JP2020091817A - Drawing recognition device, drawing recognition method, and program - Google Patents

Abstract

To provide a drawing recognition device, drawing recognition method, and program, capable of easily and appropriately extracting graphic objects, such as line segments and closed areas, from a drawing image.SOLUTION: A drawing recognition device (100) is provided, comprising: drawing image input means (101) configured to receive input of a colored drawing image; and drawing line recognition means (105, 107, 109, 111, 113, 115 or 117) configured to recognize colored areas, recognize specific drawing lines based on the colored areas, and generate line segment objects corresponding to the drawing lines.SELECTED DRAWING: Figure 3A

Description

The present invention relates to a drawing recognition device, a drawing recognition method, and a program, and more particularly to a drawing recognition device, a drawing recognition method, and a program capable of easily and accurately extracting a graphic object such as a line segment or a closed region from an image of a drawing.

A drawing (hereinafter, simply referred to as drawing) provided by handwriting, printing, screen display, or projection is read as an image (hereinafter, referred to as image, image data, drawing image or drawing image data), and the dimension line and the outer circumference are drawn from the drawing image data. Various techniques have been proposed for extracting line objects such as lines and drawing lines, and graphic objects such as closed areas, and giving meanings to fittings such as fittings, walls and columns, and rooms. Since a graphic object can be handled by a CAD application or the like, productivity of design, construction, inspection, maintenance work, etc. is significantly improved by these techniques.

For example, in Patent Document 1, horizontal and vertical line segments are extracted from a drawing image, a closed region surrounded by the line segment is extracted, and a floor plan symbol and a drawing symbol are recognized from the drawing image to identify the closed region. An architectural drawing recognition device for identifying room functions is described.

As a related technique, Patent Document 2 describes an image processing device that improves the recognition accuracy of a basic line drawing such as a straight line or an arc by smoothing a contour line included in an image of a construction drawing.

JP-A-10-124547 JP-A-2000-187730 JP, 2006-195622, A

However, drawings have various specifications depending on industries and creators, and it is difficult to uniformly define rules for extracting line segments, closed regions, and the like from drawing images. In addition, when the drawings are complicated, it is often difficult to identify the line segment or the closed region to be extracted from the image. Therefore, it is desired to establish a technique for easily and accurately identifying a line segment, a closed region, or the like to be extracted from the image data of the drawing, and extracting it as a graphic object.

The present invention has been made to solve these problems, and provides a drawing recognition device, a drawing recognition method, and a program capable of easily and accurately extracting a graphic object such as a line segment or a closed region from an image of a drawing. With the goal.

A drawing recognition apparatus according to an embodiment of the present invention recognizes a drawing image input unit for inputting a colored drawing image, the colored area, and a specific drawing line based on the area. , And a drawing line recognition means for generating a line segment object corresponding to the drawing line.
In the drawing recognition device according to an embodiment of the present invention, the drawing image input means inputs the drawing images in which a plurality of kinds of drawing lines are colored with different colors for each kind, and the drawing images are drawn. The line recognition means recognizes the colored region for each color, recognizes the drawing line of a specific type based on the region, and generates a line segment object corresponding to the drawing line.
A drawing recognition apparatus according to an embodiment of the present invention recognizes an area colored with a predetermined color in the drawing image, recognizes an outer peripheral line based on the area, and detects a line segment object corresponding to the outer peripheral line. It further has a peripheral line recognition means for generating.
A drawing recognition apparatus according to an embodiment of the present invention recognizes a region colored in a predetermined color in the drawing image, recognizes a dimension line based on the region, and forms a grid line based on the dimension line. It further has a dimension line recognition means for generating.
In the drawing recognition device according to an embodiment of the present invention, the drawing line recognition means recognizes a region colored with the predetermined color, recognizes a line segment element based on the region, and the line segment A line segment object corresponding to the drawing line is generated by correcting the element according to the grid line.
In the drawing recognition device according to an embodiment of the present invention, the drawing line recognition means recognizes a region colored with the predetermined color, and generates a line segment element in the region that matches the grid line. Thus, a line segment object corresponding to the drawing line is generated.
In the drawing recognition device according to an embodiment of the present invention, the peripheral line recognition means recognizes a region colored with the predetermined color, recognizes a line segment element based on the region, and detects the line segment element. A line segment object corresponding to the outer peripheral line is generated by performing correction according to the grid line.
In the drawing recognition device according to one embodiment of the present invention, the outer peripheral line recognition means recognizes a region colored with the predetermined color, and generates a line segment element in the region that matches the grid line. Thus, a line segment object corresponding to the outer peripheral line is generated.
The drawing recognition apparatus according to an exemplary embodiment of the present invention further includes a closed area recognition unit that recognizes a closed area formed by the drawing line.
A drawing recognition method according to an embodiment of the present invention includes a drawing image input step in which a computer having a drawing image input unit and a drawing line recognition unit causes the drawing image input unit to input a colored drawing image. A drawing line recognition step in which the drawing line recognition means recognizes the colored region, recognizes a specific drawing line based on the region, and generates a line segment object corresponding to the drawing line. ..
A program according to an embodiment of the present invention is a program for causing a computer to execute the drawing recognition method.

