Crack map creation method and crack map creation processing device

Abstract

To provide a crack map creation method that makes it possible to identify the exact position and size of a crack in a concrete floor and create a map of cracks occurred to the concrete floor by a compact device composition.SOLUTION: The crack map creation method includes: a whole image generation step S4 for generating a whole image of a concrete floor by synthesizing a plurality of partial images; a crack image detection step S5 for detecting a crack image included in the whole image; a crack identification step S6 for identifying the position and size of a crack in the concrete floor on the basis of a reference dimension and the crack image; and a map creation step S7 for extracting an image of the identified crack from the whole image and superimposing the extracted image of crack on the image of a concrete floor construction drawing, thereby creating a crack map.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method for creating a crack map and a processing apparatus for crack map creation.

Background Art

[0002] As a technique of this kind, for example, Patent Document 1 detects cracks in a concrete floor by the following method. Specifically, first, an imaging device is fixed to a moving body that travels on the concrete floor, and while the moving body is traveling, the concrete floor is photographed by the imaging device. Cracks in the concrete floor are detected from the photographed image, and their width and length are further calculated.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the technique of Patent Document 1, since the moving body travels on the concrete floor, the traveling state of the moving body is greatly affected by the surface state and steps of the concrete floor. Therefore, it is difficult to specify cracks with accurate positions and sizes for the entire concrete floor. Furthermore, if a wider range is to be photographed including the edges of the concrete floor, the camera needs to be separated from the floor surface, so the moving body becomes larger.

[0005] The present invention has been made in view of such points, and provides a creation method and a processing apparatus for crack map creation that can create a crack map of a concrete floor by specifying the accurate positions and sizes of cracks in the concrete floor with a compact device configuration.

Means for Solving the Problems

[0006] In view of the above problems, a crack map creation method according to the present invention is a method for creating a map of cracks generated in a placed concrete floor, the method including: an arrangement step of arranging markers on the concrete floor to specify a reference dimension for the entire image of the concrete floor; a flight imaging step of acquiring a plurality of partial images of the concrete floor by the imaging device while flying an unmanned aerial vehicle equipped with the imaging device in a horizontal direction so that the imaging ranges of the concrete floor overlap; an entire image generation step of generating an entire image of the concrete floor by synthesizing the plurality of partial images; a crack image detection step of detecting an image of the crack included in the entire image; a crack specification step of specifying the position and size of the crack with respect to the concrete floor based on the reference dimension and the image of the crack; and a map creation step of creating the crack map by extracting the image of the specified crack from the entire image and superimposing the extracted image of the crack on an image of a construction drawing of the concrete floor.

[0007] According to the present invention, in the arrangement step, by arranging markers on the concrete floor, a reference dimension for the entire image of the concrete floor can be specified. Next, in the flight imaging step, while flying an unmanned aerial vehicle (UAV) over the concrete floor, a plurality of partial images of the concrete floor can be acquired so that the captured images of the concrete floor overlap. Some of these partial images can include an image of the marker arranged on the concrete floor, and if there are cracks in the concrete floor, an image of the cracks can also be included.

[0008] Next, in the overall image generation step, in the adjacent portions of the concrete floor, among the partially acquired images obtained adjacently, the partial images can be synthesized so that the common portions overlap, and an overall image of the concrete floor can be obtained. Next, in the crack image detection step, an image corresponding to the crack of the concrete floor is detected from the overall image of the concrete floor. In this detection, in the overall image generation step, it may be detected from the generated overall image by a detector or the like that has learned crack image detection in image processing or machine learning, or it may be detected from the partial images before synthesis by the same method, and the detected cracks may be reflected in the overall image.

[0009] Next, in the crack identification step, based on the detected crack image, the position and size of the crack on the concrete floor are identified. Therefore, in the map creation step, a crack with the correct size can be reflected in the image of the construction drawing at the correct position with respect to the image of the construction drawing. In this way, with a compact device configuration, the correct position and size of the crack on the concrete floor can be identified, and a crack map generated on the concrete floor can be created.

