Drawing search device, drawing search method, and drawing search program

Abstract

The drawing search device uses a laser and a sensor to grasp the shape of the object, and searches and selects a drawing showing the shape of the object from a database.In this case, instead of using a camera to perform image recognition, the sensor is used to extract feature points from point cloud data, improving the accuracy of the object's shape. [Solution] The drawing search device is a device for searching for drawings from input data, and has a measurement unit that extracts multiple feature points from the input data obtained from the input device and measures the distance between each of the multiple feature points and a predetermined base point, a distance estimation unit that calculates the mutual distances between each of the feature points, and a search unit that uses the distances between the feature points calculated by the distance estimation unit to search for the most suitable drawing from the drawings stored in a database.

Description

[Technical field]

[0001] The present disclosure relates to a drawing search device, a drawing search method, and a drawing search program. [Background technology]

[0002] Advances in machine learning technology have led to advances in technology for grasping image data using AI (Artificial Intelligence) and for matching image data with selected information. On the other hand, to detect the current location from data such as drawings and maps, location information such as GPS (Global Positioning System, registered trademark) can be used with fewer errors, and there are many situations in which the accuracy of methods for detecting the current location from image data is inferior, leaving room for improvement. Furthermore, since GPS has limitations in spatial resolution, if accurate matching with drawings and maps could be achieved based on data input from cameras and other sensors, matching between the site and drawings at construction sites and other locations would be possible, and effective use of GPS is expected.

[0003] For example, Patent Document 1 discloses a technique for estimating from what position on a plan view and in what direction photographic data was taken. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6830561 Summary of the Invention [Problem to be solved by the invention]

[0005] Patent Document 1 uses machine learning to concatenate features of a drawing with features of a photograph (image data input from a camera) and then uses machine learning to match them; however, if it were possible to match input data from a sensor with a drawing using simpler information and a simpler mechanism, it would be possible to apply this to real-time processing.

[0006] The present disclosure has been proposed in consideration of the above-mentioned problems, and aims to provide a drawing search device that matches an input image with a drawing using a simple mechanism by extracting the lengths of the sides of the contour of an object, such as a structure, from input data entered from a sensor and matching it with the drawing using the proportions of each side. [Means for solving the problem]

[0007] In order to achieve the above-mentioned object, the drawing search device of the present disclosure is a device for searching for drawings from input data, and has a measurement unit that extracts a plurality of feature points from the input data acquired from an input device and measures the distance between each of the plurality of feature points and a predetermined base point, a distance estimation unit that calculates the mutual distances between each of the feature points, and a search unit that uses the distances between the feature points calculated by the distance estimation unit to search for the most suitable drawing from the drawings stored in a database.

[0008] In order to achieve the above-mentioned object, the drawing search method disclosed herein is a method for searching for drawings from input data, and includes a measurement step of extracting a plurality of feature points from the input data acquired from an input device and measuring the distance between each of the plurality of feature points and a predetermined base point, a distance estimation step of calculating the mutual distances between each of the feature points, and a search step of searching for the most suitable drawing from the drawings stored in a database using the distances between the feature points calculated in the distance estimation step.

[0009] In order to achieve the above-mentioned object, the drawing search program of the present disclosure is a program for searching for drawings from input data, and includes a measurement step of extracting a plurality of feature points from the input data acquired from an input device and measuring the distance between each of the plurality of feature points and a predetermined base point, a distance estimation step of calculating the mutual distances between each of the feature points, and a search step of searching for the most suitable drawing from the drawings stored in a database using the distances between the feature points calculated in the distance estimation step. Effect of the Invention