According to the present invention, it is possible to provide a drawing recognition device, a drawing recognition method and a program capable of easily and accurately extracting a graphic object such as a line segment or a closed region from an image of a drawing.

It is an example of image data of a colored drawing. It is an example of image data of a drawing that has been subjected to preprocessing. 3 is a block diagram showing a schematic hardware configuration of the drawing recognition device 100. FIG. 1 is a block diagram showing a schematic functional configuration of a drawing recognition apparatus 100 according to a first exemplary embodiment. FIG. 4 is a block diagram showing a schematic functional configuration of a drawing recognition device 100 according to a second exemplary embodiment. It is an example of building size data. FIG. 6 is a diagram showing a processing process by a dimension line recognizing means 105. FIG. 6 is a diagram showing a processing process by a dimension line recognizing means 105. FIG. 6 is a diagram showing a processing process by a dimension line recognizing means 105. It is a figure which shows the process of the outer peripheral line recognition means 107. FIG. 7 is a diagram showing a processing process by a drawing line (opening) recognition means 109. FIG. 6 is a diagram showing a processing process by a drawing line (partition) recognition means 111. It is a figure which shows the process of the closed area|region recognition means 119. It is an example of output data. It is an example of output data. It is an example of a grid creation process. It is an example of a grid creation process. It is an example of a grid creation process. It is an example of image data of a colored drawing. It is an example of image data of a colored drawing.

<Embodiment 1>
Specific embodiments to which the present invention is applied will be described in detail with reference to the drawings. The drawing recognition apparatus 100 according to the first exemplary embodiment of the present invention obtains image data of a previously colored drawing as shown in FIG. 1A, and based on the coloring, a graphic object to be extracted (for example, a peripheral line, Objects such as openings such as windows, partitions such as walls, fittings such as doors) and their types are specified. That is, the predetermined element included in the drawing needs to be colored in advance with a predetermined color according to its type.

For example, the outer peripheral line is colored in red and yellow, the drawing line of the portion corresponding to the opening in the outer peripheral line is colored in yellow, the drawing line indicating the partition is colored in blue, and the drawing line indicating the fitting is colored in green. The coloring at this time is preferably performed so as to cover the outer peripheral line and the drawing line and to transmit the outer peripheral line and the drawing line. For example, in a paper drawing in which the outer peripheral line and the drawing line are printed in black, coloring is performed by tracing the outer peripheral line and the drawing line with yellow, blue, and green line markers. Also, the dimension line is colored in green so as to surround it. Peripheral lines, drawing lines, dimension lines and the like are typically preferably drawn in black so that they can be clearly distinguished from coloring.

FIG. 2 is a block diagram showing a schematic hardware configuration of the drawing recognition device 100. The drawing recognition apparatus 100 is an information processing apparatus including a CPU 11, a volatile memory 13, a non-volatile memory 14, an interface 15, an interface 16, a bus 20, an input/output device 70, and a photographing device 80.

