a method of narrowing it down as the position W ₂ of the other surface when the highest peak position xp is the position W ₁ of the _one surface of the wall. FIG. 13 is an explanatory diagram of this narrowing-down process.
Is the positions W ₁ and W _{2 on} both sides of the wall on both sides of the highest peak position xp
It is an example of the case where there exists. In this case, the highest peak position xp is estimated as a candidate for the core line position of the wall.

FIG. 13B shows an example in which the highest peak position xp is the position W ₁ on _one surface of the wall, and the position W ₂ on the other surface exists on one side thereof. In this case, the longer peak position can be estimated as the outer wall position candidate of the outer contour shown by the double line in FIG. 2, and the shorter peak position in the vicinity thereof can be estimated as the inner wall position candidate. Where rate is the first half of the analysis
Set small (when the value of the nest variable is small) and small in the latter half (when the value of the nest variable is large) (for example, initially set rate = 0.70). min is the minimum value of the black dot number measurement data within a width of about 4 W which is spread to the left and right 2 W on both sides of the position xp of the maximum peak P currently under consideration as shown in (c) of FIG.

In this way, the validity as a wall is confirmed in steps 27 and 28 based on the result narrowed down in step 26 of FIG. First, in step 27, it is judged whether or not the wall thickness We exceeds its maximum value Wmax (for example, 30 cm), and if it exceeds, it is determined that the wall recognition is unknown, and the process proceeds to the next half-way analysis in step 34. Proceed to.
If it does not exceed, step 28 is proceeded to, where it is determined whether or not the wall thickness We is less than the minimum value Wmin (for example, 2.5 cm).

As a result, if the wall thickness We is less than the minimum value Wmin, the process proceeds to step 30, and if not, the process proceeds to step 29. In step 30, the information is obtained assuming that a peak other than the wall (for example, tatami mat, window, sliding door, etc.) is recognized. In step 29, whether or not the condition as a wall is satisfied is determined from the candidate region of the wall by a half-division search method described later, and if it is recognized as a wall, in step 31 both surfaces of the wall are recognized. Information such as the positions W ₁ and W ₂ and the thickness We are saved (stored) and the process proceeds to step 32.

If the wall condition is not satisfied in step 29, the process proceeds to step 32. In step 32, it is determined whether the both-sided positions (coordinates) W ₁ and W ₂ of the wall are both inside the region currently being processed, and if it is inside, the process proceeds to step 33. If not, the process proceeds to step 34. In step 33, the current processing area is subdivided, and W ₁ ,
W ₂ and We are saved (stored). Also, (W ₁ + W ₂ ) /
2 is the core position of the wall. In step 34, the count value i of the large loop number counter is incremented by 1 to perform the next half-way ahead processing, and then the process returns to step 23 to repeat the above processing.

In this manner, the analysis processing is performed halfway from one end (leading element) of the target black dot number measurement array data, and when the analysis is completed up to the opposite end, the analysis of the current range is completed. The horizontal and vertical directions are separately analyzed for the black dot number measurement array data, and the next analysis range is limited to the range that can be recognized as a wall in this analysis. Therefore, the results of analyzing the black dot number measurement array data in the horizontal and vertical directions are combined to perform a brute force case classification.

In this embodiment, it is considered that the core line position of a wall is arranged such that the interval between the wall and the adjacent wall is an integral multiple of half a unit. Then, the range up to which the characteristic peak can be found is effectively used as analysis data. Conversely, if no feature corresponding to a wall within a certain range to be analyzed cannot be found in both the horizontal and vertical directions, the analysis ends. The analysis range of one time is made to cover the entire drawing at first, and the method is limited to the range separated by the wall that was discovered after that, so that the peak of the wall can be easily found and the analysis of the details is easy. Try to be.

Next, the half-division search method for judging "whether the condition as a wall is satisfied" in step 29 of FIG. 6 will be described with reference to the flowchart of FIG. When the process of the flow shown in FIG. 7 is started, first, in step 41, the bisector search method is initialized. That is, the number n of area dividing elements for wall analysis is set to 1, k that indicates the smallest No. of array elements of dividing elements whose wall portion or non-wall portion is undetermined is set to 0, and As a split array element,
S

[0], E

[0] The input start and end image addresses are assigned to each to initialize.

