Deterioration assessment device, deterioration assessment method, and program
The deterioration evaluation device and method address the challenge of evaluating degraded areas by detecting equipment and deteriorated regions and estimating pixel resolution, enabling precise size assessment regardless of imaging conditions.
Patent Information
- Application Number
- JP2022576912
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-01-22
- Publication Date
- 2024-11-14
- Estimated Expiration
- 2041-01-22
AI Technical Summary
Conventional techniques for quantitatively evaluating degraded areas in structures require specific imaging conditions, making it difficult to accurately assess deterioration without these conditions.
A deterioration evaluation device and method that includes equipment area and deteriorated area detection, along with pixel resolution estimation, allowing for the estimation of actual size values of degraded areas based on image analysis.
Enables accurate evaluation of deteriorated areas without being limited to certain imaging conditions, providing a unified pixel resolution map for precise estimation of actual size values.
Smart Images

Figure 0007583300000003 
Figure 0007583300000004 
Figure 0007583300000005
Abstract
Description
[Technical field]
[0001] The present disclosure relates to a degradation assessment device, a degradation assessment method, and a program. [Background technology]
[0002] There are methods that use image processing to detect deterioration that occurs on the surface of a structure. In recent years, segmentation methods in deep learning have been used. Known techniques for quantitatively evaluating deteriorated areas include a technique that includes a crack scale in an image and quantitatively evaluates the deteriorated area from the scale (see Non-Patent Document 1), and a technique that acquires images using photographic equipment with a mechanism for maintaining a constant distance from the subject and quantitatively evaluates the deteriorated area from pixel resolution (see Non-Patent Document 2). [Prior art documents] [Non-patent literature]
[0003] [Non-Patent Document 1] Yusuke Fujita and 2 others, "Image synthesis and semi-automatic crack evaluation for visual inspection of concrete structure surfaces," Journal of the Japan Society of Civil Engineers, Vol. 74, No. 1, pp. 18-32, 2018. [Non-Patent Document 2] Kenichi Handa and 5 others, "Evaluation of Road Tunnel Integrity by Crack Extraction Using Digital Image Processing," Proceedings of the 37th Symposium on Rock Mechanics, January 2008. Summary of the Invention [Problem to be solved by the invention]
[0004] Conventional techniques for quantitatively evaluating degraded areas require some kind of modification to the captured image, which means that it is difficult to quantitatively evaluate degraded areas using image processing algorithms such as segmentation methods for images captured without satisfying certain shooting conditions, leaving room for improvement.
[0005] In view of the above circumstances, an object of the present disclosure is to provide a degradation assessment device, a degradation assessment method, and a program that are capable of assessing the actual size of a degraded area without being limited to certain shooting conditions. [Means for solving the problem]
[0006] In order to solve the above-mentioned problems, the deterioration assessment device of the present disclosure includes an equipment area detection unit that detects an equipment area from an acquired image, a deteriorated area detection unit that detects a deteriorated area from the image, a pixel resolution estimation unit that estimates the pixel resolution of at least a portion of the image based on the detected equipment area, and a deterioration size estimation unit that estimates the size of the deteriorated area based on the pixel resolution.
[0007] In addition, the deterioration assessment method of the present disclosure is a deterioration assessment method for assessing deterioration using a deterioration assessment device, and includes the steps of detecting an equipment area from an acquired image, detecting a deteriorated area from the image, estimating the pixel resolution of at least a portion of the image based on the detected equipment area, and estimating the size of the deteriorated area based on the pixel resolution.
[0008] In addition, a program according to the present disclosure causes a computer to function as the degradation assessment device according to the present disclosure. Effect of the Invention
[0009] According to the present disclosure, it is possible to evaluate the actual size of a degraded area without being limited to certain shooting conditions. [Brief description of the drawings]
[0010] [Figure 1] FIG. 1 is a diagram for explaining a degradation assessment system according to an embodiment of the present disclosure. [Diagram 2] FIG. 1 is a simplified diagram showing a tunnel. [Diagram 3] 1 is a block diagram showing a configuration of a degradation assessment device according to an embodiment of the present disclosure. [Figure 4A]FIG. 4 is a diagram showing an example of an image acquired by the degradation assessment device. [Figure 4B] FIG. 4 is a diagram showing an example of an image acquired by the degradation assessment device. [Diagram 5] 2 is a block diagram showing a configuration of a pixel resolution estimation unit according to an embodiment of the present disclosure. FIG. [Figure 6] 13 is a diagram showing an example of an image when the number of objects N is determined to be N=1. FIG. [Figure 7A] 13 is a diagram showing another example of an image when the number of objects N is determined to be N=1. FIG. [Figure 7B] 13 is a diagram showing another example of an image when the number of objects N is determined to be N=1. FIG. [Figure 8] FIG. 13 is a diagram for explaining a pixel resolution map. [Figure 9] FIG. 13 is a diagram showing an example of an image when the number of objects N is determined to be N=2. [Figure 10A] FIG. 13 is a diagram showing another example of an image when the number of objects N is determined to be N=2. [Figure 10B] FIG. 13 is a diagram showing another example of an image when the number of objects N is determined to be N=2. [Figure 11A] 13 is a diagram showing an example of an image when it is determined that the number of objects N is N≧3. [Figure 11B] 13 is a diagram showing another example of an image when the number of objects N is determined to be N≧3. [Figure 12] 4 is a block diagram showing a configuration of a degradation magnitude estimation unit according to an embodiment of the present disclosure. FIG. [Figure 13A] FIG. 11 is a diagram for explaining an example of coordinates of a deteriorated region. [Figure 13B] FIG. 13 is a diagram for explaining a state in which the coordinates of a degraded area are plotted on a pixel resolution map. [Figure 14A] FIG. 2 is a diagram showing the operation of the degradation assessment system according to an embodiment of the present disclosure. [Figure 14B] FIG. 2 is a diagram showing the operation of the degradation assessment system according to an embodiment of the present disclosure. [Figure 15A]FIG. 4 is a diagram illustrating an operation of a control unit of the degradation assessment device according to an embodiment of the present disclosure. [Figure 15B] FIG. 4 is a diagram illustrating an operation of a control unit of the degradation assessment device according to an embodiment of the present disclosure. [Figure 16A] 13 is a diagram for explaining an equipment area correction unit according to the first modified example. FIG. [Figure 16B] 13 is a diagram for explaining an equipment area correction unit according to the first modified example. FIG. [Figure 17A] 13 is a diagram for explaining an equipment area correction unit according to the first modified example. FIG. [Figure 17B] 13 is a diagram for explaining an equipment area correction unit according to the first modified example. FIG. [Figure 18] 13 is a diagram for explaining an equipment area correction unit according to the first modified example. FIG. [Figure 19] FIG. 11 is a diagram showing one example of an image acquired by a degradation assessment device according to Modification 2. [Figure 20] FIG. 11 is a diagram for explaining a pixel resolution map according to Modification 2. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0011] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings as appropriate. In each drawing, the same or corresponding parts are given the same reference numerals. In the description of the present embodiment, the description of the same or corresponding parts will be omitted or simplified as appropriate. The embodiment described below is an example of the configuration of the present disclosure, and the present disclosure is not limited to the following embodiment.
[0012] <Configuration of Deterioration Assessment System> An example of the configuration of a degradation assessment system 1 according to this embodiment will be described with reference to FIG.
[0013] The deterioration assessment system 1 is a system that quantitatively assesses the deterioration occurring in the main body of a structure by using known information about the equipment from a photographed image of the structure. In the deterioration assessment system 1, an image processing algorithm is used to detect deteriorated areas and known equipment from the photographed image. Next, the known dimensions of the detected equipment are compared with the number of pixels on the image to estimate the pixel resolution at each position. Then, the actual size of the deteriorated area is calculated based on the estimated pixel resolution. This makes it possible to estimate the actual size of the deteriorated area without being limited to certain photographing conditions.