The drawing recognition apparatus 100 may be a single information processing apparatus, or may be configured by a plurality of information processing apparatuses. For example, all the functions of the drawing recognition device 100 may be implemented in one mobile terminal device or the like. Alternatively, some functions may be installed in a mobile terminal device or the like, and other functions may be installed in a server computer or the like. In this case, the server computer may be provided using technologies such as cloud computing, edge computing, and fog computing.

The CPU 11 (central processing unit) reads out the program stored in the non-volatile memory 14 via the bus 20 and executes information processing according to the program to realize a specific function.

The non-volatile memory 14 is a storage device that retains a storage state regardless of the power state of the drawing recognition apparatus 100, and is, for example, a hard disk or SSD. Generally, programs and data stored in the non-volatile memory 14 are expanded in the volatile memory 13 when the program is executed.

The volatile memory 13 is a storage device that stores programs and data expanded from the non-volatile memory 14, temporary calculation data, data input or output through the input/output device 70, and the like.

The input/output device 70 includes a data output device such as a display, a data input device such as a keyboard, and a communication interface that controls communication with an external device (for example, another information processing device connected via a network). The display data output from the CPU 11 is displayed on the display via the interface 15. Instructions and data input from the keyboard are passed to the CPU 11 via the interface 15. The communication interface acquires the transmission data output by the CPU 11 through the interface 15 and outputs the transmission data to an external device. Further, the communication interface acquires the received data from the external device and transfers it to the CPU 11 via the interface 15.

The photographing device 80 is a device that photographs a drawing with a camera or the like and outputs image data. The image data is passed to the CPU 11 via the interface 16.

FIG. 3A is a block diagram showing a schematic functional configuration of the drawing recognition device 100 according to the first exemplary embodiment of the present invention. The drawing recognition device 100 includes a drawing image input unit 101, a preprocessing unit 103, a dimension line recognition unit 105, a peripheral line recognition unit 107, a drawing line (opening) recognition unit 109, a drawing line (partition) recognition unit 111, and a drawing line. It has (joint) recognition means 117, closed area recognition means 119, and output means 121.

(Enter drawing image)
The drawing image input unit 101 acquires image data of a drawing provided by handwriting, printing, screen display, projection, or the like. The drawing image input means 101 typically captures a drawing by an image capturing device 80 to obtain image data. Alternatively, the created image data may be read from the volatile memory 13, the non-volatile memory 14, or the input/output device 70.

Here, the image data needs to be colored in advance as described above. Further, when a drawing frame is drawn in the drawing, it is preferable to fit the drawing frame in the image. FIG. 1A shows an example of drawing image data read by the drawing image input means 101. In this example, the drawing is colored using four colors, color 1 to color 4, the dimension line and fittings are color 1, the outer peripheral line (excluding the opening) is color 2, the opening is color 3, and the partition is Color 4 is assumed.

(Enter building size data)
The preprocessing unit 103 acquires drawing image data from the drawing image input unit 101. In addition, building size data as shown in FIG. 4 can also be acquired together. The building size data is information indicating the maximum external dimensions of the building drawn in the drawing. In the example of FIG. 4, the building size data is given as a csv file in which the drawing file name and the maximum external dimension in the X direction and the maximum external dimension in the Y direction of the building depicted in the drawing are described in text format. There is. For example, in the 1st record of FIG. 4, in the drawing file “IMG — 0000”, the maximum external dimension of the building is width (width in the X direction)=“10060”, height (depth in the Y direction)=“6560”. It is shown. The file format and format are not limited to this example, and can be arbitrarily determined. Further, the building size data may be given by another method such as direct input from a keyboard.

(Remove shadows and noise)
The pre-processing unit 103 can perform image processing for removing shadows, noises, etc. included in the image data of the drawing. As a result, it is possible to suppress the occurrence of an error in the line segment acquisition processing, which will be described later.