[0010] According to the present disclosure, light is irradiated onto an object to be searched for in a drawing, and the reflected light is acquired by an input device such as an optical sensor as input data, and drawing data that matches the input data is searched for. This makes it possible to capture the shape of the object more accurately than when image data is used as input data. Furthermore, the search method according to the present disclosure can be realized in a simple and easy manner. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram showing an example of a usage scenario of the drawing retrieval device 10. [Diagram 2] FIG. 2 is a diagram illustrating a hardware configuration of the drawing retrieval device 10. [Diagram 3] FIG. 2 is a block diagram showing basic functions of the drawing retrieval device 10. [Figure 4] 1A to 1C are diagrams illustrating an example of feature points and distances between feature points calculated by the drawing search apparatus 10. [Diagram 5] 10 is a diagram showing an example of feature points and distances between feature points calculated from a drawing selected by the drawing search device 10. FIG. [Figure 6] 10 is a diagram showing an example of feature points and distances between feature points calculated from a drawing selected by the drawing search device 10. FIG. [Figure 7] FIG. 13 is a block diagram showing basic functions of a second modified example of the drawing retrieval device 10. [Figure 8] FIG. 13 is a block diagram showing basic functions of a third modified example of the drawing retrieval device 10. [Figure 9] FIG. 1 is a diagram showing an example of a usage scenario of the drawing retrieval device 10. [Figure 10] 13 is a diagram showing an example of feature points of a ceiling surface and distances between the feature points in the third modified example of the drawing retrieval device 10. FIG. [Figure 11] 13 is a diagram showing an example of feature points of a floor surface and distances between the feature points in the third modified example of the drawing retrieval device 10. FIG. [Figure 12] 10 is a flowchart showing an example of a control process performed by the drawing retrieval device 10. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] The present disclosure will now be described with reference to the drawings.

[0013] (Scenes where the drawing search device 10 is used) Fig. 1 is a diagram showing a scene in which a drawing search device 10 is used. A structure 20 is shown as an example, but it is a target of drawing search, and may be a building such as a structure, such as a bridge, or a movable property (construction) such as a machine. Any building, structure, or movable property (construction) for which a drawing exists may be a target.

[0014] The plan search device 10 is directed toward a structure 20 for which plans are to be searched, and acquires input data through an input device such as a sensor. Plans to be searched may be floor plans of buildings, but may also be other architectural plans such as elevations, or may be plans of structures such as bridges. Plans of movable property, such as machine plans, may also be searched.

[0015] (Hardware configuration of drawing search device 10) 2 is a diagram showing a hardware configuration of the drawing search device 10. The drawing search device 10 can be configured by a general-purpose computer.

[0016] As shown in FIG. 2, the drawing search device 10 includes a processor 11, a memory 12, a storage 13, a communication IF 14 (note that IF is written as an abbreviation of Interface. The same applies below), and an input / output IF 15.

[0017] The processor 11 is hardware for executing an instruction set written in a program, and is composed of an arithmetic unit, a register, peripheral circuits, and the like.

[0018] The memory 12 is for temporarily storing programs and data to be processed by the programs, and is realized by a volatile memory such as a DRAM (Dynamic Random Access Memory).

[0019] The storage 13 is a storage device for saving data such as programs, and is realized by, for example, a flash memory, an SSD (Solid State Drive), or an HDD (Hard Disk Drive).

[0020] The communication IF 14 is an interface for transmitting and receiving signals so that the drawing search device 10 can communicate with other devices. However, the communication IF 14 is not an essential component of the drawing search device 10.

[0021] The input / output IF15 functions as an interface for an image input device such as a camera, an input device such as a sensor, and / or an input device such as a keyboard, mouse, or numeric keypad for accepting input from a user, as well as an output device such as a display for presenting information to the user.

[0022] 3 is a block diagram for explaining basic functions of the drawing search device 10. The drawing search device 10 includes a communication unit 110, a storage unit 120, and a control unit .

[0023] The communication unit 110 performs processing for communicating with other devices. For example, when a database exists outside via a network, the communication unit 110 may acquire data such as drawings via the communication unit 110. Note that the communication unit 110 is not an essential component.