[0010] In a more preferred embodiment, before the overall image generation step, an image conversion step of converting a plurality of the partial images into partial ortho-images is included, and in the overall image generation step, from the plurality of the partial ortho-images, an ortho-overall image of the concrete floor is generated as the overall image of the concrete floor.

[0011] According to this embodiment, since the plurality of partial images are image-converted into non-distorted partial ortho-images, a partial image facing the imaging range of the surface of the concrete floor can be obtained. Thereby, in the overall image generation step, the ortho-overall image generated from the plurality of converted partial ortho-images becomes a non-distorted image conforming to the construction drawing of the concrete floor. For this reason, a crack map reflecting the correct position and size of the crack on the concrete floor can be created on the construction drawing.

[0012] As a more preferred embodiment, in the arranging step of arranging the markers, a plurality of the markers are arranged on two perpendicular axes on the surface of the concrete floor, and for each of the plurality of partial images captured in the aerial imaging step, a point cloud data acquisition step of acquiring three-dimensional partial point cloud data of the concrete floor corresponding to the partial image, and in the image conversion step, using the partial point cloud data, converting the partial image into the partial ortho-image, and in the entire image generation step, using the partial point cloud data, acquiring three-dimensional entire point cloud data corresponding to the ortho entire image, and the crack identification step includes a measured distance calculation step of calculating, from the three-dimensional point cloud data of the image of each of the markers included in the entire image, on each of the two axes, the distance between the markers as the measured distance, and a scale correction step of correcting the scale of the ortho entire image in the direction along the two axes so that the measured distance matches the actual distance between the markers serving as the reference dimension, and in the map creation step, overlapping the image of the crack included in the ortho entire image after scale correction with the image of the construction drawing of the concrete floor.

[0013] According to this embodiment, since the three-dimensional partial point cloud data corresponding to the partial image is used to convert it into the partial ortho-image, a more distortion-free entire ortho-image can be obtained. Here, the ortho entire image includes three-dimensional entire point cloud data. Since the three-dimensional entire point cloud data includes relative position information having a length (size), information on the actual survey length (measured length) is associated with the ortho entire image. Therefore, the measured distance calculated in the measured distance calculation step is the length between the markers along the two axes measured using the unmanned aircraft. Thus, in this embodiment, since the scale of the ortho entire image in the direction along the two axes is corrected according to the respective actual distances, the survey error (measurement error) at the time of imaging can be corrected. As a result, in the crack identification step, based on the image of the detected crack, the position and size of the crack with respect to the concrete floor can be accurately identified.

[0014] As a more preferred embodiment, the series of steps from the flight imaging step to the map creation step are repeated a plurality of times at intervals, and the crack maps created for each series of steps are overlaid to determine the progress of the cracks over time.

[0015] According to this embodiment, by repeating a series of steps a plurality of times at intervals, the progress of the cracks generated in the concrete floor can be determined over time. Thereby, it is possible to consider the necessity of repair for filling the cracks in the concrete floor.

[0016] The crack map creation processing apparatus according to the present invention is a crack map creation processing apparatus that performs processing for creating a map of cracks generated in a placed concrete floor. The processing apparatus includes a floor information registration unit in which concrete floor information including a reference dimension specified by a plurality of markers arranged on the concrete floor and an image of the construction drawing of the concrete floor is registered, and an unmanned aerial vehicle equipped with an imaging device is flown horizontally, and a partial image acquisition unit that acquires a plurality of partial images of the concrete floor by the imaging device so that the imaging ranges of the concrete floor overlap, an overall image generation unit that generates an overall image of the concrete floor by synthesizing the plurality of partial images, a crack detection unit that detects an image of the crack included in the overall image, a crack specifying unit that specifies the position and size of the crack with respect to the concrete floor based on the reference dimension and the image of the crack, and a map creation unit that creates the crack map by extracting the specified image of the crack from the overall image and overlaying the extracted image of the crack on the image of the construction drawing of the concrete floor.