After that, the process proceeds to step 42, and if the start address S [k] and the end address E [k] of the investigation area are equal, the process proceeds to step 53. If they are not equal, the process proceeds to step 4
Go to 3. In step 43, the target area of the image data is divided, and the original area is divided into two equal parts by removing the decimal error or less. That is, the start and end image addresses S1 and E1 of the first half area to be divided are
S1 = S [k], E1 = (1/2) (S [k] + E [k])
And the start and end image address S of the latter half area
2, E2, S2 = (1/2) (S [k] + E [k]) +
1, E2 = E [k].

As a result, as shown in FIG. 14, for example, in the image data of the construction drawing, the original area width of the line (indicated by the alternate long and short dash line) where the wall candidate exists is initially divided into two equal parts at the position of 1. To do. After that, until the presence / absence of the wall can be determined, the position 2 shown in FIG.
The second division is repeated at the fourth and fourth positions, so that the wall survey in steps 44 and 45 is performed in the subdivided region. In steps 44 and 45, it is investigated whether or not the halved regions P1 and P2 are filled with walls, and the process proceeds to step 46.

Here, as a result of analyzing the black dot number measurement array data in the designated area (from the start address to the end address), the height of the peak of the black dots (keeping the spread of the wall thickness) is designated. Compared with the width (length) of the area, the following items are judged. : When 5% or less, v = 0: Judge as non-wall part: When 95% or more, v = 1: Judge as wall part: Other than the above, v = 2: Neither can be judged Then, as shown in FIG.

In step 46, with respect to the first half region P1 divided in step 43, if it cannot be determined whether or not a wall exists (v = 2), the process proceeds to step 47, and if not, the process proceeds to step 49. Step 47
, The storage array elements S [k], E [k], v [k] in the area range before being divided in step 43 are converted into the latter half area data (start and end image addresses S2, S2). Replace E2 with the determination result v2).

In step 48, new storage array elements S [n], E [n], v [n] are set as step 4
The first half area data (start and end image addresses S1, E1 and determination result v1) divided in 3 are saved, and the process proceeds to step 51. In step 49, the storage array elements S [k], E [k], v [k] in the area range before being divided in step 43 are set to the first half area data (start and end image addresses) divided in step 43. Replace S1 and E1 with the determination result v1).

In step 50, new storage array elements S [n], E [n], v [n] are set as step 4
The latter half area data (start and end image addresses S2 and E2 and determination result v2) divided in 3 are saved, and the process proceeds to step 51. In step 51, the variable n indicating the new storage array element No. is incremented by 1 so as to indicate the new storage array element of the area address, and then the process proceeds to step 52.

In step 52, the classification code v in the k element, which points to the smallest No. of the divided elements for which it is uncertain whether the wall portion or the non-wall portion is unknown, does not indicate whether a wall exists (v = 2). In the case of), the process returns to step 42 and the process of further subdividing is repeated. Otherwise, go to step 53. In step 53, it is assumed that one of the contents, which is not known whether or not there is a wall, has been solved, and the index k is incremented by 1, and then the process proceeds to step 54.

In step 54, the number n of area dividing elements for wall analysis is equal to k indicating the smallest No. of array elements of dividing elements whose wall portion or non-wall portion is undetermined. If it is equal, it is determined that the dividing process for wall recognition is completed, and the process proceeds to step 55.
If they are not equal, the process returns to step 52. In step 55, sorting is performed in the order of start address (ascending order) so that the divided area address data (array) are arranged in ascending order, and the process proceeds to step 56. That is, S

[0] to S [n-1], E

[0] to E [n-1], v

The data of [0] to v [n-1] are sorted in ascending order using S [] as a key.

