KR20220057937A - Method for measuring architecture using image - Google Patents

Abstract

The present invention relates to a method for measuring a building using an image. The method for measuring a building using an image includes: a step of obtaining an image about a part of a structure in a building including a standard material by using a camera, and receiving input of the same; a step of identifying the standard material included in the inputted image and calculating a distortion degree of the image and a standard specification of the standard material in advance; a step of specifying a risk candidate position by referring to a blueprint on the structure of the building, which is stored in advance; a step of determining a risk candidate area corresponding to the specified risk candidate position in the image after matching the blueprint and the inputted image; a step of generating a correction image by reflecting the distortion degree of the image calculated in advance to the structure or a member adjacent or corresponding to the determined risk candidate area; and a step of calculating a size and deformation of the structure or the member by comparing the generated correction image and the standard specification of the standard material. According to embodiments of the present invention, the method for measuring a building using an image can measure the size of the multiple members or the structure in the single image without individual measurement on an object by initially obtaining the image of the building including the standard material.

Description

Method for measuring architecture using image}

The present invention relates to a technology for detecting a building, in particular, obtaining an image of the building, measuring the size of a specific member or structure of the building from this, and inspecting whether there is any deformation from the measured size, and if there is any deformation, to what extent It relates to a method of measuring whether there is deformation, a recording medium recording the method, and a building detection device.

In principle, architectural design and construction are consistent, but errors that inevitably occur during site surveying or building construction are permitted within the scope stipulated by the applicable building laws.

In the conventional construction and engineering supervision work, field inspections so far have been carried out by carrying a primitive measuring tool such as a tape measure, a drawing for confirmation, and, if necessary, expensive equipment such as a light breaker. However, these measurement tools have disadvantages in that a lot of time and money are wasted as they are required to be corrected or supplemented due to the inconvenience of use and a large number of measurement errors.

In order to solve this problem, a small and portable measuring tool such as a laser rangefinder was developed and started to be used. For example, a laser distance meter is a device that irradiates a laser beam on a target and measures the distance between buildings according to the reflected shape and time of the beam, and can measure a distance of several meters to several tens of meters instantaneously, It has an advantage that there is little error compared to a conventional measuring tool such as a tape measure.

On the other hand, as a considerable amount of time elapses after the building is constructed, deformation occurs in the building due to various external or internal forces. By accurately grasping such deformation and repairing it in advance, it is possible to prevent serious defects or defects in the structure of the building that may occur later. To this end, it is necessary to identify the current state by referring to the initial design drawings or information about the construction state, and various measurement tools can be utilized in this process as well.

However, despite the development of these measurement tools, the process itself of grasping errors or deformations of buildings is still not significantly different from conventional methods. For example, an operator carrying a drawing can determine whether there is an abnormality by specifying a member or structure that needs to be inspected at the site, directly measuring it using a measuring tool, and comparing it with the original design drawing again.

In the prior art literature introduced below, technical means are proposed to more easily perform the process of detecting a building using a simple measuring tool and comparing it with a drawing, but still each member or structure is manually installed by a worker. Fundamental improvement is lacking in that it has to be individually detected and recorded.

Korean Patent Laid-Open Publication No. 2005-0060227, published on June 22, 2005, "a building detection device using a laser rangefinder and a recording medium storing a program for driving the detection device"

The technical problem to be solved by the present invention is, in the detection of a conventional building, each member or structure, which is an individual detection target, is manually surveyed by using a detection means, and then the measurement result is compared again with the drawing or in an electronic drawing It eliminates the inconvenience of having to input one by one, and eliminates the limitation of the need for manual labor in the process of linking the measured values and drawings at the site even though advanced detection equipment and electronic drawings are used. In selecting an area where errors or dangers are likely to occur, we try to overcome the weakness of specifying the detection target depending on the subjective experience of the operator.