-Shadow Removal The pre-processing unit 103 can remove, by image processing, a shadow caused by, for example, uneven illumination when a drawing is photographed by a camera. One of the techniques for removing shadows is to use the unevenness coefficient. According to Patent Document 3, the unevenness coefficient is a value obtained by dividing each pixel value by the average luminance value of all the pixels in the block, which is calculated in pixel block units divided into several pixel square units. By performing the binarization process while automatically changing the threshold value using the unevenness coefficient, it is possible to obtain clear drawing image data with shadows removed, without being affected by low frequency components caused by uneven lighting. You can

-Removal of noise The pre-processing unit 103 can remove noise included in the drawing data. Morphological transformation is one of the noise removal techniques. For example, by performing processing called contraction (erosion) and expansion (dilation) on a binary image after the shadow removal processing is performed, clear drawing image data from which noise is removed can be obtained.

(Trimming with picture frame)
The pre-processing means 103 can perform correction for trimming the image data with a drawing frame drawn in the drawing to adjust the shape. This makes it possible to efficiently proceed with the line segment acquisition processing, which will be described later, and the like. Since the drawing frame is a rectangle drawn on the outermost side of the drawing, the preprocessing unit 103 recognizes the drawing frame by extracting the largest closed region included in the image data. Next, trimming is performed with the recognized drawing frame as a boundary, and the region after trimming is corrected into a rectangle by a known method such as perspective transformation.

(Color conversion)
The pre-processing unit 103 performs a process of converting the color of the colored portion into a color that is easy to handle in the process using the color mask. As a result, it is possible to suppress the occurrence of an error in the line segment acquisition processing, which will be described later. The color conversion processing includes, for example, color enhancement and color clustering.

Color enhancement The preprocessing unit 103 can enhance the color of the colored portion by binarizing the lightness and the saturation, for example. The pre-processing unit 103 converts the color space of the image into a color space (YCbCr, HSV, etc.) having saturation and color difference signals as parameters, and then performs binarization processing on the color difference signals and the saturation channel. As a result, the difference between the colored portion and the other portions in the drawing image is clarified.

Color clustering Further, the pre-processing unit 103 can perform processing of absorbing variation in colors used in the image and suppressing the number of colors. This process is called color clustering, color reduction, or simplification. If you try to extract only a region colored with a specific color from an image of a colored drawing or an image obtained by subjecting the image to color binarization (enhancement) processing, the desired region may not be extracted correctly. .. This is because the RGB values may be slightly different even in a portion colored with the same color due to a change in the hue of the colored portion, color unevenness, or the like. Color clustering can be realized by a known technique such as the K-means method. For example, by performing color clustering with the number of colors K=6, colored portions can be classified into 6 colors (white, black, colors close to green, colors close to pink, colors close to orange, colors close to blue). .. The pre-processing unit 103 performs processing such as marking the clusters on the pixels classified into each cluster and replacing the color of the pixel with a representative value (average value, etc.) of the clusters.

FIG. 1B shows an example of drawing image data after the preprocessing unit 103 has performed the above-described processing of removing shadows and noises, trimming in a drawing frame, and color conversion.

(Get dimension line)
The dimension line recognizing means 105 recognizes the dimension line in the drawing image. In the present embodiment, as shown in FIG. 1B, the circumference of the dimension line is colored with color 1.

As shown in FIG. 5A, the dimension line recognizing unit 105 extracts a colored region of color 1 (region shown in white in FIG. 5A) in the image. Also, a closed area inside the extracted area is extracted. Further, as shown in FIG. 5B, a straight line component is extracted from the drawing line inside the closed region. Since the extraction of the colored area and the extraction of the closed area can be realized by various known techniques, detailed description thereof will be omitted. The extraction of the straight line component is performed by extracting a drawing line included in the closed region, calculating an approximate line segment of the drawing line, and generating a line segment object. As a well-known method used for extracting the line segment element from the image, there is LSD (Line Segment Detector) as a typical method.