[0014] In this embodiment, the structure refers to a cable tunnel. A cable tunnel is an underground tunnel for laying communication cables. The structure is not limited to this and includes, for example, tunnels for laying gas pipes or power lines, manholes, etc. In this embodiment, the equipment refers to reinforcing steel that is part of a metal equipment. The deterioration to be detected is damage, dirt, etc. that has occurred on the main body of the cable tunnel in the captured image. The deterioration may also be damage, dirt, etc. that has occurred on the equipment.
[0015] FIG. 2 is a simplified diagram of a tunnel. The height in the Z-axis direction from the top to the bottom of the tunnel and the width in the X-axis direction are, for example, about 3 meters. The tunnel may have a circular cross-sectional shape as shown in FIG. 2, or a rectangular cross-sectional shape. Workers enter the tunnel to perform work such as laying, connecting, maintaining, repairing, and removing the communication cable. The tunnel is equipped with hardware equipment for accommodating the communication cable, lighting, ventilation equipment, drainage equipment, and the like. The hardware equipment includes a flat reinforcing bar fixed to the wall surface of the tunnel, and a receiving bar attached to the reinforcing bar to support the communication cable. In this embodiment, the reinforcing bar has a width of about 8 cm, but is not limited thereto. The reinforcing bar extends in the Z-axis direction of FIG. 2 and is installed at regular intervals in the Y-axis direction, which is the depth direction of the tunnel. The Z-axis direction of FIG. 2 indicates the longitudinal direction of the reinforcing bar, and the Y-axis direction indicates the lateral direction of the reinforcing bar.
[0016] 1 again, the degradation assessment system 1 includes an image capturing device 10, a degradation assessment device 20, and a server device 30. The image capturing device 10, the degradation assessment device 20, and the server device 30 are connected to each other so as to be able to communicate with each other via wired or wireless communication. There are no particular limitations on the communication method for transmitting and receiving information between each device.
[0017] The photographing device 10 is a device having a function of photographing the inside of the tunnel, and is, for example, a smartphone, a tablet terminal, a notebook PC (personal computer), an unmanned aerial vehicle, etc. The photographing device 10 transmits data of the photographed image to the deterioration assessment device 20. At least a part of the equipment inside the tunnel is reflected in the photographed image. The number of photographing devices 10 is not limited to one, and may be multiple. The photographing device 10 may be integrated with the deterioration assessment device 20.
[0018] The deterioration assessment device 20 is any electronic device such as a smartphone or a tablet terminal. The deterioration assessment device 20 may be a general-purpose computer, a dedicated computer, or a PC (Personal Computer). The deterioration assessment device 20 may be used by an operator who performs inspection inside a cable tunnel. The deterioration assessment device 20 receives data of a captured image from the photographing device 10. As will be described in detail later, the deterioration assessment device 20 detects an equipment area and a deteriorated area based on the received photographed image, estimates the pixel resolution, and estimates the size of the deteriorated area. The deterioration assessment device 20 transmits information indicating the actual size value of the estimated deteriorated area to the server device 30 via a network.
[0019] The server device 30 may be installed in a facility such as a data center. The server device 30 receives information indicating the actual size values of the deteriorated area of the equipment via a network from the deterioration assessment device 20. The server device 30 stores the information indicating the actual size values of the deteriorated area in a storage unit of the server device 30.
[0020] <Deterioration detection device> An example of the configuration of a degradation assessment device 20 according to this embodiment will be described with reference to FIGS. 3 to 13B.
[0021] 3, the degradation assessment device 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, and an output unit 25. The control unit 21 includes an equipment area detection unit 211, a degraded area detection unit 212, a pixel resolution estimation unit 213, and a degradation magnitude estimation unit 214.
[0022] The storage unit 22 includes one or more memories, and may include, for example, a semiconductor memory, a magnetic memory, an optical memory, etc. Each memory included in the storage unit 22 may function, for example, as a main storage device, an auxiliary storage device, or a cache memory. Each memory does not necessarily need to be provided inside the deterioration assessment device 20, and may be configured to be provided outside the deterioration assessment device 20. The storage unit 22 stores information used for the operation of the deterioration assessment device 20 and information obtained by the operation of the deterioration assessment device 20. The storage unit 22 can store, for example, information indicating the actual value of the width of the reinforcing bars, information indicating the installation interval of the reinforcing bars, information of a created pixel resolution map, information of the actual value of the deterioration area estimated by the control unit 21, etc.
[0023] The communication unit 23 includes at least one communication interface. The communication interface is, for example, a LAN interface. The communication unit 23 receives information used in the operation of the degradation assessment device 20, and transmits information obtained by the operation of the degradation assessment device 20. The communication unit 23 receives data of a captured image from the imaging device 10. The communication unit 23 transmits information on the actual size values of the estimated deteriorated area to the server device 30.
[0024] The input unit 24 includes at least one input interface. The input interface is, for example, a physical key, a capacitance key, a pointing device, a touch screen integrally provided with a display, or a microphone. The input unit 24 accepts an operation to input information used for the operation of the deterioration assessment device 20. The input unit 24 may be connected to the deterioration assessment device 20 as an external input device instead of being provided in the deterioration assessment device 20. The connection method may be any. For example, an operator can input image data captured inside a tunnel, such as an inspection image, to the deterioration assessment device 20 by performing a predetermined operation using the input unit 24.
[0025] Output unit 25 includes at least one output interface. The output interface is, for example, a display or a speaker. The display is, for example, a liquid crystal display (LCD) or an organic EL (electro luminescence) display. Output unit 25 outputs information obtained by the operation of deterioration assessment device 20. Output unit 25 may be connected to deterioration assessment device 20 as an external output device instead of being provided in deterioration assessment device 20. The connection method may be any method. For example, output unit 25 may notify the worker of information on the actual size value of the estimated deteriorated area by image or sound.
[0026] The control unit 21 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a central processing unit (CPU) or a graphics processing unit (GPU), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC). The control unit 21 executes processes related to the operation of the degradation assessment device 20 while controlling each part of the degradation assessment device 20. The control unit 21 has a function of acquiring an image acquired inside the communication tunnel, estimating the pixel resolution of the image based on the equipment area detected from the image using an image processing method such as segmentation deep learning, and quantitatively evaluating the detection result of the degradation area.
[0027] The equipment area detection unit 211 detects reinforcing bars as an equipment area from the acquired image, and outputs the detected area to the pixel resolution estimation unit 213. The object to be detected as an equipment area is not limited to reinforcing bars, and may be freely set. The detection is performed by an image processing method such as segmentation deep learning.
[0028] The deteriorated area detection unit 212 detects a deteriorated area occurring in the main body of the tunnel from the acquired image, and outputs the detected area to the deterioration degree estimation unit 214. The detection may be performed by the same image processing method as that used by the equipment area detection unit 211.
[0029] Fig. 4A shows an example of an acquired image. Fig. 4B shows an equipment area detected by the equipment area detection section 211 and a degraded area detected by the degraded area detection section 212. The equipment area detection section 211 and the degraded area detection section 212 may be integrated to simultaneously detect an equipment area and a degraded area. When an equipment area is not detected from the acquired image, the equipment area detection section 211 may notify the imaging device 10 of this fact via the output section 25 and prompt the imaging device 10 to take an image again.
[0030] The equipment area detected by equipment area detection unit 211 and the deteriorated area detected by deteriorated area detection unit 212 may be displayed to an operator using deterioration assessment device 20 via output unit 25. In this case, the operator can operate input unit 24 to manually set the ranges of the equipment area and the deteriorated area.
[0031] The pixel resolution estimation unit 213 estimates the pixel resolution of at least a part of the image based on the equipment area detected by the equipment area detection unit 211. Specifically, the pixel resolution, which is the actual value of the length per pixel of the image, is estimated for each pixel from the width or installation interval of the reinforcing bars, and a pixel resolution map is created for the entire image, which is output to the degradation magnitude estimation unit 214. One pixel means one pixel.