In order to solve the above technical problem, the building detection method according to an embodiment of the present invention is (a) using a camera, the building detection device has a shape in which a plurality of straight lines intersect at right angles and at least one standard is determined acquiring and receiving an image of a part of a structure of a building including the above standard materials; (b) the building detection device identifies the standard material included in the input image, and uses a shape defined by a straight line and a right angle in the identified standard material to determine the standard size of the standard material and the degree of distortion of the image calculating in advance; (c) With reference to the design drawing for the structure of the building in which the building detection device is stored in advance, at least one of bending moment and torsion according to tensile force, compressive force, shear force, or a combination thereof in the structure is specified. to do; (d) determining, in the image, a risk candidate area corresponding to the specified risk candidate location after the building detection device matches the input image with the design drawing; (e) generating a correction image by reflecting the distortion degree of the image previously calculated for the member or structure adjacent to or corresponding to the risk candidate area determined by the building detection device; and (f) calculating the size and deformation of the member or structure by comparing the corrected image generated by the building detection device with the standard standard of the standard material.

In the method for detecting a building according to an embodiment, the step (a) of acquiring and receiving an image of a part of the structure of the building includes: (a1) A plurality of straight lines set as a reference for correction for distortion of a camera lens intersect at right angles A method comprising: attaching or mounting at least one standard material having a shape to a structure and having a standard determined as a criterion for determining the size of a member or structure to be compared to a building to obtain an image together with a part of the structure of the building and receiving an input; and (a2) additionally receiving curvature information of the camera lens in addition to the input image.

In addition, in the building detection method according to an embodiment, the step (b) of pre-calculating the standard size of the standard material and the degree of distortion of the image is, (b1) image recognition or the standard material included in the input image. identifying by user input; and (b2) a shape defined by a straight line and a right angle set as a reference for correction in the identified standard material and the degree of distortion of the image based on the distortion characteristic of the lens by using the curvature information of the camera lens for each region of the image The step of calculating in advance and reading the determined reference standard of the standard material: may include.

In the method of detecting a building according to an embodiment, the step (c) of specifying a risk candidate location includes: (c1) reading a design drawing for the structure of the building stored in advance in response to the input image; and (c1) the risk that at least one of tensile force, compressive force, shear force, or bending moment or torsion according to a combination of tensile force, compressive force, shear force, or a combination thereof acts in the structure according to the direction of load transmission in consideration of the connection state of the member or structure in the read design drawing Including, but, the risk candidate position, when the direction of the force applied to the end and the central portion of the member or structure is different from the action of the force applied from both sides of the end of the member or structure When the direction is different, when the line of action of two or more forces applied to one side of the member or structure is out of line, and when the line of action of the force applied to one side of the member or structure is deviating from the direction of the load, at least one of It can be set as an action point.

In the building detection method according to an embodiment, the step (d) of determining the risk candidate area in the image is to match the design drawing with the input image based on feature lines, feature points, or a combination thereof in the image. Then, a risk candidate area corresponding to the risk candidate location specified in the design drawing may be determined in the image.

In the building detection method according to an embodiment, in the step (e) of generating a corrected image, the lens of the camera reflects the degree of distortion of the image calculated in advance for a member or structure adjacent to or corresponding to the risk candidate area. A corrected image can be created by correcting for barrel distortion, pincushion distortion, or moustache distortion of the image caused by

In the building detection method according to an embodiment, the step (f) of calculating the size and deformation of a member or structure includes: (f1) comparing the generated corrected image with the standard standard of the standard material, and deriving a size of a member or structure corresponding to or adjacent to the risk candidate area based on the; and (f2) calculating a deformed displacement value from the derived size of the member or structure with reference to the connection relationship in the design drawing.

On the other hand, below, a computer-readable recording medium in which a program for executing the above-described building detection method on a computer is recorded is provided.

Embodiments of the present invention can measure the size of a number of members or structures within one image without individual measurement of the object to be detected by first acquiring an image of a building including standard materials together, and measure the measurement result again separately Information can be integrated and managed by matching with electronic design drawings without the need to compare or input with drawings of detection may be induced.

1 is a view illustrating a regulation regarding a construction tolerance in the technical field to which embodiments of the present invention belong.
2 is a view for explaining the inclination of a vertically constructed building.
3 is a flowchart illustrating a method for detecting a building using an image according to an embodiment of the present invention.
4 is a view for explaining the idea of determining the size of another structure using an image of a building including standard materials.
5 is a diagram for explaining lens distortion of a camera.
6 to 9 are views for explaining the type of deformation according to the force acting on the building structure.
10 is a view for explaining a process of grasping the deformation shown in the building structure through the displacement of the member.
11 is a block diagram illustrating an apparatus for detecting a building using an image according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. However, detailed descriptions of well-known functions or configurations that may obscure the gist of the present invention in the following description and accompanying drawings will be omitted. In addition, throughout the specification, 'including' a certain component does not exclude other components unless otherwise stated, but means that other components may be further included.