[0024] The storage unit 120 stores data held by the drawing search device 10. For example, the storage unit 120 includes a drawing database 121 and stores drawings to be searched. The storage unit 120 may also store a program to be executed by the drawing search device 10.

[0025] The control unit 130 controls the operation of the drawing search device 10. Specifically, the control unit 130 includes a measurement unit 131, a distance estimation unit 132, and a search unit 133.

[0026] The measurement unit 131 extracts multiple feature points from input data acquired through the input device 30 by reflecting light from an object, and measures the distance and angle between each of the multiple feature points, using an optical sensor that detects the reflected light as a base point. For example, by using LiDAR (Light Detection And Ranging), it is possible to irradiate a laser in all directions and acquire point cloud data. By linking together these point cloud data, it is possible to detect the shape of an object that can be grasped by the sensor. Compared to extracting feature points from an image acquired by a camera, the advantage is that the shape of an object can be grasped more accurately.

[0027] The measurement unit 131 performs processing to detect the contour and shape of the object by performing contour enhancement on the acquired point cloud data using image processing technology or the like. Then, feature points are extracted from the shape of the object. When an object has a contour, feature points are points that are the end points of each straight line or curve that constitutes the contour. In other words, feature points are points where multiple straight lines intersect, points where a straight line and a curve intersect, and points where a curve intersect with another curve. These include, for example, points that form corners where a straight line or curve intersects with another straight line or curve.

[0028] After extracting the feature points as described above, the measurement unit 131 measures the distance and angle between the sensor serving as the base point and the extracted feature point.

[0029] The measurement unit 131 measures the distance and angle between the base point sensor and each feature point and acquires the data, making it possible to calculate not only the relationship between the base point and the feature points, but also the distance between the feature points.

[0030] The distance estimation unit 132 calculates the mutual distances between the feature points extracted by the measurement unit 131. Since the measurement unit 131 calculates the distance and angle from the sensor serving as the base point to each feature point, it is possible to extract the distances between the feature points using this information.

[0031] When extracting the distances between feature points, the distance estimation unit 132 does not need to calculate the distances for all combinations of feature points, but only needs to calculate the distances between adjacent feature points along the contour calculated by the measurement unit 131, in other words, the distances of the straight lines or curves that form the contour.

[0032] By calculating the distance of a straight line or a curve between feature points in distance estimation section 132, it becomes possible to grasp the general shape of the contour of the object from the distance.

[0033] First, the search unit 133 reads any drawing data from the drawing DB 121. At this time, the distance between feature points may be stored in advance as a numerical value in the drawing data, and the distance between feature points may be acquired at the same time. In addition, when the distance between feature points is not stored in the drawing data as a numerical value, the distance may be calculated from the distance between feature points drawn on the drawing, for example, by the following method.

[0034] The feature points that the search unit 133 grasps from the drawing data are points that can be captured as the end points of each straight line or curve that constitutes the drawing, similar to the processing in the measurement unit 131. In other words, points where multiple straight lines intersect, points where a straight line and a curve intersect, and points where a curve intersect are captured as feature points.

[0035] The search unit 133 selects any two adjacent feature points for one drawing read from the drawing DB 121, and calculates the distance on the drawing. In other words, the search unit 133 calculates the distance of the straight lines or curves that make up the drawing. Then, for the read drawing, the search unit 133 calculates the lengths of all the straight lines and curves that make up the drawing.

[0036] Secondly, the search unit 133 selects any one straight line or curve from the loaded drawing, and calculates the ratio between the selected straight line or curve and the adjacent straight line or curve.

[0037] Thirdly, the search unit 133 determines whether or not the ratio of adjacent straight lines or curves selected from the drawing matches the ratio of two adjacent straight lines or curves constituted by the feature points calculated by the distance estimation unit 132.

