Working face monitoring program and working face monitoring terminal

Abstract

To provide a cutting face monitoring program and a cutting face monitoring terminal capable of supporting accurate evaluation of the degree of progress of weathering deterioration of a cutting face surface.SOLUTION: An evaluation target image acquisition step of acquiring an evaluation target image included in a captured image of a working face of a bedrock in a tunnel pit; Acquiring at least an a * value as a color specification value among an L value, the a * value, and a b * value indicating a color of the pixel, acquiring an evaluation value defined according to the color specification value of the pixel, and causing a display device to display an evaluation value distribution image in which a color corresponding to the evaluation value is assigned to each of the plurality of pixels.SELECTED DRAWING: Figure 10

Description

[Technical Field]

[0001] The present invention relates to a face monitoring program for evaluating weathering alteration on the face of rock mass inside a tunnel, and a face monitoring terminal device. [Background technology]

[0002] Conventionally, methods for evaluating the degree of weathering of a tunnel face based on photographed images of the tunnel face have been known. For example, in Patent Document 1, photographed images of the tunnel face are divided into multiple evaluation target images, and each evaluation target image is input to the input layer of a learning model that has completed deep learning. The output layer of the learning model outputs an evaluation category that indicates the degree of weathering for each evaluation target image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6784239 Summary of the Invention [Problem to be solved by the invention]

[0004] Generally, the color of a photographed image is expressed using RGB values. The color of the face changes depending on the degree of weathering of the face, but if the slight difference in color change is not expressed with sufficient sensitivity as RGB values, there is a risk that the weathering will not be accurately evaluated.

[0005] An object of the present invention is to provide a face monitoring program and a face monitoring terminal device that can assist in accurate evaluation of the degree of progress of weathering of the face. [Means for solving the problem]

[0006] 1) A face monitoring program according to at least one embodiment of the present invention, On the computer, an evaluation target image acquisition step of acquiring an evaluation target image included in a photographed image of a rock face inside a tunnel; a color value acquisition step of acquiring, for each of a plurality of pixels constituting the image to be evaluated, at least the a* value among the L value, a* value, and b* value indicating the color of the pixel as a color value; an evaluation value acquisition step of acquiring an evaluation value defined in accordance with the color specification value of the pixel; an image display step of displaying, on a display device, an evaluation value distribution image in which a color corresponding to the evaluation value is assigned to each of the plurality of pixels; Execute the following.

[0007] According to the above configuration 1), an evaluation value for each pixel is acquired based at least on the a* value, which varies depending on the redness or greenness of the rock, among the L value, a* value, and b* value that indicate the color of the image to be evaluated. An evaluation value distribution image, in which each pixel is assigned a color corresponding to the evaluation value, is then displayed on the display device. According to the inventor's knowledge, the rock becomes more reddish as weathering progresses, and more greenish as alteration progresses. Therefore, the evaluation value distribution image determined based at least on the a* value accurately indicates the degree of weathering on the face, contributing to the accurate judgment of workers visually checking the actual degree of weathering on the face. This realizes a face monitoring program that can support accurate evaluation of the progress of weathering on the face.

[0008] 2) In some embodiments, the face monitoring program described in 1) above is The computer, a progress degree acquisition step of acquiring a progress degree of weathering of the face of the workpiece shown in the evaluation target image based on the evaluation value distribution image; Then run In the image display step, an evaluation result image generated by superimposing the progress level on the evaluation value distribution image is displayed on the display device.

[0009] According to the configuration 2) above, an evaluation result image is displayed in which the degree of weathering determined based on the evaluation value distribution image is superimposed on the evaluation value distribution image. This allows workers to immediately visually confirm the mechanical determination result of the degree of weathering. Therefore, the face monitoring program can support accurate evaluation of the degree of weathering on the face.

[0010] 3) In some embodiments, the face monitoring program according to 2) above, In the progress degree acquisition step, the progress degree for each of a plurality of evaluation target regions that divide the evaluation target image is acquired based on the color value for each of two or more of the pixels included in the evaluation target region; In the image display step, the progress level for each of the evaluation target regions is superimposed on the evaluation distribution image and displayed on the display device.

[0011] According to the configuration of 3) above, the degree of weathering is evaluated for each of the multiple evaluation target areas that divide the evaluation target image. This helps the worker evaluate the degree of weathering for each evaluation target area.

[0012] 4) In some embodiments, the face monitoring program according to 2) or 3) above is The computer, a data transmission step of transmitting correspondence data that associates the evaluation target image with the evaluation result image to a first computer located away from the computer; Further execute the following.

[0013] According to the configuration 4) above, the face observed during the tunnel excavation work can be stored as data at any time.

[0014] 5) At least one embodiment of the present disclosure relates to a face monitoring terminal device, an evaluation target image acquisition unit that acquires an evaluation target image included in a photographed image of the rock face inside the tunnel; a color value acquisition unit that acquires, for each of a plurality of pixels constituting the image to be evaluated, at least the a* value among the L value, a* value, and b* value that indicate the color of the pixel as a color value; an evaluation value acquisition unit that acquires an evaluation value defined in accordance with the color specification value of the pixel; a display control unit that causes a display device to display an evaluation value distribution image in which a color corresponding to the evaluation value is assigned to each of the plurality of pixels; Equipped with.