The terms used in the present invention are only used to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as "comprises" or "comprises" are intended to designate that the described feature, number, step, operation, component, part, or combination thereof exists, but is one or more other features or It should be understood that the existence or addition of numbers, steps, operations, components, parts, or combinations thereof does not preclude the possibility of addition.

Unless specifically defined otherwise, all terms used herein, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted to have meanings consistent with the context of the related art, and are not to be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. .

1 is a view illustrating the regulations regarding building tolerance in the technical field to which embodiments of the present invention pertain, and shows that even errors that inevitably occur in the construction process can be tolerated within the range set by the Building Act.

As is widely known, in general, it is not easy to exactly match the shape and total floor area of the site with the blueprint. For districts such as housing site development projects, the numerical accuracy can be high, but in urban areas that have been neglected for a long time, the land area often does not match the actual measurement for study purposes, and this may encroach on the boundaries of surrounding buildings. In addition, errors often occurred during the design process, because in the stage of reading the accumulation of cadastral maps and transferring them to numerical values, a difference of up to 10 cm may appear depending on the thickness of the lines. During construction, when members such as bricks are not produced uniformly or depending on the construction method, there may be significant differences. Buildings can be used after receiving approval for use, but the legal status of the building is achieved by registration in the building register. need to manage

2 is a view for explaining the inclination of a vertically constructed building. Figure 2 (a) illustrates a state in which the vertical construction is correctly performed according to the design, and Figure 2 (b) illustrates a situation in which the building is tilted during construction due to incorrect construction. In particular, it is very important to manage the construction so that it can be constructed in compliance with regulations because the width of the error actually increases even if the ratio is the same as the height of the building increases.

On the other hand, as introduced above, it has been pointed out that deformation may occur in the building due to various external or internal forces as a considerable amount of time elapses even after the building is constructed. By accurately identifying such deformation and repairing it in advance, it is possible to prevent serious defects or defects in the structure of a building that may occur later or to prevent unnecessary cost increase.

Accordingly, the embodiments of the present invention presented below are devised from recognizing this problem, and while simplifying the conventional cumbersome field measurement process into a single act, it is easier to determine whether the actual measurement value matches the initial design matter in the drawing. We would like to propose a technical means to integrate measurement, input and inspection through one electronic means. Furthermore, in the inspection of the risk factors of the building, it was induced to help the judgment of the operator by selecting/determining the location that could become a danger later in the drawing in an automated manner based on the structural relationship.

3 is a flowchart illustrating a method for detecting a building using an image according to an embodiment of the present invention, and includes a series of processing steps performed by a building detecting device having at least one processor. Each of these processing processes may be implemented as a computer program including a plurality of instruction sets, loaded on the storage device of the building detection device, read out, and then performed by the processor.

In step S310, the building detection apparatus acquires and receives an image of a part of the structure of a building having a shape in which a plurality of straight lines intersect at right angles and including at least one standard material whose standard has been determined. In this process, the object to be measured (part of the structure of a building, etc.) and the standard material to be compared are taken together with the camera. At this time, as the standard material is known, it becomes a standard for judging the size or degree of deformation of the measurement target thereafter.

In addition, it is preferable to select a standard material having a shape in which a plurality of straight lines intersect at right angles in order to post-resolve image distortion caused by a camera lens. For example, a tile for finishing a wall or floor of a building or a cement block constituting a wall may be selected as a standard material. In particular, since the standard material includes a shape defined by a straight line, it can be used as a means of auxiliary distortion correction when correcting lens distortion afterward. That is, a shape defined as a straight line in a standard material may appear as a curve in the first image captured by the camera, meaning that such a distorted shape must be restored to a straight line after correction. Furthermore, since the plurality of straight lines are defined as a shape that intersects at right angles, it can help to correct distortion of the intersection angle of the straight lines thereafter. Therefore, in acquiring the first image through the camera, including a standard material including a straight line and a right angle in the image is a very important pre-work in accurately determining the size of the detection target after distortion correction.