[0032] As shown in FIG. 5, the pixel resolution estimation unit 213 includes an object number extraction unit 2131, an object number determination unit 2132, an object width extraction unit 2133a and an object width extraction unit 2133b, a thinning unit 2134, an installation interval extraction unit 2135, an actual dimension comparison unit 2136a, an actual dimension comparison unit 2136b and an actual dimension comparison unit 2136c, and a pixel resolution map creation unit 2137a, a pixel resolution map creation unit 2137b and a pixel resolution map creation unit 2137c.
[0033] The object number extraction unit 2131 extracts the number of reinforcing bars detected as equipment areas by dividing the detected reinforcing bars into individual objects from the equipment area detection result by the equipment area detection unit 211. For example, from the equipment area detection result shown in FIG. 4B, the object number extraction unit 2131 extracts that the number of reinforcing bars (reinforcing bar number) N is 2. The object number extraction unit 2131 outputs the extracted result to the object number determination unit 2132.
[0034] The object number determination unit 2132 determines whether the number of reinforcing metal objects N extracted by the object number extraction unit 2131 is N=1, N=2, or N≧3. As shown in Fig. 5, when the number of reinforcing metal objects N=1, the object width extraction unit 2133a, the actual dimension comparison unit 2136a, and the pixel resolution map creation unit 2137a execute the processes in this order. When the number of reinforcing metal objects N=2, the thinning unit 2134, the installation interval extraction unit 2135, the actual dimension comparison unit 2136b, and the pixel resolution map creation unit 2137b execute the processes in this order. When the number of reinforcing metal objects N≧3, the object width extraction unit 2133b, the actual dimension comparison unit 2136c, and the pixel resolution map creation unit 2137c execute the processes in this order.
[0035] The following describes a case where the object number determination unit 2132 determines that the number of reinforcing bars N=1.
[0036] The object width extraction unit 2133a measures the number of pixels corresponding to the width of the reinforcing bar, and outputs it to the actual dimension comparison unit 2136a. FIG. 6 shows an example of an image when the object number determination unit 2132 determines that the number of reinforcing bars N=1. FIG. 6 shows one reinforcing bar area detected as an equipment area and a background area. The object width extraction unit 2133a measures the number of pixels corresponding to the width of the reinforcing bar indicated by an arrow symbol in FIG. 6. The vertical position of the width of the object to be measured may be freely set. For example, as shown in FIG. 6, the number of pixels corresponding to the width at multiple positions of the reinforcing bar may be measured. By measuring the number of pixels for the width at multiple positions, the accuracy of a pixel resolution map described later can be improved.
[0037] 7A and 7B are diagrams showing other examples of acquired images. FIG. 7A is an image taken looking up at the reinforcing metal, and FIG. 7B is an image taken looking down at the reinforcing metal. The upper part of the reinforcing metal in FIG. 7A and the lower part of the reinforcing metal in FIG. 7B are displayed smaller due to perspective. When the width of the reinforcing metal changes on the image due to perspective in this way, the object width extraction unit 2133a measures the number of pixels for the width at multiple positions. For example, the object width extraction unit 2133a may measure the number of pixels for the width at the top position, the center position, and the bottom position of the reinforcing metal in the vertical direction.
[0038] The actual size comparison unit 2136a calculates the pixel resolution for each pixel and outputs it to the pixel resolution map creation unit 2137a. Specifically, the actual size comparison unit 2136a compares the number of pixels corresponding to the width of the reinforcing bar in the image measured by the object width extraction unit 2133a with the actual value of the width of the reinforcing bar, and calculates the pixel resolution by the following formula (1).
[0039]
number
[0040] In formula (1), R is the pixel resolution per pixel, and its unit is mm / pixel. B is the actual value of the width of the reinforcing bar, and its unit is mm. C is the number of pixels (i.e., the number of pixels) equivalent to the width of the reinforcing bar. In this way, the actual dimension comparison unit 2136a estimates the pixel resolution R by dividing the actual value B of the width of the reinforcing bar by the measured number of pixels C. The actual dimension comparison unit 2136a refers to the memory unit 22, reads out the actual value of the width of the reinforcing bar stored in advance, and calculates the pixel resolution R using this as the value of B. The actual dimension comparison unit 2136a may calculate the pixel resolution R using the actual value of the width of the reinforcing bar inputted via the input unit 24 as the value of B.
[0041] The pixel resolution map creation unit 2137a creates a pixel resolution map A indicating the pixel resolution for each pixel of the entire image by linearly approximating the relationship between the pixel resolution calculated by the actual size comparison unit 2136a and the coordinates of the image. The pixel resolution map creation unit 2137a outputs the created pixel resolution map A to the deterioration magnitude estimation unit 214. The coordinates are the vertical direction of the image, that is, the coordinates in the longitudinal direction of the reinforcing metal. The type of approximation method may be any, and may be, for example, a least squares method or other method. As shown in FIG. 7A, in an image taken from the bottom of the tunnel looking up at the reinforcing metal, the pixel resolution value of the upper part of the image corresponding to the upper part of the tunnel is large. Also, as shown in FIG. 7B, in an image taken from the top of the tunnel looking down at the reinforcing metal, the pixel resolution value of the lower part of the image corresponding to the lower part of the tunnel is large. In this way, in the case of an image in which only one reinforcing metal is detected, the image is often distorted in the vertical direction, and it is possible to increase the accuracy of the pixel resolution value of the entire image by linearly approximating the pixel resolution in the longitudinal direction of the reinforcing metal. This makes it possible to reproduce the distortion correction on the pixel resolution map. The pixel resolution map may be created by linear approximation in the longitudinal direction of the reinforcing bar, or may be created using an approximation method that takes into account angles, etc.
[0042] Fig. 8 is a diagram for explaining pixel resolution map A created by pixel resolution map creation unit 2137a when only one reinforcing bar is detected. The scale on the right side of the figure indicates the pixel resolution value in the image by shading. In Fig. 8, the reinforcing bar is photographed from below the cable tunnel. It can be seen that the pixel resolution value increases in the direction indicated by the arrow symbol in the image.
[0043] The following describes a case where the object number determination unit 2132 determines that the number of reinforcing bars N=2.
[0044] The thinning unit 2134 displays a thin line along the longitudinal direction of the detected reinforcing metal in the center of the width of the reinforcing metal, and performs thinning to represent the width of the reinforcing metal with a line. FIG. 9 shows an example of an image when the object number determination unit 2132 determines that the number of reinforcing metals N=2. FIG. 9 shows the areas of two reinforcing metals detected as equipment areas and the background area. The thinning unit 2134 displays a thin line as shown by the dotted line in FIG. 9, and represents the width of each of the two reinforcing metals in a linear form. The thinning unit 2134 outputs the result of thinning to the installation interval extraction unit 2135.
[0045] The installation interval extraction unit 2135 measures the number of pixels corresponding to the interval between the thin lines displayed by the thinning unit 2134, and outputs the number of pixels to the actual dimension comparison unit 2136b. The installation interval extraction unit 2135 measures the number of pixels corresponding to the interval between the thin lines, as indicated by the arrow symbol in Fig. 9. The vertical position of the interval to be measured may be freely set. For example, as shown in Fig. 9, the number of pixels corresponding to the interval at multiple positions of the reinforcing bar may be measured. By measuring the number of pixels for the interval at multiple positions, the accuracy of the pixel resolution map described later can be improved.
[0046] 10A and 10B are diagrams showing other examples of the acquired images. FIG. 10A is an image taken looking up at the reinforcing metal, and FIG. 10B is an image taken looking down at the reinforcing metal. The upper part of the spacing distance of the reinforcing metal in FIG. 10A and the lower part of the spacing distance of the reinforcing metal in FIG. 10B are displayed smaller due to perspective. In this way, when the spacing distance of the reinforcing metal changes on the image due to perspective, the installation interval extraction unit 2135 measures the number of pixels for the spacing distance at multiple positions. For example, the installation interval extraction unit 2135 may measure the number of pixels for the spacing distance at the top position, center position, and bottom position of the thin wire of the reinforcing metal in the vertical direction.