[0017] According to the present invention, in the partial image acquisition unit, while flying an unmanned aerial vehicle (UAV) over the sky along the concrete floor, a plurality of partial images of the concrete floor can be acquired so that the captured images of the concrete floor overlap. Some of these partial images include images of markers arranged on the concrete floor, and when cracks occur on the concrete floor, images of the cracks are included.

[0018] Next, in the overall image generation unit, in adjacent portions of the concrete floor, the partial images can be combined with each other so that common portions overlap among the adjacent partial images that are acquired, and an overall image of the concrete floor can be obtained. Next, in the crack image detection unit, an image corresponding to the crack on the concrete floor is detected from the overall image of the concrete floor. Note that the partial image acquisition unit may be composed of a detector that has learned crack detection processing by image processing or detection of crack images in machine learning from the overall image generated by the overall image generation unit, or may detect from the partial images before synthesis by the same means and reflect the detected cracks in the overall image.

[0019] Next, in the crack identification unit, based on the detected crack image, the position and size of the crack on the concrete floor are identified, so that in the map creation unit, a crack with an accurate size can be reflected in the image of the construction drawing at an accurate position with respect to the image of the construction drawing. In this way, with a compact device configuration, the accurate position and size of the crack on the concrete floor can be identified, and a crack map generated on the concrete floor can be created.

Advantages of the Invention

[0020] According to the present invention, regardless of the surface condition of the concrete floor, etc., with a compact device configuration, the cracks on the concrete floor can be detected, and a crack map generated on the concrete floor can be created.

Brief Description of the Drawings

[0021]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0022] Hereinafter, a method for creating a crack map and a processing device for creating a crack map according to an embodiment of the present invention will be described with reference to the drawings.

[0023] (First Embodiment) FIG. 1 is a schematic diagram of a processing system including a crack map creation processing apparatus according to an embodiment of the present invention. FIG. 2 is a conceptual diagram of the processing system including the processing apparatus shown in FIG. 1. As shown in FIG. 1, the processing system 1 according to the present embodiment is a system for creating a crack map of a concrete floor F using an unmanned aerial vehicle (UAV) 20 equipped with an imaging device 21. In the present embodiment, the processing system 1 is used to confirm the position and size of cracks C generated in the concrete floor F and the progress of the cracks C after placing the concrete to create the concrete floor F.

[0024] 1. Regarding the processing system 1 The processing system 1 includes an unmanned aerial vehicle 20 equipped with an imaging device 21 and a plurality of markers M (M1 to M3) visible from the unmanned aerial vehicle 20. Further, the processing system 1 includes a processing apparatus 10 that performs a process of creating a crack map of the concrete floor F based on an image of the concrete floor F captured by the imaging device 21. In addition, in the present embodiment, the processing system 1 may include a measure with marked dimensions in addition to the marker M.

[0025] The unmanned aerial vehicle 20 is a so-called drone and is operated by a control signal from an operation device (not shown). The unmanned aerial vehicle 20 is equipped with a digital camera as the imaging device 21, and the captured image is transmitted to the processing apparatus 10 via a transmitter (not shown) and received by the processing apparatus 10 via a receiver (not shown).

[0026] 2. Regarding the hardware configuration of the processing apparatus 10 As shown in FIG. 2, the processing apparatus 10 includes, as hardware, a storage device 10A configured with a ROM, a RAM, etc., in which a check program for the partition status of the construction site 9 and the like is recorded, and an arithmetic device 10B that executes this check program.

[0027] An input device 31 and an output device 32 are connected to the processing device 10. In this embodiment, the input device 31 and the output device 32 may be an integrated touch panel display. The input device 31 receives data such as information on the concrete floor F and information on the flight control of the unmanned aircraft 20. In this embodiment, the data input by the input device 31 is stored in the storage device 10A. The output device 32 displays images captured by the imaging device 21, images of crack maps calculated by the arithmetic device 10B, and the like.

[0028] In this embodiment, the processing device 10 is composed of a storage device 10A and an arithmetic device 10B. However, for example, it may include the input device 31 and the output device 32, or may be a portable terminal such as a smartphone or a tablet in which these are integrated.