Accordingly, in step S310, a plurality of straight lines set as a criterion for correction of distortion of a camera lens have a shape that intersects at right angles, and a standard is determined as a criterion for judging the size of a member or structure to be compared, at least one or more It is preferable to obtain and input an image together with a part of the structure of the building by attaching or mounting a standard material to the building. In addition, in order to secure more information on the degree of distortion thereafter, it is preferable to additionally receive information on the curvature of the camera lens together with the input image. A current camera lens manufacturer provides various information as a detailed specification of a lens, and double lens curvature information can be acquired together, which can be very usefully used in distortion correction.

In step S320, the building detection device identifies the standard material included in the image input through step S310, and using a shape defined by a straight line and a right angle in the identified standard material, the standard size of the standard material and The degree of distortion of the image is calculated in advance. More specifically, in this process, the standard material included in the input image is identified by image recognition or user input, and the shape and the camera defined by a straight line and a right angle set as a reference for correction in the identified standard material and the camera The degree of distortion of the image based on the distortion characteristic of the lens may be calculated in advance for each region of the image by using the curvature information of the lens, and the determined reference standard of the standard material may be read.

Earlier, since an image is acquired through a camera lens in step S310, radial distortion is accompanied, and depending on the type of lens, for example, barrel distortion, pincushion distortion, or wave Moustache distortion may occur. This distortion occurs because the magnifications of the center and the outer angle are different at the imaging position due to the difference in refractive index between the center and the outer angle of the lens.

In general, in the case of a camera lens, a method such as using a plurality of lenses or using an aspherical lens for optical correction is available, but in the technical field to which embodiments of the present invention are applied, it is difficult to adopt such optical correction. Inappropriate. Accordingly, in embodiments of the present invention, after receiving a distorted image, a software-based correction technique for the image is used.

To correct the radial distortion of the lens, check the distortion degree (distortion coefficient) by comparing the distorted image and the reference (normal image or non-distortion target), and then apply the opposite distortion that can cancel the distortion You can correct the image by doing this. Therefore, in step S320, it is necessary to calculate the degree of distortion in advance prior to correction. A more specific process of calculating the degree of distortion in the image will be described in more detail later through a mathematical model of lens distortion.

In step S330, the building detection device refers to the design drawings for the structure of the building stored in advance, and at least one of a bending moment and torsion according to a tensile force, a compressive force, a shear force, or a combination thereof in the structure. specify Now, it is necessary to determine the detection target from the initially designed drawing separately from the image. Of course, it is natural to include a member or structure corresponding to the previously inputted image as well as the detection target. Here, an operation for specifying a fine position to be measured particularly precisely among these members or structures is performed.

Embodiments of the present invention are implemented to specify a fine position in the structure where a risk or error is likely to occur in the future in consideration of the mechanical relationship of the member or structure shown in the design drawings. For this, various external forces and stresses applied to the building were considered, and various cases appearing vertically and horizontally were collected and classified. As specified in step S330, tensile force, which is a pulling force within the structure, compressive force, which is a pressing force, shear force that slips away from each other, bending that occurs when tensile force acts on one side and compressive force acts on the other side together. As such, it is possible to automatically specify in the design drawings where deformations appear in these structures, such as torsions appearing. If the direction of at least one force applied to one structure is defined in advance in consideration of the correlation between structured objects, it is possible to combine It is possible to pre-determine and categorize risk candidate positions where distortion/error/destruction may appear. Therefore, it is preferable to add and store the action point and action direction of the force for each member or structure in a conventional design drawing in advance.

In step S340, the building detection device matches the image input in step S310 and the design drawing stored in advance, and then determines a risk candidate area corresponding to the risk candidate location specified in step S330 in the image. do. More specifically, in this process, after matching the input image with the design drawing based on feature lines, feature points, or a combination thereof in the image, a risk candidate area corresponding to the specified risk candidate location in the design drawing can be determined within the image.

This process is a process of correlating the drawing and the corresponding configuration in the actually captured image, and only then matches the dangerous candidate location where distortion/error/destruction in the electronic drawing may appear to the area in the image representing the current construction state. will make it In order to match drawings and photos from an implementation point of view, it is necessary to set them based on a specific area or structure of both. It is also possible to set the matching criteria by using (tag). Through this, the conventional technology fundamentally solves the inconvenience of having to prepare or describe/input each member of the on-site actual measurement value and the printed physical drawing or electronic drawing, and convert the electronic image (digital image) directly into the electronic drawing ( In addition to being able to link directly to a CAD file, for example, it is possible to simultaneously specify risk candidate locations in an automated way.