[0015] The configuration 5) above provides the same technical advantages as the configuration 1). [Effects of the Invention]

[0016] According to the present invention, it is possible to provide a face monitoring program and a face monitoring terminal device that can assist in accurate evaluation of the degree of progress of weathering of the face of a tunnel. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a schematic diagram showing an overview of the present invention. [Figure 2A] FIG. 2 is a schematic diagram of a captured image. [Figure 2B] FIG. 2 is a schematic diagram of a mask image. [Figure 2C] FIG. 2 is a schematic diagram of an image to be evaluated. [Figure 2D] FIG. 10 is a schematic diagram of an evaluation value distribution image. [Figure 2E] FIG. 10 is a schematic diagram of an evaluation result image. [Figure 3] FIG. 1 is a schematic diagram of a two-dimensional space in which the magnitudes of a* and b* values ​​can be expressed. [Figure 4] FIG. 2 is a block diagram showing the electrical configuration of the terminal device. [Figure 5] FIG. 2 is a schematic diagram of color specification value data. [Figure 6] FIG. 10 is a schematic diagram of evaluation value distribution data. [Figure 7] 1 is a flowchart showing a method for determining the degree of progress of weathering. [Figure 8]FIG. 10 is a schematic diagram of a terminal device that displays an evaluation result image. [Figure 9] FIG. 10 is a schematic diagram showing a method for initial setting of a determination unit. [Figure 10] 1 is a flowchart showing a method for evaluating weathering alteration of a working face. [Figure 11] FIG. 10 is a schematic diagram of a two-dimensional space capable of expressing the magnitudes of a* and b* values ​​(modification). DETAILED DESCRIPTION OF THE INVENTION

[0018] Hereinafter, several embodiments of the present disclosure will be described with reference to the accompanying drawings. However, the dimensions, materials, shapes, relative arrangements, etc. of components described as embodiments or shown in the drawings are merely illustrative examples and are not intended to limit the scope of the present disclosure. For example, expressions expressing relative or absolute arrangement such as "in a certain direction," "along a certain direction," "parallel," "orthogonal," "center," "concentric," or "coaxial" not only express such an arrangement exactly, but also express a state in which there is a relative displacement with a tolerance or an angle or distance to the extent that the same function is obtained. For example, expressions such as "identical," "equal," and "homogeneous" that indicate that something is in an equal state not only indicate a state of strict equality, but also indicate a state in which there is a tolerance or a difference to the extent that the same function is obtained. For example, expressions representing shapes such as a square shape or a cylindrical shape not only represent shapes such as a square shape or a cylindrical shape in the strict geometric sense, but also represent shapes including uneven portions, chamfered portions, etc., to the extent that the same effect can be obtained. On the other hand, the expressions "comprise," "include," or "have" one element are not exclusive expressions that exclude the presence of other elements. Note that the same components will be denoted by the same reference numerals and the description thereof will be omitted.

[0019] <Summary of the Invention> As shown in FIG. 1, the present invention utilizes a face monitoring terminal device 5 (hereinafter referred to as the "terminal device 5"), such as a smartphone, to automatically evaluate the weathering of the face 3 of a rock mass 2 inside a tunnel. During the automatic evaluation process, an application installed on the terminal device 5 is configured to use an image I (see FIG. 2A) of the face 3 captured by the terminal device 5 as input data and output an evaluation result image C (see FIG. 2E) showing the evaluation results of the weathering of the face 3. A worker can make a final judgment on the progress of weathering while checking both the evaluation result image C displayed on the terminal device 5 and the actual face 3. Therefore, even a worker with little evaluation experience can accurately evaluate the weathering of the face 3 with the support of the terminal device 5.

[0020] <Evaluation principle of weathering according to the present invention> 2A to 2E show a series of images processed during the automatic evaluation of weathering. FIG. 2A shows a captured image I captured by a terminal device 5. FIG. 2B shows a masked image M obtained by applying a masking process to the captured image I. In this example, the masking process is applied to the captured image I, leaving only the upper half of the working face 3 above the spring line. FIG. 2C shows an evaluation target image E obtained by applying a trimming process to the masked image M to remove the masked area. The working face 3 captured in the evaluation target image E is the subject of evaluation for weathering. Furthermore, the evaluation target image E is divided into three evaluation target areas R corresponding to the top, left shoulder, and right shoulder of the working face 3, and the degree of weathering is evaluated for each evaluation target area R. The evaluation target area R includes two or more of the pixels that make up the evaluation target image E.

[0021] Here, each of the multiple pixels constituting each evaluation target region R is assigned an RGB value derived from the captured image I. In the present invention, a process is executed to convert the RGB value for each pixel into Lab values, which are values ​​in the L*a*b* color space, and the a* and b* values ​​for each pixel are acquired as color values ​​for the evaluation target image E. As weathering of the working face 3 progresses, the reddish, greenish, and yellowish hues of the working face 3 change, so the a* and b* values ​​change more sensitively in response to weathering than RGB values. The inventors have noticed that the a* and b* values ​​are more suitable for evaluating the degree of weathering, and have therefore decided to convert the RGB values ​​into a* and b* values.