[0029] 3. Regarding the software configuration of the processing device 10 In this embodiment, as shown in FIG. 3, the processing device 10 includes, as software, a floor information registration unit 11, a partial image acquisition unit 12, an overall image generation unit 13, a crack image detection unit 14, a crack identification unit 15, a map creation unit 16, and a map update unit 17.

[0030] 3-1. Regarding the floor information registration unit 11 The floor information registration unit 11 stores concrete floor information including reference dimensions LX and LY specified by a plurality of markers (for example, three) M1 to M3 arranged on the concrete floor F, an image D1 of the construction drawing of the concrete floor F, and the like. Here, the reference dimension LX is the actual distance between marker M1 and marker M2, and the reference dimension LY is the actual distance between marker M1 and marker M3. Further, the construction drawing is a plan view of the concrete floor F, and may be, for example, a foundation laying drawing, and the positions and sizes of columns, dirt floors, or piles of the building may be described. Further, as the concrete floor information, when automatically flying the unmanned aircraft 20 described later, the latitude and longitude of each corner of the concrete floor F and the position information (geographical position information) of the plurality of markers M1 to M3 arranged on the concrete floor F may be stored.

[0031] 3-2. Regarding the partial image acquisition unit 12 The partial image acquisition unit 12 acquires a plurality of partial images GP of the concrete floor F by the imaging device 21 while flying the unmanned aircraft 20 equipped with the imaging device 21 horizontally so that the imaging ranges of the concrete floor F overlap. Specifically, as shown in FIGS. 1 and 4, the partial image acquisition unit 12 calculates the flight route R from the latitudes and longitudes of the respective corner portions of the concrete floor F, and in accordance with this flight route R, while automatically flying the unmanned aircraft 20 at an altitude at which the entire image GT can be generated and the crack C can be detected, a plurality of partial images GP are acquired. At this time, while performing the SmF process described later, a plurality of partial images GP of the concrete floor F are acquired so that the image can be converted into an ortho partial image.

[0032] 3-3. Regarding the entire image generation unit 13 As shown in FIG. 4, the entire image generation unit 13 generates the entire image GT of the concrete floor F by synthesizing a plurality of partial images GP. The entire image generation unit 13 generates the ortho entire image of the concrete floor F as the entire image GT of the concrete floor F from a plurality of converted partial ortho images. Further, the entire image generation unit 13 acquires the three-dimensional entire point group data ET corresponding to the ortho entire image using the partial point group data EP.

[0033] 3-4. Regarding the crack image detection unit 14 The crack image detection unit 14 detects an image CT of a crack C included in the overall image GT of the concrete floor F. The crack image detection unit 14 may detect the image of the crack from the generated overall image GT using a detector or the like that has learned the detection of the crack image in image processing or machine learning. In this case, since it is a generally known crack detection method, detailed description thereof is omitted. Note that the crack image detection unit 14 may detect the image CG of the detected crack C from the partial image GP before conversion into the partial ortho-image in the same manner and reflect the detected image CG of the crack C on the overall image GT. Since the converted ortho-image is likely to be a degraded image, the image of the crack C can be accurately detected by directly detecting the image of the crack C from the captured partial image.

[0034] 3-5. Regarding the crack identification unit 15 The crack identification unit 15 identifies the position and size of the crack C with respect to the concrete floor F based on the reference dimension and the image CG of the crack C. Specifically, scale correction described later may be performed on the overall image GT of the concrete floor F, and as shown in FIG. 5A, the ground sampling distance (GSD) described later may be identified. The crack identification unit 15 performs scale correction on the overall image GT (i.e., the ortho overall image) in the directions along the X-axis and Y-axis such that the measured distance LA (LB) between the points P1 to P2 (points P1 to P3) of the point cloud corresponding to the images MG (MG1 to MG3) of the markers M (M1 to M3) matches the actual distance LX (LY) between the markers M (M1 to M3) that is the reference dimension LX (LY).