In step S350, the building detection apparatus generates a corrected image by reflecting the distortion degree of the image previously calculated for the member or structure corresponding to or adjacent to the risk candidate area determined in step S340.

Although the detection target was specified as an image from the design drawing in step S340, distortion is still contained in the image. Therefore, before measuring the object and examining whether there is distortion or error, it is necessary to first resolve the distortion of the image. To this end, the degree of distortion of the image calculated through step S320 is used. The degree of distortion of the image may appear due to the characteristics of the lens that mutually mediates the 3D spatial coordinates where the real object is located and the 2D spatial coordinates obtained through the camera, and various methodologies for correcting this may be used. In the embodiments of the present invention, distorted values are derived for each pixel region constituting an image through image analysis among these methodologies, and the actual position of an object existing in reality is adjusted. To this end, in step S350, by reflecting the degree of distortion of the image calculated in advance for the member or structure corresponding to or adjacent to the risk candidate area, barrel distortion, pincushion distortion, Alternatively, a correction image can be created by correcting the moustache distortion. From an implementation point of view, embodiments of the present invention may specify a type of distortion through a bending direction of a straight line defined in a standard material, and may remove that type of distortion through a degree of bending.

First, barrel distortion refers to a phenomenon in which the edge of an image is curved outward, and the direction in which a straight line in the acquired image is curved appears as a larger convex shape at the edge than at the center. Second, in the pincushion distortion, the direction of straight line bending in the opposite direction to the barrel distortion appears in a larger concave shape at the edge than at the center. Third, wave distortion is a phenomenon in which barrel distortion and pincushion distortion occur together in one straight line. In this case, an image processing technique for flattening a curved straight line may be used to correct image distortion. Accordingly, it is possible to provide more accurate image shape and size information by correcting image distortion through the direction and degree of bending of the straight line. In addition, by using the vertical shape defined in the standard material, it is possible to compensate in consideration of the degree of distortion of the intersecting straight lines.

Hereinafter, with respect to a more specific distortion correction technique, a mathematical model of lens system distortion will be described.

In general, a mathematical model of lens distortion is defined in a normalized image plane from which the influence of camera internal parameters is removed. Assuming that there is no lens system distortion, a point (X _c , Y _c , Z _c ) in three-dimensional space is a point (x _{n _u} , y _{n_u} ) on the normal image plane by central projection (pinhole projection). ) and expressed as Equation 1 below (here, subscript n: normalized, u: undistorted).

However, in reality (x _{n _u} , y _{n _u} ) becomes distortion (mainly radial distortion) due to the nonlinearity of the lens system. Assuming that (x _{n _d} , y _{n _d} ) is a normalized coordinate in which the lens system distortion is reflected, the lens system distortion model is the following Equation 2 (here, the subscript d: distorted).

이다.step,

am.

In Equation 2, the first term on the right side represents radial distortion, and the second term represents tangential distortion (here, k ₁ , k ₂ , k ₃ are radial distortion coefficients, p ₁ , p ₂ is the tangential distortion coefficient). r _u represents the distance (radius) to the principal point when there is no distortion.

In this case, (x _{n _d} , y _{n _d} ) is the coordinates in the regular image plane, and the actual image pixel coordinates (x _{p_d} , y _{p _d} ) can be obtained by the following Equation 3 by reflecting the camera internal parameters.

That is, it can be arranged as in Equation 4 below.

Here, the variables of Equation 4 are intrinsic parameters of the camera, f _x , f _y are focal lengths, c _x , c _y are the lens center image coordinates (principal point), and skew_c is the image sensor's It is a skew coefficient indicating the degree of inclination of the y-axis of the cell array.

Now, let's create a corrected image from the distorted image. Before correcting the distorted image, it is preferable to first derive an intrinsic parameter through camera calibration. Mathematically, camera calibration is a process of finding a transformation relationship between three-dimensional spatial coordinates and two-dimensional image coordinates or a parameter that explains this transformation relationship, and is a process of finding a camera coordinate system, a normal image plane of a normal coordinate system, and a pixel coordinate system. It handles geometric relationships between image planes.