As shown in FIG. 5A, when there is a pair of colored regions facing each other in the X direction or the Y direction, the lengths of the dimension lines existing in both colored regions (the lengths from the end points to the end points) are the same. There are many. However, the total length of the dimension lines that face each other may differ due to factors such as the fact that the drawing line is recognized longer than it actually is due to the influence of noise. Therefore, the dimension line recognizing means 105 can compare the total lengths of both dimension lines, and if they are different from each other, the dimension line recognition means 105 can perform correction so that the total lengths match.

For example, the dimension line recognizing unit 105 can compare the end points of both dimension lines and match both end points with whichever is inside (the one whose total length of the dimension line is shorter). FIG. 5C shows an example in which the dimension line 1 extending in the X direction and the dimension line 2 facing the dimension line 1 have different total lengths. First, there is a deviation between the end point A of the dimension line 1 and the end point B of the dimension line 2. In this case, the X coordinates of the end points A and B are unified to those of the end point B. Further, there is a deviation between the end point C of the dimension line 1 and the end point D of the dimension line 2. In this case, the X coordinates of the end points C and D are unified to those of the end point C. That is, the coordinate values are aligned to the one having the larger coordinates (B) in the end point group (A, B) having the smaller coordinates in the extension direction of the dimension line, and the one having the smaller coordinates (C) in the group having the larger coordinates (C, D). ..

(Create grid lines)
The dimension line recognition means 105 creates grid lines based on the recognized dimension lines. The grid lines are a plurality of parallel lines drawn at a predetermined pitch. The grid lines include grid lines parallel to the X axis and parallel grid lines parallel to the Y axis. Basically, in the drawings, the perimeter lines, openings, partitions, etc. of a building are arranged along this grid line.

The dimension line recognizing means 105 first generates a grid line passing on the dimension auxiliary line (see FIG. 12A) (see FIG. 12B). Next, parallel grid lines are added at a predetermined pitch in the vertical or horizontal direction. That is, grid lines (shown by broken lines in FIG. 12C) are added while changing the Y coordinate for the grid line extending in the X direction and the X coordinate for the grid line extending in the Y direction at a predetermined pitch.

The grid line pitch can be calculated from the building size data acquired by the preprocessing unit 103 and the predetermined module length. The module length may be pre-recorded in a storage area, a file, or the like (not shown), or may be acquired by the user via a keyboard or the like. For example, when the maximum outer dimension of the building in the X direction is x1 (mm), the module length is x2 (mm), and the dimension line length parallel to the X axis is x3 (pixels), the grid line pitch is x2*. (X3/x1) (pixels).

(Acquisition of peripheral line)
The outer peripheral line recognizing unit 107 recognizes an outer peripheral line in the drawing image. In the present embodiment, it is assumed that the user has previously colored the outer peripheral line drawn in the drawing with color 2 or color 3. More specifically, the opening on the outer peripheral line is colored with color 3, and the other outer peripheral lines are colored with color 2.

As shown in FIG. 6, the outer peripheral line recognizing unit 107 extracts colored regions of color 2 and color 3 in the image, and extracts a closed region inside the extracted region. Further, the vertices of the closed region (point group forming the outer edge) are extracted, and each point is moved to the grid line closest to the vertex in the colored region. An approximate line segment of the point group moved on the grid line is calculated, and a line segment object indicating the outer peripheral line is generated.

(Get drawing line (opening))
The drawing line (opening) recognition means 109 recognizes the drawing line of the opening in the drawing image. In the present embodiment, it is assumed that the user has previously colored the drawing line of the opening drawn in the drawing with color 3.

As shown in FIG. 7, the drawing line (opening) recognizing unit 109 extracts a colored region of color 3 in the image, and determines the line segment object corresponding to the drawing line included in the extracted region by the dimension line recognizing unit. It is generated by the same method as 105. Further, the drawing line (opening) recognizing unit 109 performs a correction to move the position of the line segment object to the grid line closest to the line segment in the colored region. By this correction, only the portion that matches the outer peripheral line is extracted as the drawing line (opening).

(Obtain drawing line (partition))
The drawing line (partition) recognition means 111 recognizes the drawing line of the partition in the drawing image. In the present embodiment, it is assumed that the user has previously colored the partition line drawn in the drawing with color 4.