[0047] The actual dimension comparison unit 2136b calculates the pixel resolution R for each pixel and outputs it to the pixel resolution map creation unit 2137b. Specifically, the actual dimension comparison unit 2136b compares the number of pixels corresponding to the interval distance of the thin wires of the reinforcing steel in the image measured by the installation interval extraction unit 2135 with the actual value of the interval distance, and calculates the pixel resolution by formula (2).
[0048]
number
[0049] In formula (2), R is the pixel resolution per pixel, and its unit is mm / pixel. L is the actual value of the spacing distance between the reinforcing bars, and its unit is mm. C is the number of pixels (i.e., the number of pixels) corresponding to the spacing distance between the reinforcing bars. In this way, the actual dimension comparison unit 2136b estimates the pixel resolution R by dividing the actual value L of the spacing distance between the reinforcing bars by the measured number of pixels C. The actual dimension comparison unit 2136b refers to the memory unit 22, reads out the actual value of the spacing distance between the reinforcing bars that is stored in advance, and calculates the pixel resolution R as the value of L. The actual dimension comparison unit 2136b may calculate the pixel resolution R by using the actual value of the spacing distance between the reinforcing bars inputted via the input unit 24 as the value of L.
[0050] The pixel resolution map creation unit 2137b creates a pixel resolution map B indicating the pixel resolution for each pixel of the entire image by linearly approximating the relationship between the pixel resolution calculated by the actual size comparison unit 2136b and the coordinates of the image. The pixel resolution map creation unit 2137b outputs the created pixel resolution map B to the degradation magnitude estimation unit 214. Details of the pixel resolution map creation unit 2137b are similar to those of the pixel resolution map creation unit 2137a, so a description thereof will be omitted.
[0051] If the actual size of the spacing between the reinforcement bars in the tunnel is unknown, the pixel resolution may be calculated from the width of the reinforcement bars, as in the case where the object number determination unit 2132 determines that N=1. Alternatively, the pixel resolution of the entire image may be calculated based on the number of pixels and actual size measured for both the width of the reinforcement bars and the spacing distance between the fine wires of the reinforcement bars. This makes it possible to create a pixel resolution map with greater accuracy.
[0052] The following describes a case where the object number determination unit 2132 determines that the number of reinforcing bars N is 3 or more.
[0053] The object width extraction unit 2133b measures the number of pixels corresponding to the width of the reinforcing bars, and outputs the number of pixels to the actual size comparison unit 2136c. FIG. 11A and FIG. 11B show an example of an image when the object number determination unit 2132 determines that the number of reinforcing bars N is 3 or more. The images in FIG. 11A and FIG. 11B show three reinforcing bars detected as equipment areas and a background area. The image was taken toward the back of the cable tunnel, with the right side of the image being the front side of the cable tunnel and the left side being the back side. As shown by the arrow symbol, due to perspective, the reinforcing bars as equipment areas detected are displayed smaller as they move toward the back side. Other details of the object width extraction unit 2133b are the same as those of the object width extraction unit 2133a, so a description thereof will be omitted.
[0054] The object width extraction unit 2133b may measure the number of pixels of the width of each of the detected equipment areas. Alternatively, the object width extraction unit 2133b may identify multiple reinforcing bars in descending order of the proportion of the size of the equipment area in the image, and measure the number of pixels of the width of only the identified reinforcing bars. This makes it possible to omit the process of measuring the number of pixels corresponding to the width of reinforcing bars that have an extremely small proportion of the size of the equipment area in the image, i.e., reinforcing bars that exist at the back of the cable tunnel. Alternatively, when a deteriorated area is detected at the back of the image, the object width extraction unit 2133b may also measure the number of pixels corresponding to the width of reinforcing bars that exist at the back of the cable tunnel. This improves the accuracy of the pixel resolution map created by the pixel resolution map creation unit 2137c, compared to the case where the width is measured only for the identified reinforcing bars, and allows the actual size value of the deteriorated area at the back of the image to be estimated with high accuracy. In this way, the object width extraction unit 2133b can freely determine which of the detected equipment areas to measure the number of pixels of the width of.
[0055] As in the above-mentioned case where the object number determination unit 2132 determines that the number of reinforcing bars N=2, the thinning unit 2134 and the installation interval extraction unit 2135 may be provided in addition to the object width extraction unit 2133b or instead of the object width extraction unit 2133b. This allows the thinning unit 2134 and the installation interval extraction unit 2135 to flexibly measure the number of pixels of the interval distance even when the actual width value of some of the multiple reinforcing bars is unknown. FIG. 11B is a diagram for explaining a case where the thinning unit 2134 and the installation interval extraction unit 2135 measure the number of pixels of the interval distance instead of the object width extraction unit 2133b. In FIG. 11B, the thinning unit 2134 displays the widths of three or more reinforcing bars in a linear form, and the installation interval extraction unit 2135 measures the number of pixels corresponding to the interval distance of each line. Similar to the above-mentioned object width extraction unit 2133b, the thinning unit 2134 and installation interval extraction unit 2135 can freely determine which of a plurality of intervals between a plurality of reinforcing bars the number of pixels of which is to be measured.
[0056] The actual size comparing unit 2136c calculates the pixel resolution for each pixel and outputs it to the pixel resolution map creating unit 2137c. Details of the actual size comparing unit 2136c are similar to those of the actual size comparing unit 2136a or the actual size comparing unit 2136b, and therefore description thereof will be omitted.
[0057] The pixel resolution map creation unit 2137c creates a pixel resolution map showing the pixel resolution of each pixel of the entire image by linearly approximating the relationship between the pixel resolution calculated by the actual size comparison unit 2136c and the coordinates of the image. The pixel resolution map creation unit 2137c outputs the created pixel resolution map C to the deterioration size estimation unit 214. The coordinates are the left-right direction of the image, that is, the coordinates in the short direction of the reinforcing bars. The type of approximation method may be any, and may be, for example, a least squares method or other method. As shown in Figures 11A and 11B, in an image in which multiple reinforcing bars are photographed from the front side of the tunnel to the back, the pixel resolution value of the left side of the image corresponding to the back side of the tunnel is large, and the pixel resolution value of the right side of the image corresponding to the front side of the tunnel is small. Note that when the left side of the image corresponds to the front side of the tunnel and the right side of the image corresponds to the back side of the tunnel, the pixel resolution value of the right side of the image is large and the pixel resolution value of the left side of the image is small. In this way, when an image contains three or more reinforcing bars, the image is often distorted in the left-right direction, so by linearly approximating the pixel resolution in the short direction of the reinforcing bars, it is possible to improve the accuracy of the pixel resolution value of the entire image. This allows the distortion correction to be reproduced on the pixel resolution map.
[0058] The degradation magnitude estimation unit 214 estimates the actual size value of the degraded region detected by the degraded region detection unit 212, based on the pixel resolution map created by the pixel resolution estimation unit 213. This allows the degradation magnitude estimation unit 214 to evaluate the degraded region. As shown in FIG. 12, the degradation magnitude estimation unit 214 includes a degraded region coordinate acquisition unit 2141, a degraded region length estimation unit 2142, and a degraded region area estimation unit 2143.
[0059] The deteriorated area coordinate acquisition unit 2141 acquires the coordinates in the image of the area detected as the deteriorated area. Depending on the acquired coordinate value, the deteriorated area coordinate acquisition unit 2141 determines which of the deteriorated area length estimation unit 2142 and the deteriorated area area estimation unit 2143 will perform the next process, and outputs the acquired coordinates to either the deteriorated area length estimation unit 2142 or the deteriorated area area estimation unit 2143. For example, when the deterioration in the photographed inside the tunnel is linear deterioration with an indefinite shape such as cracks, the deteriorated area coordinate acquisition unit 2141 determines which processing the deteriorated area length estimation unit 2142 will perform next from the acquired coordinate value. Then, outputs the acquired coordinates to the deteriorated area length estimation unit 2142. For example, when the deterioration in the photographed inside the tunnel is deterioration such as hole-shaped defects with a certain area, the deteriorated area coordinate acquisition unit 2141 determines which processing the deteriorated area area estimation unit 2143 will perform next from the acquired coordinate value. Then, outputs the acquired coordinates to the deteriorated area area estimation unit 2143.
