Calculation program and calculation method

Abstract

To provide a calculation program and a calculation method which calculate a thickness of a tree from an image.SOLUTION: A calculation program allows a computer to execute processing of determining a first region of a tree and a second region of a tape horizontally stuck to the tree, which are included in an image, on the basis of a learned model for determining the first region and the second region to calculate the thickness of the tree on the basis of the relation between a horizontal width corresponding to a short direction of the first region and a vertical width of the second region.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a calculation program and a calculation method. [Background technology]

[0002] There is a known technology that can continuously measure the actual sizes of multiple subjects contained in a captured image by analyzing the captured image while comparing the measurement target, which is the subject of the captured image, with a reference target (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-181035 Summary of the Invention [Problem to be solved by the invention]

[0004] When erecting towers or laying access roads in mountainous areas, unnecessary trees that obstruct the construction or laying of the work are sometimes cut down. When cutting down trees, a report is prepared that records the tree's diameter (specifically, diameter at breast height), type of tree, location, etc., and this report is used to estimate compensation to landowners in the mountainous area and the costs incurred by cutting down the trees.

[0005] When preparing the records, workers must manually measure each tree using a tool called a ruler and record the tree's diameter, which is time-consuming and labor-intensive. This can be especially time-consuming if there are thousands of trees to measure.

[0006] Therefore, in one aspect, an object is to provide a calculation program and a calculation method for calculating the thickness of a tree from an image. [Means for solving the problem]

[0007] In one embodiment, the calculation program is 、 tree is the entire area The first area and the tree are pasted horizontally. rectangular tape is the entire area and a second region based on a trained model for discriminating between the first region and the second region, and based on the relationship between the width corresponding to the short side direction of the first region and the length of the second region, trunk diameter at breast height The computer is caused to execute the process of calculating the above. [Effects of the Invention]

[0008] The thickness of the tree can be calculated from the image. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 shows an example of the hardware configuration of the imaging device. [Figure 2] FIG. 2 shows an example of the functional configuration of the imaging device. [Figure 3] Figure 3 is an example of the record information. [Figure 4] FIG. 4 is a flowchart showing an example of processing executed by the imaging device. [Figure 5] FIG. 5 is a diagram illustrating an example of image acquisition by the imaging device. [Figure 6] Figure 6(a) is an example of a first screen displayed on the display unit, and Figure 6(b) is an example of a second screen displayed on the display unit. [Figure 7] 7(a) is an example of a third screen displayed on the display unit, and FIG. 7(b) is an example of a fourth screen displayed on the display unit. [Figure 8] FIG. 8 is a flowchart showing an example of the area discrimination process. [Figure 9] FIG. 9 is a flowchart showing an example of the first pixel number counting process. [Figure 10] FIG. 10 is a flowchart showing an example of the second pixel number counting process. [Figure 11] FIG. 11 is a flowchart showing an example of the area detection process. [Figure 12] FIG. 12 is a flowchart showing an example of the thickness calculation process. [Figure 13] FIG. 13 is a flowchart showing an example of the display process. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] First, the hardware configuration of the image capturing device 100 that executes the calculation method will be described with reference to Fig. 1. As shown in Fig. 1, the image capturing device 100 includes a CPU (Central Processing Unit) 100A as a processor, and RAM (Random Access Memory) 100B, ROM (Read Only Memory) 100C, and NVM (Non-Volatile Memory) 100D as memories.

[0012] The photographing device 100 also includes an RF (Radio Frequency) circuit 100E, an acceleration sensor 100F, and a camera 100G. An antenna 100E' is connected to the RF circuit 100E. A CPU (not shown) that realizes a communication function may be used instead of the RF circuit 100E. The camera 100G includes an image sensor such as a CMOS (Complementary Metal Oxide Semiconductor) or a CCD (Charge Coupled Device).