[0038] When the ratio of adjacent lines or curves selected from the drawing matches the ratio of two adjacent lines or curves calculated by distance estimation unit 132, search unit 133 further determines whether the ratio of adjacent lines or curves in the drawing matches the ratio of further adjacent lines or curves calculated by distance estimation unit 132, continues searching as long as there is a match, and finally calculates the number of matches. At this time, when determining whether the ratio of adjacent lines or curves selected from the drawing matches the ratio of two adjacent lines or curves calculated by distance estimation unit 132, a threshold may be set and an error within the threshold may be allowed.

[0039] When the ratio of adjacent straight lines or curves selected from the drawing does not match the ratio of two adjacent straight lines or curves formed by the feature points calculated by the distance estimation unit 132, the search unit 133 shifts any straight line or curve that starts the search on the drawing and sets it to a straight line or curve, and then performs a similar search.

[0040] The search unit 133 changes the line or curve that starts the search on the drawing, and determines whether it matches the ratio of the line or curve formed by the feature points calculated by the distance estimation unit 132, and the value with the highest number of matches is determined as the matching degree of the drawing.

[0041] Fourthly, the search unit 133 searches all the drawings in the drawing DB 121, and finally determines the drawing with the highest degree of matching as the drawing of the structure searched by the drawing search device 10.

[0042] The search unit 133 searches for drawings, i.e., matches drawings, using only the ratio of straight lines or curves, making it possible to search for drawings while reducing the amount of calculations, while absorbing errors in the scale of the drawings and errors in the distances detected by the sensors.

[0043] The search unit 133 searches for drawings using the ratio of straight lines or curves. Searching for drawings from three-dimensional real space tends to be a complicated problem, as it is related to the problem of how to characterize and search information in the real space. However, searching can be easily performed by simply using the straight lines or curved portions of the contour that make up the real space.

[0044] By searching the drawing using a ratio instead of the distance of a straight line or curve, the search unit 133 can eliminate the need to calculate the actual distance from the scale on the drawing, and can perform a search even if the drawing has an incorrect scale.

[0045] (Specific example of processing performed by control unit 130) Fig. 4 shows a specific example of feature points extracted by the measurement unit 131. Black circles (●) indicate feature points. Fig. 4 also shows the distance of each straight line.

[0046] Fig. 4 is a specific example assuming a scene of a room of a building, and when the drawing search device 10 acquires point cloud data using a sensor, point cloud data is generated along the walls, floor, ceiling, pillars, etc. of the room. The measurement unit 131 performs contour enhancement to enhance the straight lines or curves that constitute the walls, floor, ceiling, pillars, etc., and extracts the end points of such straight lines or curves as feature points. In Fig. 4, P101 to P112 respectively indicate feature points.

[0047] The distance estimation unit 132 calculates the distance between adjacent feature points, for example, the distance P101-P102, the distance P102-P103, and the distance P101-P112. In this calculation, since the measurement unit 131 knows the distance and angle from the sensor (drawing search device 10) to each feature point, it is possible to perform the calculation using the distance from the sensor to the feature point and trigonometric functions. In addition, the method is not limited to this, and it is also possible to calculate the distance from the sensor to each feature point.

[0048] The search unit 133 selects any one drawing from the drawing DB 121. Then, from the selected drawing, the distances of the straight lines or curves that make up the drawing are read.

[0049] When the distance of the straight lines or curves constituting the drawing is not stored as numerical data in the drawing data of the drawing DB 121, the search unit 133 calculates the distance of the straight lines or curves constituting the drawing after grasping the characteristic points.

[0050] FIG. 5 is a specific example showing a drawing and its characteristic points. Black circles (●) indicate characteristic points, and points that are the end points of lines or curves are shown as characteristic points. FIG. 5 also shows the distances of the lines or curves that make up the drawing. Specifically, the search unit 133 grasps the distances of the lines or curves, that is, the distances P201-P202, P202-P203, P203-P204, and P204-P201 by reading or calculating them, respectively.