In step 56, when the wall portion and the non-wall portion are continuous, the degeneracy process (represented by one range) is performed and the process ends. That is, consecutive v
[] If the value is 0 or 1, S [], E
[] Compress the data. At this time, if an array element having a content (v = 2) in which it is not known whether or not a wall exists is included, if the elements before and after the element indicate a wall, the data is changed to wall data, and If a non-wall is indicated, the data is changed to non-wall data for processing.

FIG. 16 shows an example of analysis of wall positions in image data by the above-described bisector search process. In FIG. 16A, an image image W of the wall (hatched portion) W and its investigation target area are shown by broken lines, and this investigation target area is along the existing position of the previously recognized wall candidate. Is set. And S

[0] = 0 is the first start address of the investigation area, E

[0] = 15 is the first end address. The numbers in the middle of 4, 5, 7, etc. are the start or end addresses (both are image addresses) of the surveyed area after division.

FIG. 16B shows the judgment result v regarding the start address S, the end address E, and the presence / absence of a wall of each investigation target area in each investigation stage of the variables n = 1 to 10, and the determination status and sorting status. , And the result of the degeneration process are shown together with the value of k. Finally, it recognizes that a wall exists at the image addresses 5 to 12.

Next, referring to FIG. 17, an example of recognizing a wall which is a contour and a dividing line of a construction drawing will be described. In FIG. 17,
The nest variable (nest) and the actual state of the wall recognized as the area division state are shown. First, nest variable = 0
Then, the wall is extracted with the entire construction drawing as the object of wall position survey. As a result, it is assumed that the outline of the house in the construction drawing and the positions of the horizontal and vertical wall candidates can be recognized as indicated by the solid line in (A). However, among the recognized wall candidates, the actual wall is only the portion shown in (a), but it is still unknown.

Then, next, the nest variable is set to 1 and
Walls are extracted by limiting the investigation target area for each rectangular area delimited by the core lines of the recognized wall candidates shown in (A).
As a result, in (B), the wall candidates newly indicated by thick lines are recognized in the four survey target areas indicated by ,, and among the previously recognized wall candidates, the actual wall is shown in (a). It is confirmed that only the part where the wall is formed is present, and the state of the wall shown in (b) is recognized.

Further, the nest variable is set to 2, and the four regions in which the new wall candidate is recognized in (B) are divided by the newly recognized walls to further limit the survey target region. Extract the wall. As a result, in (C), the wall candidate newly indicated by the thick line is newly recognized in the two investigation target areas, and the wall state shown in (c) is recognized.

After that, the nest variable is set to 3, and the two regions in which the new wall candidate is recognized in (C) are divided by the newly recognized walls to further limit the survey target region. Extract the wall. As a result, if it is not possible to extract a new wall candidate in any of the divided areas, the wall position investigation is ended, and
It is determined that the wall position shown in (c) is the final wall recognition result, and the data is stored in the memory.

In this way, until no new wall candidate is extracted in any of the divided survey target areas, the survey target area is subdivided and walls are extracted. Thereby,
Even a small wall can be surely recognized, and a part where the wall actually exists and a part where the wall does not actually exist can be accurately discriminated.

Next, information (analysis result data) such as the position and size of the wall corresponding to the outline and skeleton of the construction drawing recognized as described above is stored in the memory 4 and the external storage device 11 shown in FIG. An example of contents stored in the storage medium will be described with reference to FIG. In FIG. 18, (A)
Is the number of nests, (B) is unique nest information, (C) Horizontal wall information, (D) is vertical wall information, (E) is a horizontal wall model, and (F) is a vertical wall model. Indicates.

The nest number indicates the depth of nesting (parent-child relationship) of the child nest pointer, and if the wall is not recognized at all, the nest number = 0. The unique nest information includes a nest No., a NEXT sibling pointer, a child nest pointer, the number of walls (horizontal direction and vertical direction), and a wall information pointer (horizontal direction and vertical direction).

The NEXT sibling pointer has the next address of the simultaneous hierarchy nest information. Therefore, the nest No. in the unique nest information of the location indicated by this pointer. Is the same value as that of the processing. The child nest pointer has the address of the top data of the hierarchical nest information one level below.
Therefore, the nest No. in the unique nest information at the location indicated by this pointer is a value obtained by adding +1 to that of the process.