[0022] Furthermore, in the present invention, an evaluation value that correlates with the degree of weathering is obtained for each pixel based on the converted a* and b* values. The evaluation value is a discrete value such as "1," "2," "3," or "4," and corresponds to the evaluation criteria (evaluation criteria 1 to 4) disclosed in the tunnel rock mass classification assessment manual posted by a public institution. More specifically, the evaluation values ​​"1," "2," "3," and "4" are defined as follows, with the higher the evaluation value, the worse the degree of weathering. Evaluation value "1": No weathering or alteration, and face 3 is sound. Evaluation value "2": Discoloration along the rock grain, and the strength of face 3 has slightly decreased. Evaluation score: "3": Overall discoloration, and the strength of face 3 has significantly decreased. Rating "4": Face 3 is muddy, clayey, fractured, or unconsolidated from the beginning.

[0023] For the sake of convenience in classifying the evaluation, the evaluation value "4," which indicates a very severe degree of weathering and alteration, includes cases where the face 3 is "unconsolidated from the beginning." This is because the risk of collapse of the face 3 in this case is the same as the risk of collapse when the face 3 becomes muddy or clayey over time, or when it fractures.

[0024] In the following, the evaluation values ​​"1," "2," "3," and "4" will be referred to as the "first evaluation value," "second evaluation value," "third evaluation value," and "fourth evaluation value," respectively, and may be referred to collectively as "evaluation value" when not distinguishing between them.

[0025] The following explains which evaluation value is assigned to each pixel depending on the a* and b* values. If the face 3 is muddy, clayey, fractured, or unconsolidated, the red and yellow hues of the face 3 will be very strong. Conversely, if the face 3 is sound, the red, green, and yellow hues of the face 3 will be weak. Therefore, for a specific pixel within the evaluation area R, if the a* and b* values ​​are large, the degree of weathering and alteration of the face 3 within the pixel can be considered to be poor, and if the values ​​are small, the degree of weathering and alteration can be considered to be sound. In other words, the larger the a* and b* values, the more likely it is that the fourth evaluation value will be assigned, and the smaller the values, the more likely it is that the first evaluation value will be assigned.

[0026] The criteria for assigning the first to fourth evaluation values ​​according to the a* and b* values ​​will be explained in more detail. Fig. 3 shows a two-dimensional space Da in which the magnitudes of the a* and b* values ​​can be expressed using two mutually perpendicular coordinate axes. The horizontal axis of the figure represents the magnitude of the a* value, and the vertical axis represents the magnitude of the b* value. The two-dimensional space Da of the present invention is a space in which the a* value is equal to or greater than 0 and the b* value is equal to or greater than 0.

[0027] The two-dimensional space Da is composed of a first specific region A1, a second specific region A2, a third specific region A3, and a fourth specific region A4. The first specific region A1 is bounded by the horizontal axis, the vertical axis, and a first boundary line L1. The second specific region A2 is bounded by the horizontal axis, the vertical axis, the first boundary line L1, and the second boundary line L2. The third specific region A3 is bounded by the horizontal axis, the vertical axis, the second boundary line L2, and the third boundary line L3, and the fourth specific region A4 is bounded by the horizontal axis, the vertical axis, and the third boundary line L3. The first specific region A1 and the second specific region A2 are adjacent to each other across the first boundary line L1. The second specific region A2 and the third specific region A3 are adjacent to each other across the second boundary line L2. The third specific region A3 and the fourth specific region A4 are adjacent to each other across the third boundary line L3.

[0028] In the present invention, for each pixel, it is determined which of the first to fourth specific regions A1 to A4 the coordinate point indicating the magnitude of the a* and b* values ​​is contained in. If the coordinate point is contained in the first specific region A1, the pixel is deemed to be a pixel (first specific pixel) having a first evaluation value. If the coordinate point is contained in the second specific region A2, the pixel is deemed to be a pixel (second specific pixel) having a second evaluation value. If the coordinate point is contained in the third specific region A3, the pixel is deemed to be a pixel (third specific pixel) having a third evaluation value. If the coordinate point is contained in the fourth specific region A4, the pixel is deemed to be a pixel (fourth specific pixel) having a fourth evaluation value.

[0029] The first boundary line L1, the second boundary line L2, and the third boundary line L3 are all expressed by b* = -A × a* + B (A > 0, B > 0). The first boundary line L1, the second boundary line L2, and the third boundary line L3 are all inclined so that their acute angles (α1, α2, α3) relative to the horizontal axis are greater than 45° and less than 90°. As a result, the first boundary line L1, the second boundary line L2, and the third boundary line L3 have a steep slope relative to the horizontal axis. As the a* value of a pixel increases, the coordinate point immediately shifts toward the fourth specific region A4 (arrow Q). As a result, as the reddishness of the face 3 increases, the evaluation value increases, indicating that the degree of weathering deterioration is quickly worsening (a modified example of Figure 8 will be described later). While the first boundary line L1, the second boundary line L2, and the third boundary line L3 are depicted parallel to one another, the present invention is not limited to this configuration.