The distorted image was named I _d and corrected I _u . The basic idea of correction is to fill each pixel value in I _u with the corresponding pixel value in I _d when the pixel coordinates are distorted. If a point in the image I _u is (x _{p_u} , y _{p _u} ), it is converted into normalized coordinates (x _{n_u} , y _{n_u} ) by inversely applying camera parameters.

That is, it is expressed as Equation 6 below.

Next, the distance to the center r _u (r _u ² = x _{n _u} ² + y _{n _u} ² ) is obtained, and the distortion model of Equation 2 is applied to obtain the distorted coordinates (x _{n_d} , y _{n_d} ).

Finally, if (x _{n _d} , y _{n _d} ) is converted back to the pixel coordinate system, the coordinates (x _{p_d} , y _{p_d} ) in the distorted image of (x _{p_u} , y _{p_u} ) can be obtained as in Equation 8 below. there is.

Again, returning to FIG. 3, in step S360, the building detection device calculates the size and deformation of the member or structure by comparing the reference standard of the standard material with the corrected image generated through step S350. Now, if the distortion had been removed from the image, the standard material that was included in the original image would have been correctly restored to straight and square as defined. Based on these standard materials, the size of a member or structure specified as a target can be detected, and deformation or errors of the member or structure can be derived with reference to the design drawing.

4 is a diagram for explaining the idea of determining the size of another structure using a building image containing a standard material, illustrating an actual example in the field performed in the process of devising the embodiments of the present invention.

In the building detection method according to an embodiment of the present invention, a plurality of straight lines set as a criterion for correction of camera lens distortion have a shape that intersects at right angles, and as a criterion for judging the size of a member or structure to be compared It has been described that by attaching or mounting at least one standard material of which the standard has been determined to a building, an image is obtained along with a part of the structure of the building and input is received. Referring to FIG. 4 , a cement block is included in one image as a standard material and taken. Each image exemplifies after the distortion has already been corrected.

Among the exemplified images, the original image of FIG. 4(a) was photographed from the front and had little distortion, but the original image of FIG. 4(b) was photographed at an angle and thus the distortion was more severe. In both cases, although there is a difference in degree, it was inappropriate to measure the size of the target object (member indicated by hatching) from the original image using a standard material (cement block). Therefore, through the image of FIG. 4 in which the distortion is resolved by applying the embodiment of the present invention, it was possible to measure the height of the target object buried under the ground in two situations, and found a discrepancy with the original design drawing could take action.

5 is a diagram for explaining lens distortion of a camera.

Lens distortion largely includes radial distortion and tangential distortion. Radial distortion is caused by the refractive index of the lens, and as illustrated in FIG. 5 , the degree of image distortion refers to distortion that is differentially determined according to the separation distance from the center. On the other hand, tangential distortion is a distortion that occurs because the camera lens and the image sensor (CCD, CMOS, etc.) are not level or the center of the lens itself is not aligned during the camera manufacturing (assembly) process. Distortion distribution is different in the form.

Such information on lens distortion may be obtained by analyzing the characteristics of an object existing in an image, or may be directly provided by a lens manufacturer. In the embodiments of the present invention, the degree of distortion was identified by using standard materials (vertical and horizontal) specifically defined in the image, and additional information regarding the degree of distortion of the lens was secured and utilized together.

6 to 9 are views for explaining the type of deformation according to the force acting on the building structure. In the embodiments of the present invention, in specifying a risk candidate location, a design drawing of a structure of a building stored in advance is read in response to an input image, and the load is calculated in consideration of the connection state of a member or structure in the read design drawing. According to the transmission direction, it is possible to specify a risk candidate position in the structure where at least one of a bending moment and a torsion according to a tensile force, a compressive force, a shear force, or a combination thereof acts. In particular, such a risk candidate position is, when the direction of action of the force applied to the end and the central portion of the member or structure is different, when the direction of action of the force applied from both ends of the member or structure is different, on one side of the member or structure In at least one of a case in which the line of action of two or more applied forces is deviated and the line of action of a force applied to one side of the member or structure is deviating from the load direction, it is preferable to set the point of action of the force. The types of deformation according to the force acting on the building structure described above will be sequentially examined as follows.