The horizontal wall information shown in (C) is stored with the address indicated by the horizontal wall information pointer in the unique nest information shown in (B) as the start address. The wall information is a NEXT pointer indicating the position of the start address of the next horizontal wall information, and the starting point coordinates (x coordinate: a,
It is composed of y coordinate: b), lateral (x direction) size of the wall: c, vertical (y direction) size of the wall: wall thickness d. These a ~
The model of the horizontal wall shown in (E) can be stored and reproduced by d.

Similarly, the address indicated by the vertical wall information pointer in the unique nest information is set as the start address,
The vertical wall information shown in (D) is stored. The wall information is a NEXT pointer indicating the position of the start address of the next vertical wall information, the wall start point coordinates (x coordinate: e, y coordinate: f), the wall vertical (y direction) size: g, the wall Lateral (x direction) size: consists of wall thickness h. With these e to h, it is possible to memorize and reproduce the model of the vertical wall shown in (F).

Here, an example of recognizing the contours and ruled lines (dividing lines) of the table will be described with reference to FIG. 19 in the same manner as the wall recognition in the construction drawing described with reference to FIG. Most of the contours and ruled lines of such a table extend in the horizontal or vertical direction. When recognizing a table having complicated ruled lines as shown in FIG. 19 (A), after inputting the image data of the image, first, with the nest variable value = 0, the entire image is set as the area to be investigated, and the contour and ruled lines are set. Recognize. As a result, the outline indicated by the thick line and the long ruled line indicated by the thin line in (B) are recognized.

Next, the nest variable is set to 1, and the contours and the ruled lines are recognized by using the respective regions separated by the contours and the ruled lines recognized in FIG. As a result, a short ruled line indicated by a thick line in (C) is newly recognized, and it is recognized that there is no contour or ruled line in the portion indicated by the broken line. Therefore, the nest variable is set to 2, and the contours and ruled lines are recognized by using each region subdivided by the newly recognized ruled lines as a region to be investigated. As a result, a short ruled line indicated by a thick line in (D) is newly recognized.

Further, with the nest variable = 3, the contours and ruled lines are recognized as the areas to be investigated, which are each subdivided by the newly recognized ruled lines. I can't recognize it.
As a result, the survey of contour and ruled line position is completed, and (D)
It is confirmed that the positions of the contours and ruled lines shown in are the final recognition results, and the data are stored in the memory. In this way, even in a table having a non-rectangular contour or a complicated short ruled line, the contour and each ruled line can be accurately recognized.

In the above-described embodiment, since the positive image is the recognition target, the number of black dots in the horizontal and vertical directions of the binarized image data is measured (counted) to create the black dot number measurement array data. However, when a negative image is to be recognized, if the number of white dots in the horizontal and vertical directions of the binarized image data is measured (counted) and white dot number measurement array data is created, that image The contours and the dividing lines can be similarly recognized. Further, the recognition data relating to the contours and dividing lines (walls, ruled lines, etc.) of the drawings and tables recognized as described above can be converted into vector data for CAD, and the compatibility of CAD data between different models. Can be obtained.

[0101]

As described above, according to the image recognizing method and the image recognizing apparatus of the present invention, a drawing having a relatively low contrast such as blue-green and the like, which have been difficult to be accurately recognized in the past, and a large amount of noise are generated. Even a drawing with poor image quality such as a drawing or an old construction drawing, or a table with poor image quality such as an old form can easily and accurately recognize the contours and dividing lines thereof, that is, walls and ruled lines.

In particular, the target range is limited so as to subdivide the recognition target based on the contour and the dividing line initially recognized for the entire image, and the horizontal and vertical directions of the image data are determined for each range. Creating black or white dot number measurement array data in the vertical direction, and recognizing the contours and dividing lines of the image within each of the above-mentioned limited ranges based on the dot number measurement array data in both directions, a new contour or By repeating the process until the dividing line cannot be recognized, a short wall, ruled line, or the like can be surely recognized, and there is no risk of erroneously recognizing the contour or ruled line of a portion that does not actually exist.