[0013] Furthermore, the photographing device 100 includes a touch panel 100H, a display 100I, and a speaker 100J. The CPU 100A to the speaker 100J are connected to one another via an internal bus 100K. That is, the photographing device 100 can be realized by a smart device such as a smartphone or a tablet terminal, or a computer including a PC (Personal Computer).

[0014] The calculation program stored in the ROM 100C or the NVM 100D is temporarily stored in the RAM 100B by the CPU 100A. By executing the stored calculation program, the CPU 100A realizes various functions to be described later and executes various processes to be described later. The calculation program may be one that corresponds to the flowchart to be described later.

[0015] The functional configuration of the image capturing device 100 will be described with reference to Fig. 2. Note that Fig. 2 shows the main functional parts of the image capturing device 100.

[0016] As shown in FIG. 2, the photographing device 100 includes a memory unit 110, a processing unit 120, an input unit 130, a display unit 140, an imaging unit 150, and a detection unit 160. The memory unit 110 can be realized by one or both of the RAM 100B and NVM 100D described above. The processing unit 120 can be realized by the CPU 100A described above. The input unit 130 can be realized by the touch panel 100H described above. The display unit 140 can be realized by the display 100I described above. The imaging unit 150 can be realized by the camera 100G described above. The detection unit 160 can be realized by the acceleration sensor 100F described above.

[0017] Therefore, the memory unit 110, processing unit 120, input unit 130, display unit 140, imaging unit 150, and detection unit 160 are connected to one another. The memory unit 110 includes a model memory unit 111 and a record memory unit 112. The processing unit 120 includes a discrimination unit 121 and a calculation unit 122.

[0018] The model storage unit 111 stores trained models. As will be described in detail later, trained models are information for distinguishing between a tree region (hereinafter referred to as a first region) that fits within an image and a region of tape attached horizontally to the tree (hereinafter referred to as a second region). Specifically, the tree is the trunk, which is a part of the tree. The trained model is generated by inputting, as training data, an image including the tree and tape and an image in which the first region in the image is filled in red and the second region is filled in green into a known machine learning model. The trained model is generated in advance and stored in the model storage unit 111.

[0019] The record storage unit 112 stores record information corresponding to a record. As shown in FIG. 3, the record information includes items such as an investigation number, a tape number, tree diameter, and tree type. An identification number assigned to each tree under investigation is registered as the investigation number. A number written on a tape affixed to the tree is registered as the tape number. If the tape number and information on the tree's position are associated in advance and managed within the imaging device 100, the position of the tree to which the tape is affixed can be identified based on the tape number. The diameter at breast height of the trunk is registered as the tree diameter. The tree diameter is calculated by the calculation unit 122. For example, acacia, chestnut, or other tree types are registered as the tree type.

[0020] The discrimination unit 121 discriminates between a first region and a second region that are respectively contained within the image based on a trained model. After discriminating between the first region and the second region, the discrimination unit 121 displays the first region and the second region in a distinguishable manner. For example, the discrimination unit 121 displays the first region and the second region by filling them in with different colors so that they do not overlap each other. This makes it possible to avoid color mixing in the overlapping portion of the first region and the second region. After displaying, if the discrimination unit 121 detects a first instruction to re-discriminate between the first region and the second region, the discrimination unit 121 discriminates between the first region and the second region again. In addition, the discrimination unit 121 executes various other processes, the details of which will be described later.

[0021] When the calculation unit 122 detects a second instruction requesting calculation of the tree's thickness, it calculates the tree's thickness based on the relationship between the width corresponding to the short side of the first region and the vertical width of the second region. As described above, the first region is the tree region, but by limiting it to the short side, it is possible to represent the width corresponding to the tree's diameter at breast height. As described above, the second region is the tape region. When attaching the tape to the tree, the length of the tape varies depending on the position where it is cut. However, since the vertical width of the tape is constant and known before the inspection, the thickness of the tree can be calculated from the vertical width of the tape by using the ratio between the first number of pixels in the tree's width and the second number of pixels in the tape's vertical width. The calculation unit 122 also performs various other processes, the details of which will be described later.