[0030] Although the evaluation values ​​may be displayed on the terminal device 5 as specific numerical values ​​such as "1" to "4," displaying them in colors corresponding to the evaluation values ​​allows workers to visually understand them. Therefore, in the present invention, an evaluation value distribution image F (see FIG. 2D ) is generated in which a color corresponding to an evaluation value is assigned to each of a plurality of pixels constituting the evaluation target image E. The evaluation value distribution image F is obtained by superimposing a distribution image in which colors corresponding to the evaluation values ​​are distributed on the evaluation target image E in which the working face 3 is displayed. In other words, the evaluation target image E includes a first layer in which the evaluation target image E is displayed and a second layer in which a color distribution image is displayed, and the second layer is disposed on the first layer. The colors corresponding to the evaluation values ​​may be any color expressed, for example, by RGB values. If the evaluation values ​​are expressed in four levels from "1" to "4," four colors are prepared for each evaluation value.

[0031] The evaluation target image E shown in FIG. 2D indicates the degree of weathering for each pixel, but it is necessary to evaluate the degree of weathering for the entire working face 3. Therefore, the degree of weathering is calculated for each evaluation target area R based on the distribution of evaluation values ​​in each evaluation target area R. In other words, the degree of weathering in the evaluation target area R is calculated based on the proportion of the first, second, third, and fourth specific pixels within the evaluation target area R. FIG. 2E shows an evaluation result image C displaying the evaluation results of the weathering degree. The evaluation result image C is an image in which an object Ob indicating the evaluation result of the weathering degree is superimposed on the evaluation value distribution image F. In other words, the evaluation target image E includes, in addition to the first and second layers described above, a third layer containing the object Ob, and the third layer is arranged on the second layer. The worker compares the evaluation result image C with the actual face 3 and makes a final judgment about the weathering and alteration of the face 3.

[0032] As described above, as the weathering of the rock 2 progresses, the redness of the rock 2 increases, and as the alteration of the rock 2 progresses, the greenness and yellowness increase. In this regard, according to the above configuration, of the L value, a* value, and b* value that indicate the color of the evaluation target image E, the a* value, which changes in accordance with the redness and greenness, and the b* value, which changes in accordance with the yellowness, are used to evaluate the progress of weathering alteration of the working face 3. This allows for a more accurate evaluation of the progress of weathering alteration of the working face 3.

[0033] <Details of the configuration of the terminal device 5> The configuration of the terminal device 5 will be described in detail with reference to Figs. 4 to 9. Fig. 4 is a block diagram showing the electrical configuration of the terminal device 5. The terminal device 5 comprises a camera 6, a display 7, an operation unit 8, and a control unit 30 electrically connected to these three components. The display 7, which is an example of a display device, may be a touch panel display, that is, the operation unit 8 may be incorporated into the display 7. The control unit 30 comprises a processor and a storage device that stores data processed by the processor.

[0034] The control unit 30 functions as a shooting control unit 31, an evaluation target image acquisition unit 32, a color value acquisition unit 33, an evaluation value acquisition unit 34, an evaluation value distribution image acquisition unit 35, a progress acquisition unit 36, a display control unit 37, a progress update control unit 38, a data transmission unit 39, and a boundary line update control unit 40.

[0035] The shooting control unit 31 sends a shooting command to the camera 6 and acquires the captured image I (see FIG. 2A) generated by the camera 6. The captured image I is sent to the evaluation target image acquisition unit 32. The evaluation target image acquisition unit 32 generates a mask image M (see FIG. 2B) by performing a mask process on the captured image I, and also generates an evaluation target image E (see FIG. 2C) by performing a trimming process on the mask image M. The evaluation target image E is sent to the color value acquisition unit 33.

[0036] The colorimetric value acquisition unit 33 acquires the a* and b* values ​​from the L, a*, and b* values ​​for each of the multiple pixels that make up the evaluation target image E. Specifically, the L, a*, and b* values ​​are acquired by applying a predetermined calculation formula to the RGB values ​​assigned to each pixel, and then the L value is excluded from these three values, and the remaining a* and b* values ​​are acquired as colorimetric values. As a result, the colorimetric value data 22 shown in FIG. 5 is acquired. The colorimetric value data 22 indicates the a* and b* values ​​assigned to each of the multiple pixels that make up the evaluation target image E. The colorimetric value data 22 acquired by the colorimetric value acquisition unit 33 is sent to the evaluation value acquisition unit 34.

[0037] The evaluation value acquisition unit 34 acquires an evaluation value for each of the plurality of pixels constituting the evaluation target image E based on the color specification value data 22. Which of the first to fourth evaluation values ​​is assigned to each pixel is determined by a determination unit 321 incorporating a predetermined algorithm of the evaluation value acquisition unit 34. The determination unit 321 determines which of the first to fourth specific areas A1 to A4 (see FIG. 3) the coordinate point indicated by the a* value and b* value for each of the plurality of pixels indicated by the color specification value data 22 is included in. Based on the determination result of the determination unit 321, evaluation value distribution data 24 (see FIG. 6) indicating which evaluation value is assigned to each of the plurality of pixels is generated. The evaluation value distribution data 24 acquired by the evaluation value acquisition unit 34 is sent to the evaluation value distribution image acquisition unit 35.