Further, if the recognized data is loaded into a personal computer or the like and used, it is possible to easily and quickly create / change a construction drawing such as a sketch at the time of extension / renovation, or create / change a revision format of a table. It will be possible. Further, the image data of the image can be input as the run-length encoded image data by FAX communication or the like, and the contour and the dividing line of the image can be recognized. In addition, by analyzing whether or not there is a peak value in the dot number measurement array data at every predetermined interval corresponding to the minimum partition line interval (wall interval or ruled line interval) in general construction drawings and forms, The line recognition accuracy can be improved.

Further, by displaying the result recognized by the contour / divider line recognition means, the recognition result can be confirmed and confirmed. In this case, by displaying the input image image data at the same time as the recognition result or selectively, the display contents can be compared and examined to find the erroneous recognition portion or the unrecognized portion.
If the input image data and the recognition result are displayed in a superimposed manner, it becomes easier to find and correct the erroneously recognized portion or the unrecognized portion.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an example of an image recognition apparatus for implementing an image recognition method according to the present invention.

FIG. 2 is a diagram showing a display example of image data of a remapped recognition result on a display unit 9 of FIG.

FIG. 3 is a diagram showing an example in which input image data of a construction drawing that is also skew-corrected is superimposed and displayed on image data of a recognition result.

FIG. 4 is a diagram for explaining the definition of an outer contour, an outer wall, an inner wall, and a skeleton in a construction drawing.

5 is a flowchart showing a procedure for recognizing the position of a wall by the image recognition device shown in FIG.

6 is a flow chart of a subroutine of a wall extraction (recognition) process of steps 7 and 9 in FIG.

FIG. 7 is a flow chart for executing the bisected search method for recognizing the position of the wall.

FIG. 8 is an explanatory diagram showing a wall candidate in the investigation target area and an example of area division by the wall candidate.

FIG. 9 is a nest N of region groups subdivided by wall candidates
o. FIG. 6 is an explanatory diagram showing an example of an analysis processing sequence of the above.

FIG. 10 is a diagram showing a specific example of image data of a building drawing (floor plan of a house) and horizontal and vertical black dot number measurement array data created from the entire area thereof.

FIG. 11 is a diagram showing another specific example of the same.

FIG. 12 is an enlarged view showing vertical black dot number measurement array data shown in FIG. 10 (c).

FIG. 13 is an explanatory diagram of a process of narrowing both sides of the wall to both sides or one side of the peak in step 26 of FIG.

14 is an explanatory diagram of region width division processing in step 43 of FIG. 7. FIG.

15 is an explanatory diagram of the determination by wall survey of the target area in steps 44 and 45 of FIG.

16 is an explanatory diagram showing an example of analysis of wall samples (wall candidates) by the bisector search process shown in FIG. 7. FIG.

FIG. 17 is an explanatory diagram of an example of recognizing a wall of a construction drawing according to the present invention.

FIG. 18 is an explanatory diagram showing an example of stored contents of analysis result data in a memory.

FIG. 19 is an explanatory diagram showing an example of recognizing contours and ruled lines of a table according to the present invention.

[Explanation of symbols]

 1: Overall control unit 2: Image reading unit 3: Communication control unit 4: Memory 5: Automatic skew correction unit 6: Dot number measurement array data creation unit 7: Contour / partition line recognition unit 8: Remapping control unit 9: Display Part 10: Operation Input Part 11: External Storage Device 12: Printing Device 13: Bus

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] November 9, 1995

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] 0027

[Correction method] Change

[Correction content]

This apparatus comprises an overall control unit 1, an image reading unit 2, a communication control unit 3, a memory 4, an automatic skew correction unit 5,
Dot number measurement array data creation unit 6, contour / dividing line recognition unit 7, remapping control unit 8, display unit 9, operation input unit 1
0, an external storage device 11, a printing device 12, and a bus 13 connecting them. It should be noted that an interface section required between each of these sections (or devices) and the bus 13 is omitted in the drawing.