[0060] It may be determined whether the next process is to be performed by the degraded region length estimation unit 2142 or the degraded region area estimation unit 2143, depending on a recording item or an inspection item of the degraded region that is set in advance.
[0061] The deteriorated area length estimation unit 2142 calculates the length of the deteriorated area from the coordinates indicating the deteriorated area, and stores the result in the storage unit 22. Here, an example of calculating the maximum length of the deteriorated area will be described, but the length to be calculated may be set freely. FIG. 13A shows an example of the coordinates of the acquired deteriorated area on the image. Referring to FIG. 13A, cracks as deteriorated areas are shown by dotted lines. The deteriorated area length estimation unit 2142 extracts, from the coordinates of the deteriorated area, point A where the value of the vertical axis (Y axis) is the maximum value, point B where the value of the horizontal axis (X axis) is the maximum value, point C where the value of the vertical axis (Y axis) is the minimum value, and point D where the value of the horizontal axis (X axis) is the minimum value. The deteriorated area length estimation unit 2142 creates a quadrangle formed by straight lines connecting point A to point D, and determines that, among the diagonal lines of the quadrangle, the straight line connecting point A and point C has the maximum length L'max. The deteriorated area length estimation unit 2142 may extract points that form an arbitrary polygon according to the shape of the deteriorated area. The degraded region length estimation unit 2142 may determine any one of the sides or diagonals of the polygon as the maximum length.
[0062] Next, the degraded area length estimation unit 2142 plots the coordinates of points A and C on the pixel resolution map. FIG. 13B is a diagram for explaining a state in which points A and C forming the maximum length obtained in FIG. 13A are plotted on the pixel resolution map. The degraded area length estimation unit 2142 obtains the distance between the plotted points, that is, the actual value of the maximum length of the degraded area, from the pixel resolution value indicated by the pixel resolution map. On the pixel resolution map of FIG. 13B, the degraded area length estimation unit 2142 forms a right-angled triangle whose hypotenuse is a straight line connecting the plotted points A and C. The degraded area length estimation unit 2142 measures the pixel resolution of pixels corresponding to a straight line extending downward from point A along the vertical axis and a straight line extending leftward from point C along the horizontal axis, which form a right angle, and obtains the actual value of each straight line.
[0063] In FIG. 13B, the pixel resolution value of the pixel corresponding to the line extending downward from point A is 1.1+1.1+1.1+1.1+1.1=5.5 mm / pixel, and the actual size of the line is 5.5 mm. The pixel resolution value of the pixel corresponding to the line extending leftward from point C is 1.1+1.1+1.2=3.4 mm / pixel, and the actual size of the line is 3.4 mm. The degraded area length estimation unit 2142 then calculates the length of the hypotenuse using Pythagoras' theorem. Specifically, the degraded area length estimation unit 2142 calculates the value of 41.81 mm by adding up the value of 30.25 mm obtained by squaring 5.5 mm and the value of 11.56 mm obtained by squaring 3.4 mm. Then, the square root of the value of 41.81 mm is calculated, and the value is rounded off to the first decimal place to calculate the value of 6.5 mm. In this way, the actual length of the hypotenuse L'max is determined to be 6.5 mm.
[0064] When the degraded area whose length is to be estimated is parallel to the vertical or horizontal axis on the pixel resolution map, the degraded area length estimation unit 2142 may simply sum up the pixel resolution values corresponding to a straight line between the coordinates indicating the degraded area to calculate the actual size value of the line.
[0065] The degraded region area estimation unit 2143 obtains the area of the degraded region and stores the result in the storage unit 22. Specifically, the degraded region area estimation unit 2143 plots the acquired coordinate values of the degraded region on a pixel resolution map. The degraded region area estimation unit 2143 identifies the pixel resolution value of each pixel corresponding to the degraded region based on the plotted coordinates. Next, the degraded region area estimation unit 2143 squares each of the identified pixel resolution values to obtain an actual value of the area per pixel. The degraded region area estimation unit 2143 adds up the obtained actual value of the area per pixel and calculates the sum as the actual value of the area of the entire degraded region.
[0066] <Program> The degradation assessment device 20 may be a computer capable of executing program instructions. The computer stores a program describing the processing contents for realizing each function of the degradation assessment device 20 in a storage unit of the computer, and reads and executes this program by a processor of the computer. A part of these processing contents may be realized by hardware. Here, the computer may be a general-purpose computer, a dedicated computer, a workstation, a PC, an electronic notepad, etc. The program instructions may be program code, code segments, etc. for performing necessary tasks. The processor may be a CPU, a GPU, a DSP (Digital Signal Processor), etc.
[0067] The program may be recorded on a computer-readable recording medium. By using such a recording medium, the program can be installed on a computer. Here, the recording medium on which the program is recorded may be a non-transient recording medium. The non-transient recording medium is not particularly limited, and may be, for example, a CD-ROM, a DVD-ROM, or the like. The program may also be provided by downloading via a network.
[0068] <Operation of Deterioration Assessment System 1> Next, the operation of the degradation assessment system 1 including the degradation assessment device 20 according to this embodiment will be described with reference to Figures 13A, 13B, 14A, 14B, 15A, and 15B. Of the operations of the degradation assessment system 1, the operation of the degradation assessment device 20 corresponds to the degradation assessment method according to this embodiment.
[0069] In step S1, the imaging device 10 captures images of deteriorated parts in the tunnel and the facilities in the tunnel. The imaging device 10 transmits the captured image data to the deterioration assessment device 20.
[0070] In step S2, communication unit 23 of degradation assessment device 20 receives image data from photographing device 10. Control unit 21 of degradation assessment device 20 acquires the received image.
[0071] In step S3, the equipment area detection unit 211 of the deterioration assessment device 20 detects an equipment area from the acquired image. In this example, reinforcing bars are detected as the equipment area. Here, if the equipment area detection unit 211 determines that no equipment is captured in the image, it may notify the imaging device 10 of this fact. The equipment area detection unit 211 outputs the detection result to the pixel resolution estimation unit 213.
[0072] In step S4, the degraded area detection section 212 of the degradation assessment device 20 detects a degraded area from the acquired image. The degraded area detection section 212 outputs the detection result to the degradation magnitude estimation section 214.
[0073] The order of steps S3 and S4 may be reversed, and the processes of steps S3 and S4 may be performed simultaneously.
[0074] In step S5, pixel resolution estimation unit 213 of degradation assessment device 20 estimates the pixel resolution of at least a part of the image, based on the detected equipment area. Figures 15A and 15B show a specific processing flow for estimating the pixel resolution in step S5 of Figure 14A.
[0075] First, in step S6, the object number extraction unit 2131 extracts the number of reinforcing bars detected as the equipment area. The object number extraction unit 2131 outputs the extraction result to the object number determination unit 2132.
[0076] In step S7, the object number determination unit 2132 determines whether the extracted number of reinforcing metal pieces N is N=1, N=2, or N≧3. If the number of reinforcing metal pieces N=1, then in step S8, the object width extraction unit 2133a performs processing, followed by the actual dimension comparison unit 2136a and the pixel resolution map creation unit 2137a performing processing in that order. If the number of reinforcing metal pieces N=2, then in step S11, the thinning unit 2134 performs processing, followed by the installation interval extraction unit 2135, the actual dimension comparison unit 2136b, and the pixel resolution map creation unit 2137b performing processing in that order. If the number of reinforcing metal pieces N≧3, then in step S15, the object width extraction unit 2133b performs processing, followed by the actual dimension comparison unit 2136c and the pixel resolution map creation unit 2137c performing processing in that order.