[0051] The search unit 133 selects, for example, the drawing shown in Figure 5, and selects one arbitrary straight line or curve (P201-P202) and its adjacent straight line or curve (P202-P203). At this time, the distance ratio is 2:5.4, but this distance ratio does not exist in Figure 4. Therefore, the matching degree is 0.

[0052] Even if the search unit 133 selects P202-P203 as the line or curve first in Fig. 5 and selects P203-P204 as the adjacent line or curve, the distance ratio is 5.4:4, which does not exist in Fig. 4. Similarly, even if the line or curve selected first is P203-P204 or P203-P201, there is no line or curve with a similar distance ratio in Fig. 4. Therefore, the matching degree in Fig. 5 remains 0.

[0053] The search unit 133 then selects the drawing in FIG. 6, and the first straight line to be selected is P301-P302. If the adjacent straight lines or curves are P302-P303, the ratio of their distances is 6:2. This matches the ratios of P111-P110:P110-P109 and P102-P103:P103-P104 in FIG. 4. When the search is continued, P301-P302-P303-P304-P305-P306-P301 matches P111-P110-P109-P108-P107-P112-P111 or P102-P103-P104-P105-P106-P101-P102, and the number of matching straight lines or curves is 6. Therefore, the matching degree of FIG.

[0054] The search unit 133 further reads other drawings and searches for the ratio of the most matching straight lines or curves. If the result shows that the matching degree 6 of FIG. 6 is the highest, then the drawing of FIG. 6 is selected as the drawing representing FIG. 4.

[0055] By the above-mentioned processing, it is possible to search for a drawing such as that shown in FIG. 6 from a three-dimensional real space such as that shown in FIG. 4 with simple processing.

[0056] <Variation 1> The search unit 133 of the drawing search device 10 may be configured to select an appropriate drawing using machine learning instead of performing point-by-point matching using the proportions of straight lines or curves that constitute the drawing. When using machine learning, the search unit 133 may perform, for example, the following processing.

[0057] When machine learning is used in the search unit 133, the proportions of all straight lines or curves constituting the drawing are vectorized and used as data representing each drawing. This becomes correct answer data (learned data) for each drawing.

[0058] The search unit 133 may construct a machine learning model by learning using not only the correct answer data corresponding to the drawing, but also data of other drawings and dummy data as incorrect answer data.

[0059] When searching for a drawing, the search unit 133 may input the distance between adjacent feature points calculated by the distance estimation unit 132 into a machine learning model as input data for searching for the drawing, and determine the most suitable drawing as the drawing in the search result.

[0060] By using machine learning, the search unit 133 is able to search for drawings while reducing the effects of errors and the like.

[0061] <Variation 2> As shown in FIG. 7, the control unit 130 of the drawing search device 10 may include a location estimation unit 134.

[0062] The location estimation unit 134 estimates a location where the drawing search will be performed, and narrows down the drawings to be searched for in the search unit 133. For example, the location estimation unit 134 may acquire location information by GPS and narrow down the candidates for the drawing by referring to the location information associated with the drawing.

[0063] The location estimation unit 134 may, for example, determine whether a user who is trying to use the plan search device 10 is inside a building and is trying to search for floor plans, or whether a user is outside a building and is trying to search for elevation plans, and narrow down the plans to be searched in the search unit 133. The location estimation unit 134 may use the distance of a feature point estimated by the distance estimation unit 132 to determine whether the user is in a space within the feature point or in a space outside the feature point, and narrow down the plans to be searched in the search unit 133.

[0064] By providing the location estimation unit 134, it is possible to reduce the amount of calculation by narrowing down the number of drawings, realize high-speed drawing search, and reduce the risk of selecting the wrong drawing.

[0065] <Modification 3> As shown in FIG. 8, the control unit 130 of the drawing search device 10 may include a noise removal unit 135.