[0022] Next, an overview of the processing executed by the image capturing device 100 will be described with reference to FIGS.

[0023] First, as shown in Fig. 4, the discrimination unit 121 acquires an image (step S1). For example, as shown in Fig. 5, the worker 10 points an imaging unit (not shown) in the direction of one of the trees 21 standing in a mountainous area and captures an image of the tree 21 including the portion to which the tape 22 is attached. This causes the discrimination unit 121 to acquire an image and start a loop process of displaying the acquired image on the display unit 140.

[0024] Furthermore, when an image is acquired, the discrimination unit 121 displays the photographing reference image 30, a first button image B1, and a second button image B2 on the display unit 140, as shown in FIG. 6(a). The photographing reference image 30 includes a rectangular frame image 31 shown with dashed lines and a photographing center image 32 shown with solid lines. The photographing center image 32 has a shape in which a straight line penetrates a circle that fits within the rectangular frame image 31 in the short direction of the rectangular frame image 31, the straight line being longer than the short direction of the rectangular frame image 31. The worker 10 adjusts the position and tilt angle of the photographing device 100 so that the photographing center image 32 aligns with the portion of the tape 22. The first button image B1 is a button for giving the first instruction described above, and the second button image B2 is a button for giving the second instruction described above.

[0025] When the image is acquired, the discrimination unit 121 executes an area discrimination process (step S2). More specifically, when it is determined that the photographed center image 32 matches the portion of the tape 22, the discrimination unit 121 executes the area discrimination process. In the area discrimination process, the discrimination unit 121 distinguishes between a first area and a second area based on the trained model stored in the model storage unit 111, and displays the first area filled in red and the second area filled in green. Details of the area discrimination process will be described later. However, by executing the area discrimination process, as shown in FIG. 6(b), the photographed reference image 30 disappears, and the first area 41 filled in red and the second area 42 filled in green are displayed on the display unit 140. The operator 10 can easily or quickly grasp the processing result of the area discrimination process by checking the color-coded first area 41 and second area 42.

[0026] After the discrimination unit 121 has completed the region discrimination process, the calculation unit 122 executes a first pixel counting process (step S3). The first pixel counting process is a process for counting four types of pixel counts. For example, the first pixel counting process counts the number of pixels from the left edge of the screen to the first region 41 filled in red, and the number of pixels from the right edge of the screen to the first region 41 filled in red. The number of pixels in the first region 41 can be determined by subtracting these two counted pixel counts from the number of pixels in the longitudinal direction of the entire screen. Similarly, the first pixel counting process counts the number of pixels from the top edge of the screen to the second region 42 filled in green, and the number of pixels from the bottom edge of the screen to the second region 42 filled in green. The number of pixels in the second region 42 can be determined by subtracting these two counted pixel counts from the number of pixels in the transverse direction of the entire screen.

[0027] After the first pixel counting process is completed, the calculation unit 122 executes a thickness calculation process (step S4). More specifically, the calculation unit 122 calculates the thickness of the tree 21 based on the four types of pixel counts described above. The calculated thickness of the tree 21 is not displayed at this point but is stored. Details of the thickness calculation process will be described later.

[0028] When the calculation unit 122 finishes executing the thickness calculation process, the discrimination unit 121 determines whether or not the first instruction has been detected (step S5). As described above, the first instruction is an instruction to re-discriminate the first region 41 and the second region 42. For example, as shown in FIG. 6(b), there may be a large deviation between the first region 41 and the part of the tree 21, or between the second region 42 and the part of the tape 22. In this case, even if the thickness of the tree 21 is calculated based on the largely deviated first region 41 or second region 42, the calculation accuracy of the thickness of the tree 21 may be reduced. Therefore, if the operator 10 determines through the screen that such a deviation has occurred, the operator 10 can press the first button image B1 with his or her finger. This causes the discrimination unit 121 to determine that the first instruction has been detected (step S5: YES).