[0077] First, a case where it is determined in step S7 that the number of reinforcing bars N=1 will be described. In step S8, the object width extraction unit 2133a measures the number of pixels equivalent to the width of the reinforcing bars. The position of the width of the object to be measured may be freely set, and the number of pixels equivalent to the width at multiple positions of the reinforcing bars may be measured. For example, the object width extraction unit 2133a may measure the number of pixels for the width at the topmost position, the center position, and the bottommost position in the vertical direction of the reinforcing bars. The object width extraction unit 2133a outputs the measurement results to the actual dimension comparison unit 2136a.
[0078] In step S9 of Fig. 15B, the actual dimension comparison unit 2136a calculates the pixel resolution for each pixel. Specifically, the actual dimension comparison unit 2136a compares the number of pixels corresponding to the width of the reinforcing bar in the image measured in step S8 with the actual size value of the reinforcing bar stored in advance in the storage unit 22, and obtains the pixel resolution using the above-mentioned formula (1). The actual dimension comparison unit 2136a outputs the calculation result to the pixel resolution map creation unit 2137a.
[0079] In step S10, the pixel resolution map creation unit 2137a creates a pixel resolution map A that indicates the pixel resolution for each pixel of the entire image by linearly approximating the relationship between the pixel resolution calculated by the actual dimension comparison unit 2136a and the coordinates of the image. The pixel resolution map is created by linearly approximating the longitudinal direction of the reinforcing bar. The pixel resolution map creation unit 2137a outputs the created pixel resolution map A to the degradation magnitude estimation unit 214.
[0080] Next, a case where it is determined in step S7 that the number of reinforcing bars N=2 will be described. In step S11, the thinning unit 2134 performs thinning processing to display a thin line along the longitudinal direction of the reinforcing bar at the center of the width of the detected reinforcing bar, and to represent the width of the reinforcing bar in a linear form. The thinning unit 2134 outputs the result of the thinning to the installation interval extraction unit 2135.
[0081] 15B, the installation interval extraction unit 2135 measures the number of pixels equivalent to the interval distance of the thin line displayed by the thinning unit 2134, and outputs the number of pixels to the actual dimension comparison unit 2136b. The vertical position of the interval distance to be measured may be freely set, and the number of pixels equivalent to the interval distance at multiple positions of the thin line may be measured.
[0082] In step S13, the actual dimension comparison unit 2136b calculates the pixel resolution for each pixel and outputs it to the pixel resolution map creation unit 2137b. Specifically, the actual dimension comparison unit 2136b compares the number of pixels corresponding to the spacing distance of the thin lines shown in the reinforcing bars in the image measured in step S12 with the actual value of the spacing distance of the reinforcing bars stored in advance in the storage unit 22, and calculates the pixel resolution by the above-mentioned formula (2).
[0083] In step S14, the pixel resolution map creation unit 2137b creates a pixel resolution map B indicating the pixel resolution for each pixel of the entire image by linearly approximating the relationship between the pixel resolution calculated by the actual size comparison unit 2136b and the coordinates of the image. The pixel resolution map creation unit 2137b outputs the created pixel resolution map B to the degradation magnitude estimation unit 214.
[0084] Next, a case where it is determined in step S7 that the number of reinforcing bars N is equal to or greater than 3 will be described. In step S15, the object width extraction unit 2133b measures the number of pixels corresponding to the width of the reinforcing bars. The object width extraction unit 2133b may measure the number of pixels for the width of each of all detected reinforcing bars. Alternatively, the object width extraction unit 2133b may identify multiple reinforcing bars in descending order of the proportion of the size of the equipment area that occupies the image, and measure the number of pixels for the width of only the identified reinforcing bars. The object width extraction unit 2133b outputs the measurement results to the actual dimension comparison unit 2136c.
[0085] In step S16 of Fig. 15B, the actual size comparison unit 2136c calculates the pixel resolution for each pixel. The details of the process in step S16 are the same as those in step S9 or step S13, so the description will be omitted. The actual size comparison unit 2136c outputs the calculation result to the pixel resolution map creation unit 2137c.
[0086] In step S17, the pixel resolution map creation unit 2137c creates a pixel resolution map C that indicates the pixel resolution for each pixel of the entire image by linearly approximating the relationship between the pixel resolution calculated by the actual dimension comparison unit 2136c and the coordinates of the image. The pixel resolution map is created by linearly approximating the short side direction of the reinforcing bar. The pixel resolution map creation unit 2137c outputs the created pixel resolution map C to the degradation magnitude estimation unit 214.
[0087] After performing the above-mentioned processes, pixel resolution estimation unit 213 ends the estimation of pixel resolution.
[0088] Referring again to FIG. 14A, in step S18, the deteriorated area coordinate acquisition unit 2141 of the deterioration magnitude estimation unit 214 acquires the coordinates in the image of the area detected as the deteriorated area.
[0089] In step S19 of FIG. 14B, the degraded region coordinate acquisition unit 2141 determines whether the degraded region length estimation unit 2142 or the degraded region area estimation unit 2143 will perform the next process according to the value of the acquired coordinates, and outputs the acquired coordinates to either the degraded region length estimation unit 2142 or the degraded region area estimation unit 2143. When the degraded region is linearly degraded, the degraded region coordinate acquisition unit 2141 determines that the degraded region length estimation unit 2142 will perform the next process, and outputs the acquired coordinates to the degraded region length estimation unit 2142. In this case, the process proceeds to step S20. When the degraded region is a shape having an area, the degraded region coordinate acquisition unit 2141 determines that the degraded region area estimation unit 2143 will perform the next process, and outputs the acquired coordinates to the degraded region area estimation unit 2143. In this case, the process proceeds to step S21.
[0090] First, a case will be described where it is determined in step S19 that the degraded region length estimation unit 2142 will next perform processing, and the processing proceeds to step S20. In step S20, the degraded region length estimation unit 2142 calculates the length of the degraded region from the coordinates indicating the degraded region. The degraded region length estimation unit 2142 stores the calculation result in the storage unit 22. Here, an example of calculating the maximum length of the degraded region will be described, but the length to be calculated may be freely set.
[0091] Specifically, the deteriorated area length estimation unit 2142 extracts points that form an arbitrary polygon according to the shape of the deteriorated area, and determines one of the diagonals of the polygon as the maximum length. For example, referring to FIG. 13A, a crack as a deteriorated area is indicated by a dotted line. The deteriorated area length estimation unit 2142 extracts, from the coordinates of the deteriorated area, point A with the maximum value on the vertical axis (Y axis), point B with the maximum value on the horizontal axis (X axis), point C with the minimum value on the vertical axis (Y axis), and point D with the minimum value on the horizontal axis (X axis). The deteriorated area length estimation unit 2142 creates a quadrangle formed by the straight lines connecting point A to point D, and determines that, among the diagonal lines of the quadrangle, the straight line connecting point A and point C has the maximum length L'max.
[0092] Next, the degraded region length estimation unit 2142 plots the coordinates of the points forming the maximum length on one of the pixel resolution maps A to C. Then, the degraded region length estimation unit 2142 obtains the distance between the plotted points, that is, the actual value of the maximum length of the degraded region, from the pixel resolution value indicated by the pixel resolution map. For example, FIG. 13B shows a state in which points A and C forming the maximum length are plotted on the pixel resolution map. The degraded region length estimation unit 2142 obtains the distance between the plotted coordinate values, that is, the actual value of the maximum length of the degraded region, from the pixel resolution value indicated by the pixel resolution map. On the pixel resolution map of FIG. 13B, the degraded region length estimation unit 2142 forms a right-angled triangle with a straight line connecting the plotted points A and C as the hypotenuse. The degraded region length estimation unit 2142 measures the pixel resolution of the pixels corresponding to the straight line extending downward from point A along the vertical axis and the straight line extending leftward from point C along the horizontal axis, which form a right angle, and obtains the actual value of each straight line. 13B, the degraded area length estimation unit 2142 sums up the pixel resolution values of the pixels corresponding to the line extending from point A and calculates that the actual size of the line is 5.5 mm. Also, the degraded area length estimation unit 2142 sums up the pixel resolution values of the pixels corresponding to the line extending from point C and calculates that the actual size of the line is 3.4 mm. Then, according to Pythagoras' theorem, the degraded area length estimation unit 2142 calculates that the actual size of the hypotenuse connecting points A and C is 6.5 mm.