[0038] The evaluation value distribution image acquisition unit 35 generates an evaluation value distribution image F (see FIG. 2D ) based on the evaluation value distribution data 24 and the evaluation target image E. More specifically, a distribution image in which a color corresponding to an evaluation value is assigned to each pixel is generated based on the evaluation value distribution data 24. The evaluation value distribution image F is generated by placing a second layer including this distribution image on a first layer including the evaluation target image E. The evaluation value distribution image F acquired by the evaluation value distribution image acquisition unit 35 is sent to the progress acquisition unit 36.

[0039] The progress level acquisition unit 36 ​​acquires the progress level of weathering of the working face 3 shown in the evaluation target image E based on the evaluation value distribution image F. In this example, the progress level of weathering is acquired for each of the three evaluation target regions R into which the evaluation target image E is divided, based on the distribution of evaluation values. FIG. 7 is a schematic diagram illustrating an example of the progress level determination flow. As shown in the figure, if there is at least one fourth specific pixel in each evaluation target region R, a progress level of "4" is assigned to that evaluation target region R. If there is no fourth specific pixel and at least one third specific pixel, a progress level of "3" is assigned. If there is no third specific pixel and the second specific pixel occupies 20% or more of the evaluation target region R, a progress level of "2" is assigned. Otherwise, i.e., if the first specific pixel occupies more than 80% of the evaluation target region R, a progress level of "1" is assigned. The progress level of weathering for each evaluation target region R is sent to the display control unit 37.

[0040] The display control unit 37 generates an evaluation result image C (see FIG. 2E) based on the degree of progress sent from the progress acquisition unit 36 ​​and the evaluation value distribution image F. More specifically, the evaluation result image C is generated by placing a third layer including an object Ob indicating the degree of progress for each evaluation target region R on the second layer of the evaluation value distribution image F. The display control unit 37 further sends a display command to the display 7 to generate the evaluation result image C. As a result, the evaluation result image C is displayed on the display 7 of the terminal device 5, as shown in FIG. 8. The evaluation result image C displays the degree of weathering for each evaluation target region R, allowing the degree of progress of the working face 3 to be evaluated in more detail.

[0041] The worker compares the evaluation result image C displayed on the display 7 with the actual working face 3, and makes a final judgment on the degree of weathering and alteration of the working face 3 for each evaluation target area R. If at least one of the three degrees of progress displayed in the evaluation result image C differs from the final degree of progress judged, the worker operates the operation unit 8 based on his or her own judgment to update the degree of progress displayed in the evaluation result image C.

[0042] In this case, the progress update control unit 38 shown in FIG. 3 updates the progress to be updated in the evaluation result image C to the progress updated by the worker. The updated progress is displayed on the display 7. Furthermore, the progress for each evaluation target region R, including the updated progress, is sent to the boundary line update control unit 40. Details of the boundary line update control unit 40 will be described later. Furthermore, the captured image I and the evaluation result image C (or the updated evaluation result image C if updated) are sent to a data transmission unit 39. The data transmission unit 39 transmits correspondence data associating these to a first computer 41 located away from the terminal device 5. The first computer 41 may be a cloud.

[0043] According to a configuration in which the progress degree acquisition unit 36 ​​acquires (evaluates) the progress degree based on the proportions of the first specific pixel, the second specific pixel, the third specific pixel, and the fourth specific pixel in each evaluation target region R of the evaluation target image E, the proportions of these specific pixels become the evaluation criteria for weathering, making it possible to more objectively evaluate the progress of weathering. Furthermore, the multiple pixels that make up the evaluation target image E are classified into multiple categories, such as the first specific pixel, the second specific pixel, the third specific pixel, and the fourth specific pixel, based on their color values. This allows for a more accurate evaluation of the progress of weathering of the working face 3.

[0044] <Initial setting of algorithm of determination unit 321> With reference to FIG. 9, the initial setting of the algorithm of the determination unit 321 will be described. The initial setting is performed by a second computer 42 set in a location remote from the terminal device 5. The second computer 42 performs the initial setting using a correct answer data group 26. The correct answer data group 26 includes multiple sets of correct answer data 25, each of which includes a correct answer image Im, which is an image of the tunnel face 3 photographed during past tunnel construction work, and a correct answer progress degree, which is the progress of weathering and alteration of the tunnel face 3 captured in the correct answer image Im. Here, the correct answer image Im is assigned a correct answer color value indicating the color of the correct answer image Im for each of its multiple pixels. The correct answer color value may be an RGB value, but is preferably an Lab value. The correct answer progress degree reflects the results of observation of the tunnel face 3 by a skilled worker.

[0045] The second computer 42 applies optimization analysis to the group of correct answer data 26. For the optimization analysis, one of the black-box optimization methods, Bayesian optimization, Monte Carlo method, or reinforcement learning, is used. By applying the optimization analysis, the first boundary line L1, the second boundary line L2, and the third boundary line L3 described above are initially set. Data relating to the initially set first boundary line L1, the second boundary line L2, and the third boundary line L3 is transmitted from the second computer 42 to the determination unit 321, completing the initial setting.

[0046] According to the above configuration, the first boundary line L1, the second boundary line L2, and the third boundary line L3 are initially set according to the correspondence between the correct answer image Im and the correct answer progress degree included in the correct answer data 25. The higher the reliability of the correct answer data 25, the more the initially set first boundary line L1, the second boundary line L2, and the third boundary line L3 can contribute to highly accurate evaluation of the progress degree of weathering and alteration.