[Procedure Amendment 2]

[Document name to be amended] Statement

[Name of item to be corrected] 0030

[Correction method] Change

[Correction content]

The communication control section 3 receives image data from the outside, instead of receiving the image data from the image reading section 2 , by receiving and inputting the image image data or the coded image data in which its run length is encoded by communication. In addition to being a means, it is also possible to transmit the contour and dividing line data of the image recognized by this device to an external device. Specifically, it includes a FAX modem and a personal computer communication control means.

[Procedure 3]

[Document name to be amended] Statement

[Correction target item name] 0031

[Correction method] Change

[Correction content]

The memory 4 starts with image image data read by the image reading unit 2 , image image data or coded image data received by the communication control unit 3,
Image data skew-corrected by the automatic skew correction unit 5, dot number measurement array data created by the dot number measurement array data creation unit 6, and contour / separation line recognition unit 7
It is a memory such as a large-capacity RAM or a hard disk that stores the recognition result of the contour and the demarcation line recognized by, the image data remapped by the remapping controller 8, and the like.

[Procedure amendment 4]

[Document name to be amended] Statement

[Name of item to be corrected] 0032

[Correction method] Change

[Correction content]

The automatic skew correction unit 5 serves to correct so as to adjust the angle of the image data stored in the memory 4 is coincident with the horizontal and the reference direction of the vertical device segment direction of the horizontal and vertical Yes, a known automatic skew correction technique can be used. The image data corrected by the automatic skew correction unit 5 is stored in the memory 4 again.

[Procedure Amendment 5]

[Document name to be amended] Statement

[Name of item to be corrected] 0037

[Correction method] Change

[Correction content]

The display unit 9 receives image data from the image reading unit 2 or the communication control unit 3 and is subjected to skew correction by the automatic skew correction unit 5 such as drawing image data and dot number measurement array data.
Horizontal and vertical direction black or white dot number measurement array data created by the data creation unit 6 , wall data recognized by the contour / separation line recognition unit 7, image data remapped by the remapping control unit 8, etc. For displaying, for example, a CRT or a liquid crystal display.

[Procedure correction 6]

[Document name to be amended] Statement

[Correction target item name] 0044

[Correction method] Change

[Correction content]

The external storage device 11 stores the input image data, the black dot number measurement array data created by the dot number measurement array data creation unit 6 , and the contours and partition lines recognized by the contour / partition line recognition unit 7. The data, the image data of the recognition result remapped by the remapping control unit 8 and the like are stored in a floppy disk (FD) or a magneto-optical disk (OM).
D) is a storage device for storing in a storage medium that can be taken out to the outside. The printing device 12 is a printer or plotter that prints or draws the above various data on paper and outputs the data.

[Procedure Amendment 7]

[Document name to be amended] Statement

[Correction target item name] 0048

[Correction method] Change

[Correction content]

The "outer wall" means only the portion facing the outside of the outer peripheral wall as shown in FIG.
Of the case 4 which means the part outside the double line shown in (a) of FIG. 4 and Case 5 which means the entire wall in contact with the outside (the part shown by the thick line in FIG. 4B). There are cases. The “inner wall” is a wall excluding the outer wall (wall = outer wall + inner wall) in both cases 4 and 5 as indicated by a circle in Table 2.
For case 4 is a portion indicated by a bold line in (c) of the double line of the line inside part and 4 shown in (a) of FIG. 4, in the case of the case 5 thick line in FIG. 4 (c) Is the part indicated by. In these definitions as well, unless otherwise specified, the normal meaning of Case 5 is used.

[Procedure Amendment 8]

[Document name to be amended] Statement

[Correction target item name] 0050

[Correction method] Change

[Correction content]

First, in step 1, the image of the construction drawing set in the image reading unit is read and the binarized image image data is input. Alternatively, the communication control unit 3 may externally receive and input image data of the construction drawing (image image data or encoded image data obtained by encoding the run length).