[0035] 3-6. Regarding the map creation unit 16 As shown in FIGS. 6(a) and (b), the map creation unit 16 extracts the image CG of the identified crack C from the overall image GT. Next, as shown in FIG. 6(c), the map creation unit 16 creates a crack map CM by superimposing the extracted image CG of the crack C on the image BT of the construction drawing of the concrete floor F. At this time, if the position of the image BT of the construction drawing corresponding to the position of the marker M (M1 to M3) of the overall image is specified in advance, the image of the crack C can be arranged at the accurate position of the image BT of the construction drawing.

[0036] Regarding the map update unit 17 in 3-7 As shown in FIGS. 5(a) and (b), when a crack map is created at intervals by a series of steps described later, the map update unit 17 arranges the images of the cracks C detected at intervals so as to overlap with the image BT of the construction drawing. Thereby, the image CG of the crack C in the crack map CM can be overlapped, and the progress of the crack C over time can be determined. For example, when the crack C is progressing, as shown in FIGS. 5(a) to (b), since the number of pixels of the crack C is increasing, in this case, it can be determined that the crack C is progressing.

[0037] Regarding the method for creating a crack map of the concrete floor F With reference to FIG. 7 below, the method for creating a crack map of the concrete floor F will be described. First, in the placement step S1, the concrete floor information is registered in the floor information registration unit 11. Next, as shown in FIG. 1, markers M1 to M3 for specifying the reference dimensions with respect to the overall image GT of the concrete floor F are placed on the concrete floor F. In the present embodiment, a plurality of markers M1 to M3 are placed on the intersecting X-axis and Y-axis on the surface Fa of the concrete floor F. The distance LX between the markers M1 and M2 arranged on the X-axis and the distance LY between the markers M1 and M3 arranged on the Y-axis are actually measured as the reference dimensions (actual distances). By placing the markers M (M1 to M3) on the concrete floor F in the placement step S1, the reference dimensions LX and LY with respect to the overall image GT of the concrete floor F can be specified.

[0038] Next, as shown in FIG. 1, in the flight imaging step S2, the partial image acquisition unit 12 flies the unmanned aircraft 20 equipped with the imaging device 21 horizontally along the flight route R at a predetermined altitude H, and while the imaging range of the concrete floor F overlaps, the imaging device 21 acquires a plurality of partial images GP of the concrete floor F. Some of these partial images GP may include images (MG1 to MG3) of markers M (M1 to M3) arranged on the concrete floor F, and if there is a crack C on the concrete floor F, an image CG of the crack C can be included.

[0039] In the flight imaging step S2, by means of SmF (Structure from Motion) processing, for each of the plurality of partial images GP captured in the flight imaging step S2, three-dimensional partial point cloud data EP of the concrete floor F corresponding to the partial image GP is acquired. In order to perform this SmF processing, the overlapping margin of the imaging range of the concrete floor F is such that the area is about 80% of the imaging range, and the partial images GP are acquired.

[0040] Next, in the image conversion step S3, the partial image acquisition unit 12 converts a plurality of partial images GP into partial ortho-images before the entire image generation step S4. In the present embodiment, in the image conversion step S3, the partial images GP are converted into partial ortho-images using the partial point cloud data EP. Specifically, for each pixel, it is converted into a partial ortho-image so as to be a front-facing image viewed from the same height. Since the conversion of the ortho-image using SmF processing is a generally known image conversion, a detailed description thereof is omitted. In this way, since the plurality of partial images GP are image-converted into distortion-free partial ortho-images, a partial image GP facing the imaging range of the surface Fa of the concrete floor F can be obtained.

[0041] Next, as shown in FIG. 4, in the overall image generation step S4, the overall image generation unit 13 generates an overall image GT of the concrete floor F by synthesizing a plurality of partial images GP. In the overall image generation step S4, in the adjacent portions of the concrete floor F, the partial images GP are synthesized with each other so that the common portions of the adjacent partial images GP obtained adjacent to each other overlap, and an overall image G of the concrete floor F is obtained.