[0047] The boundary line update control unit 40 described above sends the evaluation target image E and the updated progress level to the second computer 42 as new, additional supervised answer data 25. The second computer 42 may obtain color values ​​for each of a plurality of pixels based on the evaluation target image E. The second computer 42 uses the supervised answer data 25 to perform optimization analysis again and update the first boundary line L1, the second boundary line L2, and the third boundary line L3. The data related to these updated boundary lines is sent from the second computer 42 to the determination unit 321, and the determination algorithm of the determination unit 321 is updated.

[0048] According to the above configuration, the first boundary line L1, the second boundary line L2, and the third boundary line L3 are updated based on the update progress rate updated by the worker's judgment, and the first specific area A1, the second specific area A2, the third specific area A3, and the fourth specific area A4 are updated. As a result, the first specific area A1 to the fourth specific area A4 can be updated to be more suitable for accurate evaluation through an evaluation of weathering and alteration of the tunnel face 3 in the tunnel construction work currently underway. After the update, when the terminal device 5 again evaluates the tunnel face 3 shown in the evaluation target image E, an even more accurate evaluation is possible.

[0049] <Weathering evaluation method> A method for evaluating weathering of the working face 3 will be described with reference to Figures 4 and 10. Figure 10 is a flowchart of the evaluation method for the working face 3. This flowchart shows the flow of control processing executed by the processor of the terminal device 5 reading out a working face evaluation program (work face monitoring program) stored in the storage device of the terminal device 5. Hereinafter, "step" may be abbreviated as "S".

[0050] First, an operator operates the operation unit 8 of the terminal device 5, causing the processor to acquire a photographed image I of the working face 3 (S11). The processor that executes S11 is an example of the photography control unit 31. Next, the processor executes a step of executing a step of executing a masking process and a trimming process on the photographed image I acquired in S11, thereby acquiring an evaluation target image E (S13). The processor that executes S13 is an example of the evaluation target image acquisition unit 32.

[0051] Next, the processor acquires the a* value and the b* value as color specification values ​​for each of the plurality of pixels constituting the evaluation target image E acquired in S13 (S15). The processor that executes S15 is an example of the color specification value acquisition unit 33. Next, the processor acquires an evaluation value defined according to the color specification values ​​for each of the plurality of pixels (S17). The processor that executes S17 is an example of the evaluation value acquisition unit 34.

[0052] Next, the processor acquires an evaluation value distribution image F based on the evaluation value for each pixel acquired in S17 (S19). The processor executing S19 is an example of the evaluation value distribution image acquisition unit 35. Next, the processor acquires (evaluates) the degree of weathering of the working face 3 shown in the evaluation target image E based on the evaluation value distribution image F acquired in S19 (S21). In S21, the degree of weathering is acquired for each evaluation target region R in the evaluation target image E. The processor executing S21 is an example of the progress acquisition unit 36. Next, the processor generates an evaluation result image C and displays the evaluation result image C on the display 7 (S23). The processor executing S23 is an example of the display control unit 37. After S23, the worker can finally determine the degree of weathering of the working face 3 by visually comparing the evaluation result image C shown on the display 7 with the actual working face 3.

[0053] Next, if the worker determines that it is necessary to update the degree of progress in the evaluation result image C, he or she inputs a predetermined operation to the operation unit 8. In this case, the processor updates the degree of progress shown in the evaluation result image C (S25). The updated result is reflected in the evaluation result image C. The processor that executes S25 is an example of the progress update control unit 38.

[0054] Next, the processor updates the boundary lines (first boundary line L1, second boundary line L2, third boundary line L3) in accordance with the operation content of the worker in S25 (S27). More specifically, the processor associates the evaluation target image E acquired in S13 with the progress degree updated in S25 and sends them to the second computer 42. The second computer 42 performs the optimization analysis again, updates the boundary lines, and returns the updated boundary lines to the processor of the terminal device 5. This allows the processor to update the boundary lines incorporated in the determination model used in S17. The processor that executes S27 is an example of the boundary line update control unit 40.

[0055] If the progress level shown in the evaluation result image C displayed in S23 is the same as the progress level finally determined by the worker, the progress level in the evaluation result image C is not updated and S25 and S27 are skipped.

[0056] Next, the processor transmits correspondence data that associates the evaluation target image E acquired in S13 with the evaluation result image C acquired in S21 to the first computer 41 (S29). The processor that executes S29 is an example of the data transmission unit 39. Note that if the progress level is updated in S25, the updated evaluation result image C will be included in the correspondence data. Thereafter, the processor terminates this flowchart.

[0057] The evaluation result image C displayed in S23 includes an evaluation value distribution image F in which a color corresponding to an evaluation value is assigned to each of a plurality of pixels. The evaluation value distribution image F, which is determined based on the a* and b* values, accurately indicates the degree of weathering of the working face 3, and can contribute to the accurate judgment of workers who visually check the actual degree of weathering of the working face 3. This can therefore support workers in accurately evaluating the progress of weathering.

[0058] In addition, in S23, an evaluation result image C is displayed in which the degree of weathering is superimposed on the evaluation value distribution image F. This configuration allows the worker to immediately visually recognize the mechanical determination result of the degree of weathering, thereby supporting the worker in accurately evaluating the degree of weathering.