[0029] If it is determined that the first instruction has been detected, the discrimination unit 121 executes the process of step S2 again. That is, the discrimination unit 121 executes the area discrimination process again. This allows the operator 10 to adjust the position and tilt angle of the image capturing device 100 again to align the center image 32 with the tape 22. On the other hand, if the discrimination unit 121 determines that the first instruction has not been detected (step S5: NO) and the calculation unit 122 also determines that the second instruction has not been detected (step S6: NO), the loop process ends (step S7). This returns to the process of step S1, and the discrimination unit 121 acquires images continuously (or in real time).

[0030] On the other hand, for example, as shown in FIG. 7(a), there may be no or only a small misalignment between the first region 41 and the tree 21 or between the second region 42 and the tape 22. In this case, if the thickness of the tree 21 is calculated based on the first region 41 or the second region 42, the thickness of the tree 21 can be calculated with high accuracy. Therefore, if the operator 10 determines through the screen that there is no misalignment, the operator 10 can press the second button image B2 with his or her finger. This causes the calculation unit 122 to determine that a second instruction has been detected (step S6: YES).

[0031] If it is determined that the second instruction has been detected, the loop processing is terminated, and the discrimination unit 121 executes display processing (step S8), and the processing ends. The display processing is processing in which an image at the timing when the second instruction is detected is displayed as a still image on the display unit 140, and an adjustment image and the tree thickness calculated by the calculation unit 122 are displayed on the still image. The adjustment image is an image in which the width corresponding to the thickness of the tree 21 and the vertical width of the tape 22 are adjusted based on independent operations. By executing the display processing, four marker images M1, M2, M3, and M4 are displayed as adjustment images on the still image, as shown in FIG. 7(b). In particular, the marker images M1 and M2 are displayed on a dashed horizontal line 51 corresponding to the position of the diameter at breast height, and the marker images M3 and M4 are displayed on a dashed vertical line 52 passing through the center of the screen.

[0032] In addition to the display of the adjustment image, the thickness of the tree is also displayed in a balloon image 53 arranged, for example, above the marker image M2. The worker 10 can independently operate the marker images M1, M2, M3, and M4 with his fingers as needed to adjust the width corresponding to the thickness of the tree 21 and the height of the tape 22. The calculation unit 122 calculates the thickness of the tree 21 in conjunction with the operation by the worker 10, and the discrimination unit 121 displays the thickness of the tree 21 calculated by the calculation unit 122 in the balloon image 53.

[0033] Although not shown, if the discrimination unit 121 does not detect any operation on the marker images M1, M2, M3, and M4 for a threshold time or longer, the discrimination unit 121 may display a third button image on the display unit 140. The worker 10 can press the third button image with his or her finger. When the third button image is pressed, the discrimination unit 121 detects a third instruction and registers the thickness of the tree before or after adjustment using the adjustment image in the report information. If the discrimination unit 121 detects the third instruction, a still image may be stored in association with the report information, along with registering the tree thickness in the report information.

[0034] Next, the above-mentioned area discrimination process will be described in detail with reference to FIG.

[0035] First, the discrimination unit 121 inputs an image into the trained model (step S11). More specifically, the discrimination unit 121 acquires the trained model from the model storage unit 111, and sequentially inputs images into the acquired trained model. When the image is input, the discrimination unit 121 acquires a first region and a second region (step S12). As described above, the trained model is information for distinguishing between the first region and the second region that respectively fit within the image, so by inputting an image into the trained model, the first region and the second region can be acquired.