[0093] Next, a case where it is determined in step S19 that the degraded region area estimation unit 2143 will next perform processing, and the processing proceeds to step S21 will be described. In step S21, the degraded region area estimation unit 2143 calculates the area of the degraded region. The degraded region area estimation unit 2143 stores the calculation result in the storage unit 22.
[0094] Specifically, the degraded region area estimation unit 2143 plots the acquired coordinate values of the degraded region on one of the pixel resolution maps A to C. The degraded region area estimation unit 2143 identifies the pixel resolution value of each pixel corresponding to the degraded region based on the plotted coordinates. Next, the degraded region area estimation unit 2143 squares each of the identified pixel resolution values to obtain an actual value of the area per pixel. The degraded region area estimation unit 2143 adds up the obtained actual value of the area per pixel, and calculates the sum as the actual value of the area of the entire degraded region.
[0095] As shown in steps S 18 to S 21 , the degradation degree estimation unit 214 estimates the actual size value of the degraded area detected by the degraded area detection unit 212 based on the pixel resolution map created by the pixel resolution estimation unit 213 .
[0096] In step S22, control unit 21 of deterioration assessment device 20 reads out information indicating the actual size values of the estimated deteriorated area from storage unit 22, and transmits it to server device 30 via communication unit .
[0097] In step S23, the server device 30 receives the information indicating the actual size value of the deteriorated area, and stores it in the storage unit of the server device 30.
[0098] As described above, the deterioration assessment device 20 of this embodiment comprises an equipment area detection unit 211 that detects an equipment area from an acquired image, a deteriorated area detection unit 212 that detects a deteriorated area from the image, a pixel resolution estimation unit 213 that estimates the pixel resolution of at least a part of the image based on the detected equipment area, and a deterioration magnitude estimation unit 214 that estimates the size of the deteriorated area based on the pixel resolution.
[0099] According to the present embodiment, the pixel resolution is estimated from the detected equipment area, and the actual size value of the deteriorated area can be estimated by using the pixel resolution. Therefore, it is possible to provide a deterioration assessment device and a deterioration assessment method that can evaluate the actual size value of the deteriorated area without being limited to certain shooting conditions.
[0100] As described above, in the degradation assessment device 20 according to this embodiment, the pixel resolution estimation unit 213 further includes a pixel resolution map creation unit 2137 that creates a pixel resolution map of the entire image by approximating the relationship between the coordinates of the image and the pixel resolution. The degradation magnitude estimation unit 214 estimates the size of the degraded region based on the pixel resolution indicated by the pixel resolution map.
[0101] According to this embodiment, the pixel resolution map of the entire image can be grasped in a unified manner by using the pixel resolution map. By overlaying the coordinates of the degraded area on the pixel resolution map, the actual size of the degraded area can be easily estimated. Therefore, it is possible to provide a degradation assessment device and a degradation assessment method that can assess the actual size of the degraded area without being limited to certain shooting conditions.
[0102] As described above, in the deterioration assessment device 20 of this embodiment, the pixel resolution estimation unit 213 further includes an object number extraction unit 2131 that extracts the number of equipment indicated by the equipment area, and an actual size comparison unit 2136 that estimates the pixel resolution by comparing the number of pixels indicating the equipment area with the actual size value according to the extracted number of equipment.
[0103] According to the present embodiment, a method for estimating pixel resolution is selected according to the number of pieces of equipment. This allows the pixel resolution of an image to be estimated with higher accuracy. Therefore, it is possible to provide a degradation assessment device and a degradation assessment method that are capable of assessing the actual size of a deteriorated area without being limited to certain shooting conditions.
[0104] As described above, in the degradation assessment device 20 according to this embodiment, the pixel resolution estimation unit 213 further includes an object width extraction unit 2133 that measures the number of pixels equivalent to the width of the equipment when the number of equipment extracted by the object number extraction unit 2131 is 1 or 3 or more. The actual dimension comparison unit 2136 estimates the pixel resolution by dividing the actual value of the width of the equipment by the measured number of pixels.
[0105] According to this embodiment, when the number of facilities is one or three or more, the pixel resolution can be estimated with high accuracy based on the number of pixels corresponding to the width of the facilities and the known actual value of the width of the facilities. Since an image in which only one reinforcing bar is detected is an image taken by taking a relatively close-up shot of the wall surface of the tunnel, and the deteriorated area is often captured near the reinforcing bar, the actual value of the deteriorated area can be estimated with high accuracy by estimating the pixel resolution for the number of pixels corresponding to the width of the reinforcing bar and the actual value. On the other hand, an image in which three or more reinforcing bars are detected is often an image taken from the front of the tunnel toward the back, and in this case, the deteriorated area present at the back of the tunnel is likely to be captured small. By estimating the pixel resolution for the number of pixels corresponding to the width of multiple reinforcing bars and the actual value, the actual value of such a deteriorated area can also be estimated with high accuracy. Therefore, it is possible to provide a deterioration evaluation device and a deterioration evaluation method that can evaluate the actual value of a deteriorated area without being limited to certain shooting conditions.
[0106] As described above, in the degradation assessment device 20 according to this embodiment, the pixel resolution estimation unit 213 further includes a thinning unit 2134 that represents the width of the facility as a line when the number of facilities extracted by the object number extraction unit 2131 is 2, and an installation interval extraction unit 2135 that measures the number of pixels equivalent to the interval distance of the thin lines. The actual dimension comparison unit 2136 estimates the pixel resolution by dividing the actual value of the interval distance by the measured number of pixels.
[0107] According to this embodiment, when the number of pieces of equipment is two, the equipment is represented by thin lines, and the pixel resolution can be estimated with high accuracy based on the number of pixels corresponding to the spacing distance of the thin lines and the known actual value of the spacing distance of the equipment. Even if two reinforcing bars are photographed away from the center of the image, the pixel resolution of the entire image can be reproduced closer to the actual value based on the actual value of the spacing distance of the reinforcing bars and the measured number of pixels. Therefore, it is possible to provide a deterioration assessment device and a deterioration assessment method that can assess the actual value of a deteriorated area without being limited to certain shooting conditions.
[0108] Although the present disclosure has been described based on the drawings and embodiments, it should be noted that those skilled in the art can easily make various modifications and corrections based on the present disclosure. Therefore, it should be noted that these modifications and corrections are included in the scope of the present disclosure.
[0109] (Variation 1) As a modified example of the present disclosure, control unit 21 of degradation assessment device 20 may further include an equipment area correction unit 215. The equipment area correction unit 215 performs at least one of a process of deleting an overdetected area from the equipment area and a process of interpolating a divided area in the equipment area.
[0110] An over-detected area is an area where an area other than an equipment area is detected as an equipment area by the equipment area detection unit 211. For example, when a background area in an image is detected as an equipment area, the detected area is called an over-detected area.
[0111] First, the process in which the equipment area correction unit 215 detects an area overdetected by the equipment area detection unit 211 and deletes the area will be described. In FIG. 16A, the overdetected area is indicated by an arrow symbol. The detection of the overdetected area may be performed by any image analysis method. For example, the equipment area correction unit 215 binarizes the result detected by the equipment area detection unit 211 based on the density of the pixels, and refers to pixels in any peripheral range of 4 or 8 neighbors of the target pixel, and if there are pixels with the same value, the pixels with the same value are treated as one connected pixel. The equipment area correction unit 215 deletes connected pixels whose size is equal to or smaller than a threshold. The threshold in this case may be set arbitrarily. After deleting the overdetected area, the equipment area correction unit 215 generates a corrected equipment area by correcting the equipment area. FIG. 16B shows the corrected equipment area in a state in which the overdetected area in FIG. 16A has been deleted.