[0059] Furthermore, in S29, correspondence data associating the evaluation target image E with the evaluation result image C is transmitted to the first computer 41. According to this configuration, the face 3 observed in the course of the tunnel excavation work can be stored as data at any time.

[0060] <Modification> At least one of the following modifications may be applied to the above-described embodiment.

[0061] The terminal device 5 may be any portable computer equipped with a processor and a storage device for storing data processed by the processor, and may be a tablet or a mobile phone instead of the above-mentioned smartphone.

[0062] In the above embodiment, both the a* value and the b* value are used as the color values ​​of the evaluation target image E, but the evaluation value may be obtained based on the a* value alone. In this case, the a* value is not limited to values ​​equal to or greater than 0, and may be a negative value. Even when the evaluation value is obtained based on the a* value alone, it is possible to evaluate the degree of weathering of the working face 3.

[0063] The first boundary line L1, the second boundary line L2, and the third boundary line L3 shown in FIG. 8 may be replaced with the first boundary line M1, the second boundary line M2, and the third boundary line M3 shown in FIG. 11, for example. The first boundary line M1, the second boundary line M2, and the third boundary line M3 are expressed by b* = -A × a* + B (A > 0, B > 0), as in FIG. 8. Here, the first boundary line M1, the second boundary line M2, and the third boundary line M3 are inclined so that their acute angles (β1 to β3) relative to the horizontal axis are greater than 0° and less than 45°. As a result, the first boundary line M1, the second boundary line M2, and the third boundary line M3 have a steep slope relative to the vertical axis. Even if the b* value of a pixel increases slightly, the coordinate point immediately shifts toward the fourth specific region A4 (arrow S). As a result, as the yellowish color of the face 3 increases, the evaluation value increases, indicating that the degree of weathering deterioration is immediately worsening.

[0064] <Additional Notes> The following are examples of the embodiments disclosed herein.

[0065] 1) A face evaluation method according to at least one embodiment of the present invention, an evaluation target image acquisition step of acquiring an evaluation target image included in a photographed image of a rock face inside a tunnel; a color value acquisition step of acquiring, for each of a plurality of pixels constituting the image to be evaluated, at least the a* value among the L value, a* value, and b* value indicating the color of the pixel as a color value; a progress evaluation step of evaluating a progress of weathering of the face of the stonework shown in the evaluation target image based on the color value of each of the plurality of pixels; Equipped with.

[0066] According to the inventor's findings, as rock weathering progresses, the rock becomes redder, and as rock alteration progresses, the rock becomes greener. In this regard, according to the configuration of 1) above, of the L value, a* value, and b* value that indicate the color of the image to be evaluated, at least the a* value, which changes depending on the redness and greenness, is used to evaluate the progress of weathering and alteration of the face. This allows for a more accurate evaluation of the progress of weathering and alteration of the face.

[0067] 2) In some embodiments, the face evaluation method described in 1) above, In the color specification value acquisition step, the a* value and the b* value are acquired as the color specification values.

[0068] According to the inventor's findings, as the alteration of the rock mass progresses, the rock mass also becomes more yellowish. In this regard, the configuration of 2) above uses not only the a* value but also the b* value, which changes depending on the yellowness, to evaluate the degree of weathering of the rock face. This allows for a more accurate evaluation of the degree of weathering of the rock face captured in the image to be evaluated.

[0069] 3) In some embodiments, the face evaluation method described in 2) above includes: further comprising an evaluation value acquisition step of acquiring, for each of the plurality of pixels, an evaluation value defined in accordance with the a* value and the b* value included in the color specification value of the pixel; the evaluation values ​​include a first evaluation value indicating that a coordinate point indicating the magnitude of the a* value and the b* value included in the color specification values ​​is included within a first specific region of a two-dimensional space in which the magnitudes of the a* value and the b* value can be expressed using two coordinate axes that are orthogonal to each other; In the progress evaluation step, the progress of the weathering of the face is evaluated based on the proportion of the pixels whose evaluation value is the first evaluation value within the image to be evaluated.

[0070] According to the configuration of 3) above, a first specific pixel, which is a pixel having a first evaluation value, is identified from among the plurality of pixels, and the degree of weathering is evaluated according to the proportion of the first specific pixel in the image to be evaluated. Because the proportion of the first specific pixel is included in the evaluation criteria for weathering, it is possible to evaluate the degree of weathering more objectively.

[0071] 4) In some embodiments, the face evaluation method described in 3) above, The first boundary line in the two-dimensional space that defines the first specific area is initially set by optimization analysis using correct data that includes a correct image, which is an image of the rock face taken during past tunnel construction work, and the correct progress of the weathering and alteration of the face shown in the correct image.

[0072] According to the above configuration 4), the first specific area is initially set according to the correspondence between the teacher image included in the teacher data and the teacher progress level. The higher the reliability of the teacher data, the more the initially set first specific area can contribute to highly accurate evaluation of the progress level of weathering and alteration.

[0073] 5) In some embodiments, the face evaluation method described in 4) above includes: a progress update step of updating the progress acquired by the progress evaluation step to an updated progress based on human judgment; a boundary line updating step of updating the first boundary line based on the optimization analysis using the update progress and the evaluation target image whose progress has been updated; Equipped with.