The drawing line (partition) recognizing unit 111 extracts a colored region of color 4 in the image, and generates a line segment object corresponding to the drawing line included in the extracted region by the same method as the dimension line recognizing unit 105. .. FIG. 8 shows an example of the extracted line segment object. Further, the drawing line (partition) recognizing unit 111 performs a correction to move the position of the line segment object to a grid line closest to the line segment in the colored region. When a plurality of grid lines are included in the colored area, this correction may cause unnecessary lines such as double lines. In this case, only the line segment on any one of the grid lines is extracted as a drawing line (partition). For example, it is possible to set the priority of grid lines on the dimension auxiliary lines to be high and to extract only the line segments on the grid lines of high priority as drawing lines (partitions).

(Obtain drawing line (joint))
The drawing line (fitting) recognition means 117 generates and recognizes the drawing line of the fitting in the drawing image. In the present embodiment, it is assumed that the position of the drawing line of the fitting drawn in the drawing should be colored by the user with color 1 in advance. In addition, although the fittings have predetermined positions, they are often not drawn as drawing lines in the drawing (see FIG. 1A or 1B). Therefore, in the present embodiment, the process of generating the drawing line of the fitting is performed based on the grid line.

The drawing line (jointing tool) recognizing unit 117 first generates an image of a straight line (temporary drawing line) having a predetermined width and superimposed on the grid line. Further, a colored area of color 1 in the drawing image is extracted, a line segment object corresponding to the temporary drawing line included in the extracted area is generated by the same method as the dimension line recognizing unit 105, and the drawing line ( Joinery). Like the drawing line (partition) recognizing unit 111, the drawing line (joining tool) recognizing unit 117 may perform a correction to move the position of the line segment object to a grid line in the closest colored region. After these processes, the temporary drawing line may be discarded.

(Acquisition of closed area (room))
The closed area recognition unit 119 recognizes a room in the drawing image. As shown in FIG. 9, a closed area formed by superimposing the peripheral line created by the above-described processing, and the line segment objects of the drawing line (opening), the drawing line (partition) and the drawing line (joiner) To extract. In addition, the vertices of the closed region (a group of points forming the outer edge) are extracted, and correction is performed to move each point to the closest grid line. The area generated in this way is a closed area (room).

The output unit 121 outputs the result of the series of drawing recognition processes described above. For example, the output unit 121 assigns a unique identifier to each closed region (room) recognized by the closed region recognition unit 119, and sets the identifier and a set of coordinate values expressing each closed region (room) (typically And a set of XY coordinates of the vertices of the polygon in the closed region) and output. FIG. 10 shows an example in which this is output in a text file format. For example, in the record on the first line in FIG. 10, the coordinates of the vertices of the closed area (room) having the identifier “1.0” are (6000.0, 3000.0), (6000.0, 0.0), ( 10000.0,0.0) and (10000.0,3000.0). The file format and format are not limited to this example, and can be arbitrarily determined.

Similarly, the output unit 121 has coordinate values (typically the start point and the end point of the line segment) indicating the line segment objects of the outer peripheral line, the drawing line (opening), the drawing line (partitioning part), and the drawing line (joint). It is also possible to output the XY coordinates). As shown in FIG. 11, a text file can be output for each type of line segment object.

<Second Embodiment>
FIG. 3B is a block diagram showing a schematic functional configuration of the drawing recognition device 100 according to the second exemplary embodiment of the present invention. Compared with the first embodiment, the drawing recognition device 100 of the second embodiment is characterized in that it further includes a courtyard recognition means 113 and a drawing line (sleeve) recognition means 115. Other processing means and hardware configuration are the same as those in the first embodiment.

13 and 14 show an example of the drawing image data input by the drawing image input means 101 in the second embodiment. Similar to the first embodiment, the dimension line and the fitting are colored with color 1, the outer peripheral line (excluding the opening) is colored with 2, the opening is colored with 3, and the partition is colored with 4. In addition, in the present embodiment, the sleeve wall is colored 5 (see FIG. 13) and the courtyard is colored 6 (see FIG. 14).