6 is a view for explaining a case in which the direction of action of the force applied to the end and the central portion of the member or structure is different. 6, the force applied to the building from the outside, such as the load of the building itself (self-weight), the weight of an object loaded on the building (loading load), temporarily applied to the side of the building (wind pressure, seismic force), and When the force supporting the building collides, it exemplifies a situation in which a member or part of the structure is bent or broken due to the bending moment. In the case of (a) and (b) of Figure 6, the center is distorted and bent (a) or broken ( b) becomes Alternatively, as shown in (c) of FIG. 6 , in a cantilever-shaped structure, a phenomenon in which the corresponding structure is bent by a force applied to one end of the structure may appear. The member to which such a force is applied includes, for example, a beam or a slab. For the structure shown in FIG. 6 , the risk candidate position may be set to, for example, the central portion of the member.

7 is a view for explaining a case in which the direction of action of the force applied from both sides of the end of the member or structure is different. Referring to FIG. 7 , the member is crushed by the compressive force applied from both sides by the force moving in one axial direction (a), or the member is stretched or severed by the tensile force acting from the pulling force from both sides The losing situation (b) was exemplified. The member to which such a force is applied includes, for example, a column or a load-bearing wall. For the structure shown in FIG. 7 , the risk candidate position may be set to, for example, the central portion of the member.

8 is a view for explaining a case where the line of action of two or more forces applied to one side of a member or structure is misaligned. If the lines of action of two or more forces coincide, compression or tension, not shear, will occur. Referring to FIG. 8 , when the line of action is misaligned (a), a shear force is applied, which is cut (b) or pushed (c) therebetween. The member to which such a force is applied includes, for example, a shear wall in which the member is destroyed by an earthquake or the like. For the structure shown in FIG. 8 , the risk candidate position may be set, for example, between lines of action applied to the member.

9 is a view for explaining a case where an action line of a force applied to one side of a member or a structure deviates from a load direction. Referring to FIG. 9 , when a load is applied to a structure extending from one side of a beam connecting between vertical structures in a position/direction different from the stress of the structure, a situation in which torsion occurs is exemplified. For the structure shown in FIG. 9 , the risk candidate position may be set, for example, to a position at which an action line is applied to a member (eg, a point of contact with a member stacked on the member).

10 is a view for explaining a process of grasping the deformation shown in the building structure through the displacement of the member.

In the embodiments of the present invention, the process of calculating the size and deformation of a member or structure compares the generated correction image with the reference standard of the standard material, and based on the reference standard, a member corresponding to or adjacent to a risk candidate area or Derive the size of the structure. Then, with reference to the connection relationship in the design drawing, the deformed displacement value may be calculated from the derived size of the member or structure. Referring to FIG. 10, based on this structural analysis, it was displayed what kind of force was applied to the member to be detected, such as compressive force or tensile force, and it was confirmed that deformation or error occurred due to this, and the displacement value could also be calculated.

11 is a block diagram illustrating an apparatus 1100 for detecting a building using an image according to another embodiment of the present invention, wherein each execution step of the method for detecting a building of FIG. 3 is reconstructed from the viewpoint of hardware configuration. Therefore, in order to avoid duplication of description, only an overview of the operations and functions of each configuration will be briefly summarized herein.

The input unit 10 uses a camera to obtain an image of a part of the structure of a building in which the building detection device has a shape in which a plurality of straight lines intersect at right angles and includes at least one standard material for which the standard is determined. .

The storage unit 20 is a configuration for storing standard specifications of standard materials and electronic design drawings. In particular, in the case of design drawings, the action point/line of action and the direction of the corresponding force through structural analysis in which members or structures are connected/loaded. It is preferable that these are added and stored together.

The processing unit 30 identifies the standard material included in the image input through the input unit 10, and uses a shape defined by a straight line and a right angle from the identified standard material to the standard specification and image of the standard material. Calculate the degree of distortion in advance and refer to the design drawings for the structure of the building stored in advance in the storage unit 20. In the structure, at least one of tensile force, compressive force, shear force, or a bending moment or torsion according to a combination thereof acts After specifying a risk candidate location to be used, and matching the input image with the design drawing, a risk candidate area corresponding to the specified risk candidate location is determined in the image, and corresponding to or adjacent to the determined risk candidate area A correction image is generated by reflecting the degree of distortion of the image calculated in advance for a member or structure, and the size and deformation of the member or structure are calculated by comparing the generated correction image with a reference standard of the standard material.