[0074] The first boundary line and the first specific area are updated based on the update progress rate, which is updated based on human judgment. As a result, through the evaluation of weathering and alteration of the tunnel face in the ongoing tunnel construction, the first specific area used for the evaluation can be updated to one suitable for accurate evaluation. Therefore, after the boundary line update step, when evaluating the face shown in the evaluation target image, a more accurate evaluation can be performed.

[0075] 6) In some embodiments, the face evaluation method described in 3) above, the evaluation values ​​include a second evaluation value indicating that the coordinate point is included in a second specific area in the two-dimensional space that is different from the first specific area, In the progress evaluation step, the progress of the weathering is evaluated based on the proportion of the pixels in the image to be evaluated whose evaluation value is the first evaluation value, and the proportion of the pixels in the image to be evaluated whose evaluation value is the second evaluation value.

[0076] According to the above-mentioned configuration 6), the pixels constituting the image to be evaluated are classified in more detail according to the evaluation value determined by the color value of each pixel. This allows for a more accurate evaluation of the progress of weathering of the face.

[0077] 7) In some embodiments, the face evaluation method according to any one of 1) to 6) above, In the progress evaluation step, the progress for each of a plurality of evaluation target regions that divide the evaluation target image is evaluated based on the color value for each of two or more of the pixels included in the evaluation target region.

[0078] According to the above configuration 7), the degree of weathering is evaluated for each of the multiple evaluation target regions that divide the evaluation target image, thereby enabling a more detailed evaluation of the degree of weathering on the face.

[0079] 8) A face evaluation program according to at least one embodiment of the present disclosure, On the computer, an evaluation target image acquisition step of acquiring an evaluation target image included in a photographed image of a rock face inside a tunnel; a color value acquisition step of acquiring, for each of a plurality of pixels constituting the image to be evaluated, at least the a* value among the L value, a* value, and b* value indicating the color of the pixel as a color value; a progress evaluation step of evaluating a progress of weathering of the face of the stonework shown in the evaluation target image based on the color value of each of the plurality of pixels; Execute the following.

[0080] According to the configuration 8) above, the same technical advantages as those in 1) above can be obtained. [Explanation of symbols]

[0081] 2: Bedrock 3: Face 5: Terminal device (face monitoring terminal device) 6: Camera 7: Display (display device) 8 :Operation section 22: Color value data 24: Evaluation value distribution data 25: Correct data 26: Correct data set 30: Control section 31: Shooting control unit 32: Evaluation target image acquisition unit 33: Color value acquisition unit 34: Evaluation value acquisition unit 35: Evaluation value distribution image acquisition unit 36: Progress acquisition part 37: Display control section 38: Progress update control unit 39: Data transmission unit 40: Boundary line update control section 41: First computer 42: Second computer 321: Judgment section A1: 1st specific area A2:Second specific area A3:Third specific area A4: 4th specific area C: Evaluation result image Da: 2-dimensional space E: Image to be evaluated F: Evaluation value distribution image I: Photographed image Im: Correct image L1,M1: 1st boundary line L2,M2: 2nd boundary line L3,M3: 3rd boundary line M: Mask image Ob : Object Q, S: Arrows R: Area to be evaluated

Claims

1. On the computer, an evaluation target image acquisition step of acquiring an evaluation target image included in a photographed image of a rock face inside a tunnel; a color value acquisition step of acquiring, for each of a plurality of pixels constituting the image to be evaluated, at least the a* value among the L value, a* value, and b* value indicating the color of the pixel as a color value; an evaluation value acquisition step of acquiring an evaluation value defined in accordance with the color specification value of the pixel; an image display step of displaying, on a display device, an evaluation value distribution image in which a color corresponding to the evaluation value is assigned to each of the plurality of pixels; A face monitoring program that executes the above.

2. The computer, a progress degree acquisition step of acquiring a progress degree of weathering of the face of the workpiece shown in the evaluation target image based on the evaluation value distribution image; Then run In the image display step, an evaluation result image generated by superimposing the progress level on the evaluation value distribution image is displayed on the display device. The face monitoring program according to claim 1.

3. In the progress degree acquisition step, the progress degree for each of a plurality of evaluation target regions that divide the evaluation target image is acquired based on the color value for each of two or more of the pixels included in the evaluation target region; In the image display step, the progress level for each of the evaluation target regions is superimposed on the evaluation distribution image and displayed on the display device. The face monitoring program according to claim 2.

4. The computer, a data transmission step of transmitting correspondence data that associates the evaluation target image with the evaluation result image to a first computer located away from the computer; Run the 4. The face monitoring program according to claim 2 or 3.

5. an evaluation target image acquisition unit that acquires an evaluation target image included in a photographed image of the rock face inside the tunnel; a color value acquisition unit that acquires, for each of a plurality of pixels constituting the image to be evaluated, at least the a* value among the L value, a* value, and b* value that indicate the color of the pixel as a color value; an evaluation value acquisition unit that acquires an evaluation value defined in accordance with the color specification value of the pixel; a display control unit that causes a display device to display an evaluation value distribution image in which a color corresponding to the evaluation value is assigned to each of the plurality of pixels; A face monitoring terminal device equipped with:

Images