[0066] The noise removal unit 135 removes noise from the feature points estimated by the distance estimation unit 132. For example, when searching for a floor plan, the distance estimation unit 132 can extract feature points of two surfaces, the ceiling surface and the ground surface. However, due to the installation of lighting fixtures, fire alarms, and the like, feature points different from those in the drawing may be extracted from the ceiling surface, and due to the influence of installed objects, and the like, feature points different from those in the drawing may be extracted from the floor surface.

[0067] 9 shows a situation where a floor plan is to be searched for, in which there is a lighting fixture on the ceiling surface and objects, objects, or goods are installed on the floor surface. In response to this, based on the feature points extracted by the distance estimation unit 132, the noise reduction unit 135 can further extract two spaces that are closed horizontally with the ground surface and connected vertically with the ground surface, thereby extracting the ceiling surface and the floor surface.

[0068] Fig. 10 shows a specific example of feature points extracted from the ceiling surface from Fig. 9. The noise elimination unit 135 further extracts only the feature points constituting the contour shown in Fig. 10 and eliminates the feature points present inside the contour, thereby making it possible to eliminate feature points installed inside the floor plan, such as lighting fixtures. In terms of the floor surface, it becomes possible to eliminate the influence of luggage placed on the floor surface.

[0069] Fig. 11 shows a specific example of feature points extracted from the floor surface excluding the exposed floor surface or the installation location of the item from Fig. 9. The noise elimination unit 135 further extracts only the feature points constituting the outline shown in Fig. 11. Furthermore, by comparing the ceiling surface and the floor surface and tracing the common feature points, and selecting the feature points that have a larger area for the uncommon feature points, it becomes possible to extract the feature points as shown in Fig. 6 from the feature points in Fig. 10 and the feature points in Fig. 11. This makes it possible to eliminate elements that become noise, such as lighting fixtures and luggage.

[0070] For example, at a construction site, rather than building a new property, there are already occupants in the building, and various obstacles can become noise in the blueprint search. Eliminating such noise as much as possible will help improve the accuracy of the blueprint search.

[0071] (Processing flow) Hereinafter, an example of a control process in which the drawing search apparatus 10 executes a drawing search will be described with reference to FIG.

[0072] The control unit 130 of the drawing search device 10 acquires point cloud data obtained from a sensor. It applies contour enhancement processing to the data to extract the contour of the target structure, and extracts the end points of the straight lines or curves that make up the contour as feature points. It also acquires the distance and angle between the feature points and the sensor (step S101).

[0073] The control unit 130 of the drawing retrieval device 10 estimates the length of each line or curve formed by two adjacent feature points (step S102).

[0074] The control unit 130 of the drawing search device 10 selects an arbitrary drawing from the drawing DB 121 in the memory unit 120, calculates the degree of match between the distance ratio of a straight line or curve on the drawing and the distance ratio of a straight line or curve between feature points estimated by the control unit 130 of the drawing search device 10, and extracts the drawing with the highest degree of match (step S103).

[0075] (Explanation of effect) The configuration of the drawing search device 10 has been described above, but by matching drawings using the ratio of straight lines or curves of the contours that make up a structure based on point cloud data, it is possible to select drawings with a simple mechanism. Matching a three-dimensional object with a drawing often requires extensive processing, but by using the method disclosed herein, searching becomes easy.

[0076] Although the preferred embodiment of the present disclosure has been described above, the present disclosure is not limited to such a specific embodiment, and the present disclosure includes the disclosure described in the claims and its equivalents. In addition, the configurations of the devices described in the above embodiments and modifications can be appropriately combined as long as no technical contradiction occurs. [Explanation of symbols]

[0077] 10...Drawing search device, 11...Processor, 12...Memory, 13...Storage, 14...Communication IF, 15...Input / output IF, 20...Structure, 30...Input device, 110...Communication unit, 120...Memory unit, 121...Drawing DB, 130...Control unit, 131...Measuring unit, 132...Distance estimation unit, 133...Search unit, P101 to P112, P201 to P204, P301 to P306...Feature points

Claims

1. An apparatus for retrieving drawings from input data, comprising: a measurement unit that extracts a contour from input data acquired from an input device, extracts a plurality of points that are end points of straight lines or curves as feature points, and measures a distance between each of the plurality of feature points and a predetermined base point; a distance estimation unit that calculates the mutual distances between the feature points; and a search unit that searches for the most suitable drawing by comparing the ratio of the distance between the feature points calculated by the distance estimation unit with the ratio of the distance of a straight line or a curve that constitutes an object in the drawing stored in a database.