(Courtyard acquisition)
The courtyard recognition means 113 recognizes the courtyard in the drawing image. In the present embodiment, it is assumed that the user has previously colored the outer periphery of the courtyard illustrated in the drawing with color 6.

The courtyard recognition unit 113 extracts a colored region of color 6 (see FIG. 14) in the image, and extracts a closed region inside the extracted region. It is assumed that the closed region extracted here is inside the outer peripheral line extracted by the outer peripheral line recognition means 107. The courtyard recognition unit 113 further extracts the vertices of the closed region (a group of points forming the outer edge) and moves each point on the grid line closest to the vertex in the colored region. The approximate line segment of the point group moved on the grid line is calculated, and the object of the polygonal area showing the courtyard is generated.

(Get drawing line (sleeve))
The drawing line (sleeve) recognition means 115 recognizes the drawing line of the sleeve wall in the drawing image. In the present embodiment, it is assumed that the user has previously colored the drawing line of the sleeve wall drawn in the drawing with color 5.

The drawing line (sleeve) recognizing unit 115 extracts a colored region of color 5 (see FIG. 13) in the image, and determines the line segment object corresponding to the drawing line included in the extracted region as the dimension line recognizing unit 105. It is generated by the same method. Further, the drawing line (sleeve) recognition means 115 performs a correction to move the position of the line segment object to the grid line closest to the line segment in the colored region. The drawing line (sleeve wall) thus generated is distinguished from other drawing lines by the point outside the outer peripheral line.

The output unit 121 can output the coordinate value indicating the closed area of the courtyard and the line segment of the drawing line (sleeve). For example, the courtyard can be output in the same format as the room of the first embodiment. At this time, the data of the courtyard can be output as a file separate from the room. Further, the drawing line (sleeve) can be output in the same format as other drawing lines.

According to the above-described embodiment, the drawing recognition device reads a drawing in which elements to be extracted such as a dimension line and a drawing line are colored in advance as an image, and extracts a drawing object by using the coloring as a clue. This makes it possible to accurately identify the line segment, the closed region, or the like to be extracted from the image, and to obtain the drawing object with high accuracy and easily.

Further, according to the above-described embodiment, the drawing recognition device employs an extraction method optimized for each type of drawing object to be extracted. By using this together with the identification method based on the above coloring, it is possible to identify more accurately from a line segment, a closed region, or the like.

Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and can be modified as appropriate without departing from the spirit of the present invention. In the present invention, within the scope of the invention, any constituent element of the embodiment can be modified or any constituent element of the embodiment can be omitted.

For example, in the above-described embodiment, the case where the number of peripheral lines included in the drawing is one is mainly assumed. However, the present invention is not limited to this, and one drawing may include two or more peripheral lines. For example, in a twin tower, a building divided by a corridor, a building having a distance, etc., the outer peripheral line may be two or more. In this case, if the outer peripheral line recognition unit 107 recognizes a plurality of closed regions, a unique identifier can be given to each closed region (outer peripheral line). Then, the output means 121 outputs an identifier indicating the outer peripheral line, a set of coordinate values indicating the area of each outer peripheral line, an identifier of a room within each outer peripheral line, and a set of coordinate values indicating the area of each room. It can be output as a set.

Further, in the above-described embodiment, the drawing lines such as the outer peripheral line, the opening and the partition are processed on the assumption that the outer peripheral line or the drawing line is drawn in advance as a visible line. However, the present invention is not limited to this, and is applicable to the case where the outer peripheral line, the opening, and the partition are not drawn as drawing lines. That is, if there is a colored portion, a temporary drawing line can be formed on a grid line and a line segment object corresponding to the temporary drawing line in the colored area can be generated, as in the drawing line (joint) recognition means 117. it can.

Further, in the above-described embodiment, the dimension line acquisition processing and the grid line creation processing have been described by taking the case where there is a pair of dimension lines facing each other in the X direction or the Y direction as an example, but the present invention is not limited to this. However, one of the opposing dimension lines may be missing. That is, if at least one of a pair of (that is, left and right) dimension lines facing in the X direction and one of a pair of (that is, upper and lower) dimension lines facing in the Y direction are present, Although the accuracy is lower than when all the dimension lines are complete, the dimension lines can be recognized and grid lines can be created.