Claims (14)

[Claims] 1. A method for capturing image data obtained by reading an image having contours and dividing lines in drawings, tables, etc., and measuring the number of black or white dots in the horizontal and vertical directions of the image data to determine the horizontal and vertical directions. An image recognition method characterized in that the dot number measurement array data is created, and the contours and the dividing lines of the image are recognized based on the created dot number measurement array data in both directions. 2. Image data obtained by reading an image having contours and dividing lines in drawings, tables, etc. is read, and the number of black or white dots in the horizontal and vertical directions of the image data is measured to determine the horizontal and vertical directions. To create the dot number measurement array data, to recognize the contour and the dividing line of the image based on the created dot number measuring array data in both directions, and for each limited range based on the recognized contour and the dividing line. Image The number of black or white dots in the horizontal and vertical directions of the image data is measured to create the dot number measurement array data in the horizontal and vertical directions again, and the above-mentioned limitation is made based on the created dot number measurement array data in both directions. Image recognition, characterized in that recognition of contours and delimiters of the image within each range is repeated until a new contour or demarcation line cannot be recognized. Knowledge method. 3. Image data input means for inputting image image data obtained by reading an image having contours and dividing lines of drawings, tables, etc., and horizontal or vertical black or white of the image image data input by said means. A dot number measurement array data creating means for measuring the number of dots and creating dot number measurement array data in the horizontal and vertical directions, and a contour and division of the image based on the dot number measurement array data in both directions created by the means. An image recognition apparatus comprising: a contour / dividing line recognition means for recognizing a line. 4. The image according to claim 3, wherein the image data input means is an image reading means for reading an image having a contour and a dividing line of a drawing or a table drawn on a paper and inputting the image data thereof. Recognition device. 5. Image data input means for inputting encoded image data obtained by encoding run lengths of image image data obtained by reading an image having contours and division lines of drawings, tables, etc., and encoding input by the means. Dot number measurement array data creating means for creating horizontal and vertical dot number measurement array data by measuring the number of black or white dots in the horizontal and vertical directions of the original image image data from the image data, and creating by the means An image recognition device, comprising: a contour / dividing line recognition means for recognizing the contour and the dividing line of the image based on the generated dot number measurement array data in both directions. 6. The image recognition apparatus according to claim 5, wherein the image data input unit is an image data receiving unit that receives and inputs the encoded image data by communication. 7. The image recognition apparatus according to claim 3, wherein a range of the image image data is limited based on the contour and the dividing line recognized by the contour / dividing line recognizing means, The dot number measurement array data creation means again creates the dot number measurement array data in the horizontal and vertical directions for each range, and the contour / divider line recognition means based on the created dot number measurement array data in both directions. An image recognition apparatus comprising: means for recognizing the contour and the dividing line of the image within each of the limited ranges described above until the new contour or the dividing line cannot be recognized. 8. The image recognition apparatus according to claim 3, wherein the contour / partition line recognition means detects a peak value of the dot number measurement array data, and detects the peak value portion. An image recognition device characterized by being means for recognizing a contour or a partition line. 9. The image recognition device according to claim 8,
The image recognition device, wherein the contour / partition line recognition means has means for analyzing whether or not there is a peak value of dot number measurement array data for each of a plurality of types of predetermined intervals. 10. The image recognition apparatus according to claim 3, further comprising display means for displaying a recognition result by the contour / partition line recognition means. 11. The image recognition apparatus according to claim 10, wherein the display means has a means for displaying the image data inputted by the image data input means at the same time as the recognition result. . 12. The image recognition apparatus according to claim 11, wherein the means for displaying the image image data at the same time as the recognition result is a means for displaying the image image data in superposition with the recognition result. Image recognition device. 13. The image recognition apparatus according to claim 10, further comprising display selection means for changing display contents so as to selectively display the image image data input by the image data input means and the recognition result. An image recognition device characterized by. 14. The image recognition device according to claim 3, further comprising a recognition result confirmation means for confirming a recognition result by the contour / partition line recognition means and confirming the recognition result. An image recognition device having.