2. An apparatus for retrieving drawings from input data, comprising: a measurement unit that extracts a contour from input data acquired from an input device, extracts a plurality of points that are end points of straight lines or curves as feature points, and measures a distance between each of the plurality of feature points and a predetermined base point; a distance estimation unit that calculates the mutual distances between the feature points; and a search unit that compares the ratio of the distance between the feature points calculated by the distance estimation unit with the ratio of the distances of the straight lines or curves that constitute the object in the drawing stored in a database, and determines as a search result the drawing with the largest number of matching straight lines or curves.

3. An apparatus for retrieving drawings from input data, comprising: a measurement unit that extracts a contour from input data acquired from an input device, extracts a plurality of points that are end points of straight lines or curves as feature points, and measures a distance between each of the plurality of feature points and a predetermined base point; a distance estimation unit that calculates the mutual distances between the feature points; A drawing search device having a search unit that uses machine learning to determine drawings as search results using the distance ratio between the feature points calculated by the distance estimation unit as input data for a model constructed using the distance ratio of straight lines or curves that make up an object in a drawing stored in a database as learning data.

4. An apparatus for retrieving drawings from input data, comprising: a measurement unit that extracts a plurality of feature points of input data acquired from an input device and measures a distance between each of the plurality of feature points and a predetermined base point; a distance estimation unit that calculates the mutual distances between the feature points; a noise removal unit which extracts two closed spaces horizontal to the ground as the ceiling and floor surfaces, removes feature points other than those of the elements constituting the contours of the ceiling and floor surfaces as noise, compares the ceiling and floor surfaces, and removes non-common feature points as noise while tracing common feature points, the noise removal unit extracting the feature points of the ceiling and floor surfaces as the ground and the two closed spaces horizontal to the ground as the ceiling and floor surfaces, the noise removal unit removing non-common feature points as noise while comparing the ceiling and floor surfaces, the noise removal unit extracting the feature points of the floor and ceiling surfaces as the ground and the two closed spaces horizontal to the ground and the ceiling and floor surfaces as the ground and the noise removal unit removing the non-common feature points as noise while tracing common feature points; a search unit that searches for a most suitable drawing from drawings stored in a database using the distance between the feature points calculated by the distance estimation unit.

5. 1. A method for retrieving drawings from input data, comprising the steps of: a measurement step of extracting a contour from input data acquired from an input device, extracting a plurality of points that are end points of straight lines or curves as feature points, and measuring a distance between each of the plurality of feature points and a predetermined base point; a distance estimation step of calculating mutual distances between the feature points; A drawing search method for realizing, by a computer, a search step of searching for the most suitable drawing by comparing the ratio of the distance between the feature points calculated in the distance estimation step with the ratio of the distance of a straight line or a curve that constitutes the object in the drawing stored in a database.

6. A program for retrieving drawings from input data, a measurement step of extracting a contour from input data acquired from an input device, extracting a plurality of points that are end points of straight lines or curves as feature points, and measuring a distance between each of the plurality of feature points and a predetermined base point; a distance estimation step of calculating mutual distances between the feature points; A drawing search program for implementing on a computer a search step of searching for the most suitable drawing by comparing the ratio of the distance between the feature points calculated in the distance estimation step with the ratio of the distance of a straight line or a curve that constitutes the object in the drawing stored in a database.

Images