Further, the constituent elements of the present invention can be appropriately omitted within the range not impairing the gist of the present invention, and the processing order can be changed or replaced with an equivalent. For example, various processes other than acquisition of dimension lines (necessary for grid line generation) and acquisition of outer peripheral lines (necessary for discrimination between inside and outside of outer peripheral lines) may be appropriately omitted according to the purpose. Absent.

Further, in the above-described embodiment, the example in which the drawing is drawn by hand on the paper drawing is mainly shown, but the present invention is not limited to this, and the electronic file of the drawing is colored by using the image processing application, for example. You may go. Further, in the above-described embodiment, the example in which the drawing image data is obtained by photographing the colored drawing with the camera is mainly shown, but the present invention is not limited to this, and the drawing image data is acquired by, for example, a scanner. You may.

Further, the present invention may be realized by hardware, or may be realized by the CPU executing a computer program. The computer program may be supplied to the computer by various types of non-transitory computer readable medium or transitory computer readable medium.

100 Drawing Recognition Device 11 CPU
13 Volatile Memory 14 Nonvolatile Memory 15 Interface 20 Bus 70 Input/Output Device 80 Imaging Device 101 Drawing Image Input Means 103 Pre-Processing Means 105 Dimension Line Recognition Means 107 Perimeter Line Recognition Means 109 Drawing Line (Aperture) Recognition Means 111 Drawing Lines (Partition) recognition means 113 Courtyard recognition means 115 Drawing line (sleeve) recognition means 117 Drawing line (joint) recognition means 118 Closed area recognition means 121 Output means

Claims (11)

Drawing image input means for inputting a colored drawing image,
A drawing recognition device, which recognizes the colored area, recognizes a specific drawing line based on the area, and generates a line segment object corresponding to the drawing line. The drawing image input means inputs the drawing image in which a plurality of types of drawing lines are colored with different colors for each type,
The drawing line recognizing unit recognizes the colored region for each color, recognizes a specific type of the drawing line based on the region, and generates a line segment object corresponding to the drawing line. The drawing recognition device according to Item 1. The peripheral line recognition means for recognizing a region colored with a predetermined color in the drawing image, recognizing a peripheral line based on the region, and generating a line segment object corresponding to the peripheral line. Drawing recognition device. The dimensional line recognition means for recognizing a region colored in a predetermined color in the drawing image, recognizing a dimensional line based on the region, and generating a grid line based on the dimensional line. 3. The drawing recognition device described in 3. The drawing line recognition means recognizes a region colored with the predetermined color, recognizes a line segment element based on the region, and corrects the line segment element in accordance with the grid line, The drawing recognition device according to claim 4, wherein a line segment object corresponding to the drawing line is generated. The drawing line recognition means recognizes an area colored with the predetermined color and generates a line segment element matching the grid line in the area to generate a line segment object corresponding to the drawing line. The drawing recognition device according to claim 4. The outer circumference line recognizing means recognizes the area colored with the predetermined color, recognizes a line segment element based on the area, and corrects the line segment element in accordance with the grid line to obtain the outer circumference. The drawing recognition device according to claim 4, wherein a line segment object corresponding to a line is generated. The outer peripheral line recognition means recognizes a region colored with the predetermined color and generates a line segment element matching the grid line in the region to generate a line segment object corresponding to the outer peripheral line. The drawing recognition device according to claim 4. The drawing recognition device according to claim 1, further comprising a closed area recognition unit that recognizes a closed area formed by the drawing line. A computer having drawing image input means and drawing line recognition means,
The drawing image input means, a drawing image input step of inputting a colored drawing image,
The drawing line recognition means recognizes the colored region, recognizes a specific drawing line based on the region, and generates a line segment object corresponding to the drawing line. Recognition method. A program for causing a computer to execute the drawing recognition method according to claim 10.