[0042] Here, an ortho overall image of the concrete floor F is generated as the overall image GT of the concrete floor F from a plurality of transformed partial ortho images. Further, in the overall image generation step S4, three-dimensional overall point cloud data ET corresponding to the ortho overall image is acquired using the partial point cloud data EP. The ortho overall image generated from a plurality of transformed partial ortho images is an image without distortion that conforms to the construction drawing of the concrete floor F. Therefore, a crack map reflecting the accurate position and size of the crack C on the concrete floor F can be created on the construction drawing.

[0043] Next, in the crack image detection step S5, the crack image detection unit 14 detects an image CG of the crack C included in the overall image GT of the concrete floor F. Specifically, an image CG of the crack C may be detected for the overall image GT of the concrete floor F using edge detection and binarization processing. For example, an image CG of the crack C may be detected for the overall image GT of the concrete floor F using a detector that has learned the image CG of the crack C.

[0044] Next, as shown in FIGS. 1 and 4(b), in the crack identification step S6, the crack identification unit 15 identifies the position and size of the crack C with respect to the concrete floor F based on the reference dimensions LX, LY and the image CG of the crack C. Specifically, in the measurement distance calculation step S61, the crack identification unit 15 calculates the distance LA (LB) between the markers as the measurement distance from the three-dimensional point cloud data of the image MG of each marker M (M1 to M3) included in the overall image GT on each of the X-axis and Y-axis.

[0045] Next, in the scale correction step S62, the crack identification unit 15 performs scale correction on the entire image GT (i.e., the ortho entire image) in the directions along the X-axis and Y-axis so that the measured distance LA (LB) matches the actual distance LX (LY) between the markers M (M1 to M3) where the reference dimension is LX (LY). Since the ortho entire image has size information by SfM processing, for example, if the distance LX between the markers is 20 m and it is 21 m on the ortho entire image, the X-axis direction is reduced by about 5%. The same process is performed for the Y-axis direction. Next, for example, the reference dimensions LX and LY are divided by the number of pixels to calculate the ground sampling distance (GSD). Thereby, the position and size of the crack C with respect to the concrete floor F can be identified.

[0046] In addition, for both the X-axis and Y-axis, since the marker M (M2) is 10 cm and the image MG (MG2) of the marker M is represented by 10 pixels, for example, as shown in Fig. 5(a), the ground sampling distance (GSD) may be calculated as 1 cm / pixel.

[0047] Here, the ortho entire image includes three-dimensional overall point cloud data ET. Since the three-dimensional overall point cloud data ET includes relative position information with length (size), the actual survey length (measured length) information is associated with the ortho entire image. Therefore, the measured distances LA and LB calculated in the measured distance calculation step S61 are the respective lengths between the markers M1 (M1 to M3) along the X-axis and Y-axis measured using the unmanned aircraft 20. Thus, in the scale correction step S62, since the ortho entire image in the directions along the two axes is scale-corrected according to the respective actual distances LX and LY, the survey error (measurement error) at the time of imaging can be corrected. Thereby, in the crack identification step S6, based on the detected image CG of the crack C, the position and size of the crack C with respect to the concrete floor F can be accurately identified.

[0048] Next, in the map creation step S7, as shown in FIG. 6(b), the map creation unit 16 extracts the image CG of the identified crack C from the overall image GT, and creates a crack map CM by superimposing the extracted image CG of the crack C on the image BT of the construction drawing of the concrete floor F. Specifically, in the map creation step S7, the image CG of the crack C included in the ortho overall image after scale correction is superimposed on the image BT of the construction drawing of the concrete floor F. In this way, with a compact device configuration, the exact position and size of the crack C in the concrete floor F can be identified, and the crack map CM generated in the concrete floor F can be created.

[0049] Next, as shown in FIG. 7, the steps from the aerial imaging step S2 to the map creation step S7 are taken as a series of steps, and the series of steps are performed multiple times at intervals. The map update unit 17 superimposes the images CG of the crack C in the crack map CM created for each series of steps, and determines the progress of the crack C over time. For example, when the crack C is progressing, as shown in FIGS. 5(a) to 5(b), since the number of pixels of the crack C is increasing, in this case, it is determined that the crack C is progressing.