[0112] Next, the process of the equipment area correction unit 215 to interpolate the divided area in the equipment area detected by the equipment area detection unit 211 will be described. Here, the division of the equipment area occurs because an object exists between the image capture device 10 and the equipment. In FIG. 17A, an equipment area that is divided because a cable or a metal bracket for supporting the cable exists in front of a metal brace is indicated by an arrow symbol. The detection of the divided area may be performed by any image analysis method. The equipment area correction unit 215 interpolates the divided area by, for example, a morphological transformation method. In addition, although the type of morphological transformation is not limited, a closing process in which a contraction process is performed after an expansion process is effective and desirable. After the interpolation of the divided area, the equipment area correction unit 215 generates a corrected equipment area by correcting the equipment area. FIG. 17B shows the corrected equipment area in a state in which the divided area detected in FIG. 17A is interpolated.
[0113] 18, the equipment area correction unit 215 can execute processing after the processing by the equipment area detection unit 211. This allows the object number extraction unit 2131 included in the pixel resolution estimation unit 213 to extract the number of pieces of equipment indicated by the corrected equipment area generated by the equipment area correction unit 215.
[0114] The equipment area correction unit 215 may be capable of detecting divided areas in the degraded area detected by not only the equipment area detection unit 211 but also the degraded area detection unit 212 and interpolating the areas. In this case, the equipment area correction unit 215 can execute processing after the processing by the degraded area detection unit 212.
[0115] As described above, the degradation assessment device 20 according to the first modification further includes an equipment area correction unit 215 that generates a corrected equipment area by correcting the equipment area by at least one of a process of deleting over-detected areas from the equipment area and a process of interpolating divided areas in the equipment area. The object number extraction unit 2131 extracts the number of pieces of equipment indicated by the corrected equipment area.
[0116] According to this modification, it is possible to accurately grasp the number of facilities in an image by deleting overdetected areas or by interpolating divided areas. Therefore, it is possible to estimate pixel resolution with higher accuracy and estimate the actual size of a deteriorated area. Therefore, it is possible to provide a deterioration assessment device and a deterioration assessment method that are capable of assessing the actual size of a deteriorated area without being limited to certain shooting conditions.
[0117] (Variation 2) As a modification of the present disclosure, the pixel resolution map created by pixel resolution map creation section 2137a, pixel resolution map creation section 2137b, or pixel resolution map creation section 2137c may have a two-dimensional angle.
[0118] Depending on the shooting angle of the image, the pixel resolution of the whole image may change in both the left-right direction and the up-down direction of the image. For example, in the example shown in FIG. 19, the image has an angle, and three reinforcing bars are detected. In this case, as shown by the arrow symbols, the object width extraction unit 2133b, the thinning unit 2134, and the installation interval extraction unit 2135 measure pixels corresponding to the width and installation interval of the reinforcing bars, and the actual dimension comparison unit 2136b and the actual dimension comparison unit 2136c estimate the pixel resolution in both the longitudinal and lateral directions of the reinforcing bars. Then, the pixel resolution map creation unit 2137b and the pixel resolution map creation unit 2137c create a pixel resolution map D that changes two-dimensionally.
[0119] FIG. 20 is a diagram for explaining pixel resolution map D created based on the image shown in FIG. 19. The scale on the right side of the figure indicates the pixel resolution value in the image by shading. Looking at FIG. 20, it can be seen that the pixel resolution value increases along the direction indicated by the two arrow symbols in the image, that is, from the upper right to the lower left of the image. Thus, in pixel resolution map D, the pixel resolution value changes two-dimensionally with an angle.
[0120] According to this modification, it is possible to estimate the pixel resolution and create a pixel resolution map taking into account the shooting angle of the image, and based on this, it is possible to estimate the actual size value of the deteriorated area with higher accuracy. Therefore, it is possible to provide a degradation assessment device and a degradation assessment method that are capable of assessing the actual size value of the deteriorated area without being limited to certain shooting conditions. [Explanation of symbols]
[0121] 1. Deterioration assessment system 10. Imaging Device 20 Deterioration evaluation device 30 Server device 21 Control section 22 Memory section 23 Communications Department 24 Input section 25 Output section 211 Equipment area detection unit 212 Degraded area detection unit 213 Pixel Resolution Estimation Unit 2131 Object number extraction part 2132 Object number judgment unit 2133a,2133b Object width extraction part 2134 Object thinning section 2135 Installation interval extraction part 2136a, 2136b, 2136c Actual size comparison section 2137a, 2137b, 2137c Pixel resolution map creation section 214 Deterioration magnitude estimation unit 2141 Deterioration area coordinate acquisition unit 2142 Degraded area length estimation unit 2143 Deterioration area area estimation part 215 Equipment Area Correction Department
Claims
1. an equipment area detection unit that detects an equipment area from the acquired image; a degraded region detection unit for detecting a degraded region from the image; a pixel resolution estimation unit that estimates a pixel resolution of at least a part of the image based on the detected facility area; a degradation size estimation unit that estimates a size of the degradation area based on the pixel resolution; A degradation assessment device comprising: The pixel resolution estimation unit an object number extraction unit that extracts the number of pieces of equipment indicated by the equipment area; an actual size comparison unit that estimates the pixel resolution by comparing the number of pixels indicating the equipment area with an actual size value according to the number of the extracted equipment; A degradation assessment device comprising:
2. The pixel resolution estimation unit a pixel resolution map creation unit that creates a pixel resolution map of the entire image by approximating a relationship between the coordinates of the image and the pixel resolution, The degradation assessment device according to claim 1 , wherein the degradation magnitude estimating section estimates the size of the degraded region based on the pixel resolution indicated by the pixel resolution map.
3. The pixel resolution estimation unit and an object width extraction unit that measures the number of pixels corresponding to a width of the facility when the number of the facilities extracted by the object number extraction unit is 1 or 3 or more; 3 . The degradation assessment device according to claim 1 , wherein the actual dimension comparison unit estimates the pixel resolution by dividing an actual value of the width of the facility by the number of measured pixels.
4. The pixel resolution estimation unit a thinning unit that, when the number of facilities extracted by the object number extraction unit is two, represents a width of the facilities by a line; An installation interval extraction unit that measures the number of pixels corresponding to the interval distance of the lines, 3 . The degradation assessment device according to claim 1 , wherein the actual size comparison unit estimates the pixel resolution by dividing the actual size value of the gap distance by the number of measured pixels.
5. an equipment area correction unit that generates a corrected equipment area by correcting the equipment area through at least one of a process of deleting an overdetected area from the equipment area and a process of interpolating a divided area in the equipment area; The degradation assessment device according to claim 1 , wherein the object number extraction unit extracts the number of pieces of equipment indicated by the corrected equipment area.
6. A degradation assessment method for assessing degradation using a degradation assessment device, comprising: detecting an equipment area from the acquired image; detecting a degraded region from the image; Extracting the number of pieces of equipment indicated by the equipment area; A step of estimating a pixel resolution of at least a part of the image by comparing the number of pixels indicating the detected equipment area with an actual size value according to the number of the extracted equipment; estimating a size of the degraded region based on the pixel resolution; A deterioration evaluation method comprising:
7. A program for causing a computer to function as the degradation assessment device according to claim 1 .
Citation Information
Patent Citations
Automatic diagnosing method for coating film deterioration
JP1996043315A
Surface inspection device and surface inspection method
JP2009174923A
Conveyance system
JP2011066277A
Vehicle number recognition device and method for adjusting the same
JP2011086082A
System and method for evaluating pattern height on manhole iron cover
JP2016113794A