[0036] After acquiring the first and second regions, the discrimination unit 121 fills the first region with red (step S13) and fills the second region with green (step S14). The order of steps S13 and S14 may be reversed. After filling the first region with red and the second region with green, respectively, the discrimination unit 121 displays the filled first region and second region superimposed on the image (step S15), and ends the region discrimination process. The discrimination unit 121 displays the first region filled with red and the second region filled with green without overlapping each other (see FIG. 7(a)). This makes it possible to avoid color mixing in the overlapping areas of the first region and the second region.

[0037] Next, the above-mentioned first pixel number counting process will be described in detail with reference to FIG.

[0038] First, the calculation unit 122 executes a second pixel counting process with red as the first argument and the width as the second argument (step S21). The red color corresponds to the tree 21. Next, the calculation unit 122 executes a second pixel counting process with green as the first argument and the height as the second argument (step S22), and ends the first pixel counting process. The green color corresponds to the tape 22. The order of steps S21 and S22 may be reversed.

[0039] Here, when the color is specified as the first argument and the width is specified as the second argument, the second pixel counting process counts the number of pixels from the edge of the screen to the first area and the second area. For example, in the process of step S21, two numbers of pixels are counted: the number of pixels from the left edge of the screen to the first area, and the number of pixels from the right edge of the screen to the first area. In the process of step S22, two numbers of pixels are counted: the number of pixels from the top edge of the screen to the second area, and the number of pixels from the bottom edge of the screen to the second area. The second pixel counting process will be described in detail below.

[0040] The second pixel counting process will be described in detail with reference to FIG.

[0041] First, the calculation unit 122 determines whether the second argument of the second pixel counting process is the width (step S31). If the second argument is the width (step S31: YES), the calculation unit 122 executes area detection process in which the first argument is red, the second argument is left, the third argument is height, and the fourth argument is tr_lft, which is one of the saved variables (step S32). After the process of step S32 is completed, the calculation unit 122 executes area detection process in which the first argument is red, the second argument is right, the third argument is height, and the fourth argument is tr_rgt, which is one of the saved variables (step S33). After the process of step S33 is completed, the calculation unit 122 ends the second pixel counting process. Note that the order of the processes of steps S32 and S33 may be reversed.

[0042] On the other hand, if the second argument of the second pixel counting process is the vertical width (step S31: NO), the calculation unit 122 executes area detection process in which the first argument is green, the second argument is top, the third argument is width, and the fourth argument is tp_top, which is one of the saved variables (step S34). After the process of step S34 is completed, the calculation unit 122 executes area detection process in which the first argument is green, the second argument is bottom, the third argument is width, and the fourth argument is tp_btm, which is one of the saved variables (step S35). After the process of step S35 is completed, the calculation unit 122 ends the second pixel counting process. The order of the processes of steps S34 and S35 may be reversed.

[0043] Here, the area detection process specifies a color as a first argument, a line position as a second argument, a line type as a third argument, and a storage variable as a fourth argument, and then detects the number of pixels in the area from the line position to the nearest first or second area. For example, in the process of step S32, if the line position is on the left, a straight line (vertical line) extending vertically is drawn at the left edge of the screen. That is, a straight line parallel or nearly parallel to the vertical width direction of the tape 22 is drawn. Then, this straight line is sequentially moved toward the first area, and the number of pixels moved until this line overlaps the left side of the first area by more than a threshold number of pixels is detected. Steps S33, S34, and S35 are basically the same as the process of step S32. The area detection process will be described in detail below.

[0044] The area detection process will be described in detail with reference to FIG.

[0045] First, the calculation unit 122 draws a line of either vertical or horizontal type specified as a third argument at any line position specified as a second argument among the left edge, right edge, top edge, and bottom edge of the screen of the area filled with the color specified as the first argument of the area detection process (step S41). As a result, for example, in the process of step S32 described above, a vertical line is drawn at the left edge of the screen of the first area 41 (see FIG. 7(a)) filled with red.