[0050] In this way, by performing the series of steps from the aerial imaging step S2 to the map creation step S7 multiple times at intervals, the progress of the crack C generated in the concrete floor F can be determined over time. Thereby, the necessity of repair for filling the crack C in the concrete floor F can be considered.

[0051] As described above, the embodiments of the present invention have been described in detail. However, the present invention is not limited to the above-described embodiments, and various design changes can be made without departing from the spirit of the present invention described in the claims.

Explanation of Reference Numerals

[0052] 10: Processing device, 11: Floor information registration unit, 12: Partial image acquisition unit, 13: Whole image generation unit, 14: Crack image detection unit, 15: Crack identification unit, 16: Map creation unit, 17: Map update unit, 20: Unmanned aerial vehicle, 21: Imaging device

Claims

1. A method for creating a crack map that creates a map of cracks generated in a placed concrete floor, comprising: a placement step of placing markers on the concrete floor for specifying reference dimensions with respect to the entire image of the concrete floor; a flight imaging step of acquiring a plurality of partial images of the concrete floor by the imaging device such that the imaging range of the concrete floor overlaps while flying an unmanned aerial vehicle equipped with the imaging device horizontally; an entire image generation step of generating an entire image of the concrete floor by synthesizing the plurality of partial images; a crack image detection step of detecting an image of a crack included in the entire image; a crack identification step of identifying the position and size of the crack with respect to the concrete floor based on the reference dimensions and the image of the crack; a map creation step of extracting the identified image of the crack from the entire image and creating the crack map by superimposing the extracted image of the crack on an image of the construction drawing of the concrete floor; A crack map creation method characterized by including the above steps.

2. Before the entire image generation step, an image conversion step of converting a plurality of the partial images into partial ortho-images is included, In the entire image generation step, an ortho entire image of the concrete floor is generated as the entire image of the concrete floor from the plurality of partial ortho-images. The crack map creation method according to Claim 1, characterized by this.

3. In the placement step of placing markers, a plurality of the markers are placed on two perpendicular axes on the surface of the concrete floor, a point cloud data acquisition step of acquiring three-dimensional partial point cloud data of the concrete floor corresponding to the partial image for each of the plurality of partial images imaged in the flight imaging step; In the image conversion step, using the partial point cloud data, converting the partial image into the partial ortho-image; In the entire image generation step, using the partial point cloud data, acquiring three-dimensional entire point cloud data corresponding to the ortho entire image; The crack identification step includes a measured distance calculation step of calculating, from the three-dimensional point cloud data of the image of each marker included in the entire image, the distance between markers as a measured distance on each axis of the two axes; a scale correction step of correcting the scale of the entire ortho image in the direction along the two axes so that the measured distance matches the actual distance between the markers that is the reference dimension; The crack map creation method according to claim 2, wherein in the map creation step, an image of the crack included in the ortho entire image after scale correction is superimposed on an image of a construction drawing of the concrete floor.

4. A series of steps from the aerial imaging step to the map creation step is defined, and the series of steps is performed a plurality of times with an interval therebetween. The crack map creation method according to claim 1, wherein crack maps created for each of the series of steps are superimposed to determine the progress of the crack over the period.

5. A crack map creation processing device that performs processing for creating a map of cracks generated in a placed concrete floor, The processing device includes: a floor information registration unit in which floor information including a reference dimension specified by a plurality of markers arranged on the concrete floor and an image of a construction drawing of the concrete floor is registered; a partial image acquisition unit that acquires a plurality of partial images of the concrete floor by the imaging device while flying a drone equipped with the imaging device horizontally so that the imaging ranges of the concrete floor overlap; an entire image generation unit that generates an entire image of the concrete floor by synthesizing the plurality of partial images; a crack detection unit that detects an image of the crack included in the entire image; a crack specification unit that specifies the position and size of the crack with respect to the concrete floor based on the reference dimension and the image of the crack; a map creation unit that creates the crack map by extracting the specified image of the crack from the entire image and superimposing the extracted image of the crack on the image of the construction drawing of the concrete floor; A crack map creation processing device, characterized by comprising the above.

Images