According to the above-described embodiments of the present invention, it is possible to measure the size of a plurality of members or structures within one image without individual measurement of the detection target by acquiring the building image including the standard material together at first, and the measurement Information can be integrated and managed by matching the result with the electronic design drawing without the need to compare or input the result with a separate drawing again. Detection can be induced by specifying automatically.

Meanwhile, embodiments of the present invention can be implemented as computer-readable codes on a computer-readable recording medium. The computer-readable recording medium includes all types of recording devices in which data readable by a computer system is stored.

Examples of the computer-readable recording medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. In addition, the computer-readable recording medium is distributed in a computer system connected to a network, so that the computer-readable code can be stored and executed in a distributed manner. And functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the technical field to which the present invention pertains.

In the above, various embodiments of the present invention have been mainly reviewed. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention can be implemented in modified forms without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments are to be considered in an illustrative rather than a restrictive sense. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

1100: building detection device
10: input
20: storage
30: processing unit

Claims (7)

(a) using a camera, the building detection device having a shape in which a plurality of straight lines intersect at right angles and receiving an image of a part of the structure of the building including at least one standard material for which the standard is determined;
(b) identifying the standard material included in the image to which the building detection device is input, and using the shape defined by a straight line and a right angle in the identified standard material, the standard size of the standard material and the degree of distortion of the image calculating in advance;
(c) With reference to the design drawing for the structure of the building in which the building detection device is stored in advance, at least one of a bending moment and torsion according to a tensile force, a compressive force, a shear force, or a combination thereof in the structure is specified. to do;
(d) determining, in the image, a risk candidate area corresponding to the specified risk candidate location after the building detection device matches the input image with the design drawing;
(e) generating a corrected image by reflecting the distortion degree of the image previously calculated for the member or structure adjacent to or corresponding to the risk candidate area determined by the building detection device; and
(f) calculating the size and deformation of the member or structure by comparing the corrected image generated by the building detection device with the reference standard of the standard material; Containing, building detection method. The method of claim 1,
The step (a) is,
(a1) At least one standard material having a shape in which a plurality of straight lines intersect at right angles set as a criterion for correction of distortion of a camera lens, and whose standard is determined as a criterion for judging the size of a member or structure to be compared Including; and receiving an image along with a part of the structure of the building by attaching or mounting it to the building.
(a2) further comprising a; further comprising the step of receiving the curvature information of the camera lens in addition to the input image; 3. The method of claim 2,
The step (b) is,
(b1) identifying the standard material included in the input image by image recognition or user input; and
(b2) In the identified standard material, the degree of distortion of the image based on the distortion characteristics of the lens using the shape defined by a straight line and a right angle set as a criterion for correction and curvature information of the camera lens for each region of the image in advance Calculating and reading the determined reference standard of the standard material: Containing, building detection method. The method of claim 1,
Step (c) is,
(c1) reading a design drawing for the structure of the building stored in advance in response to the input image; and
(c1) A risk candidate in which at least one of tensile force, compressive force, shear force, or bending moment or torsion according to a combination of tensile force, compressive force, shear force, or a combination thereof acts in the structure according to the direction of load transmission considering the connection state of the member or structure in the read design drawing specifying a location; including, but
The risk candidate location is
When the direction of action of the force applied to the distal end and the central portion of the member or structure is different, when the direction of action of the force applied from both ends of the member or structure is different, two or more forces applied to one side of the member or structure When the line of action of the deviates, the line of action of the force applied to one side of the member or structure is set as the point of action of the force in at least one of the cases where the line of action of the force deviates from the direction of the load, the building inspection method. The method of claim 1,
The step (d) is,
After matching the input image with the design drawing based on feature lines, feature points, or a combination thereof in the image, determining a risk candidate area corresponding to the risk candidate location specified in the design drawing in the image, Building inspection methods. The method of claim 1,
Step (e) is,
Barrel distortion, pincushion distortion, or wave distortion of the image by the lens of the camera by reflecting the degree of distortion of the image calculated in advance for the member or structure corresponding to or adjacent to the risk candidate area A method of detecting a building that produces a corrected image by correcting (moustache distortion). The method of claim 1,
The step (f) is,
(f1) comparing the generated correction image with a reference standard of the standard material, and deriving the size of a member or structure corresponding to or adjacent to the risk candidate area based on the reference standard; and
(f2) with reference to the connection relationship in the design drawing, calculating a displacement value deformed from the derived size of the member or structure; Containing, building detection method.

Images