[0046] After drawing the line, the calculation unit 122 starts a loop process of moving the line by one pixel in the direction opposite to the line position specified as the second argument, either left, right, up, or down (step S42). As a result, for example, if a vertical line is drawn at the left edge of the screen in the process of step S41 described above, the line is moved by one pixel in the direction opposite to the left edge of the screen, toward the first region 41.

[0047] After the line moves, the calculation unit 122 then determines whether the number of pixels of the color specified as the first argument on the line of either the vertical or horizontal line type specified as the third argument is equal to or greater than the threshold number of pixels (step S43). The threshold number of pixels can be, for example, several tens of pixels to several hundreds of pixels. For example, if a vertical line drawn at the left edge of the screen moves just one pixel toward the first region 41, there is little or no chance that this line will be located within the first region 41. Therefore, in this case, the calculation unit 122 determines that the number of pixels of the color specified as the first argument on the line of either the vertical or horizontal line type specified as the third argument is not equal to or greater than the threshold number of pixels (step S43: NO). If this determination is made, the calculation unit 122 ends the loop processing (step S44) and returns to the processing of step S42. That is, the calculation unit 122 continues to move the line until it determines that the number of pixels of the color specified as the first argument on the line of either the vertical or horizontal line type specified as the third argument is equal to or greater than the threshold number of pixels.

[0048] On the other hand, if a vertical line drawn at the left edge of the screen moves several tens to several hundreds of pixels toward the first region 41, there is a possibility that this line will be located on the first region 41. Therefore, if the number of pixels in the first region 41 where this line is located is equal to or greater than the threshold number of pixels, the calculation unit 122 determines that the number of pixels of the color specified as the first argument on the line of either the vertical or horizontal line type specified as the third argument is equal to or greater than the threshold number of pixels (step S43: YES). If this determination is made, the calculation unit 122 exits the loop processing, stores the number of pixels moved in a saved variable specified as the fourth argument (step S45), and terminates the region detection processing. For example, if a vertical line moving from the left edge of the screen reaches the first region 41, the number of pixels in the portion where this line overlaps the first region 41 is stored in one of the saved variables, tr_lft. The remaining saved variables, tr_rgt, tp_top, and tp_btm, are basically the same as tr_lft. In this way, the region detection process can detect the number of pixels in the four regions as four stored variables.

[0049] Next, the above-mentioned thickness calculation process will be described in detail with reference to FIG.

[0050] First, the calculation unit 122 acquires four types of stored variables (step S51). That is, the calculation unit 122 acquires tr_lft, tr_rgt, tp_top, and tp_btm, which are stored variables in which the number of pixels is stored. After acquiring the stored variables, the calculation unit 122 then calculates the number of pixels in the width of the tree (step S52). For example, the number of pixels in the vertical and horizontal directions of the display unit 140 or the screen displayed on the display unit 140 is often known. Therefore, the calculation unit 122 can calculate the number of pixels in the width of the tree by subtracting the sum of the number of pixels stored in tr_lft and the number of pixels stored in tr_rgt from the number of pixels in the horizontal direction of the display unit 140 or the screen.

[0051] After calculating the number of pixels in the width of the tree, the calculation unit 122 then calculates the number of pixels in the height of the tape (step S53). When calculating the number of pixels in the height of the tape, the calculation unit 122 can calculate the number of pixels in the height of the tape in the same way as in the processing of step S52. That is, the calculation unit 122 can calculate the number of pixels in the height of the tape by subtracting the sum of the number of pixels stored in tp_top and the number of pixels stored in tp_btm from the number of pixels in the vertical direction of the display unit 140 or screen. Note that the processing order of steps S52 and S53 may be reversed.

[0052] After calculating the number of pixels in the vertical width of the tape, the calculation unit 122 calculates the thickness of the tree (step S54) and ends the process. Specifically, the calculation unit 122 has calculated the number of pixels in the horizontal width of tree 21 and the number of pixels in the vertical width of tape 22 through the processes of steps S52 and S53 described above. Meanwhile, the length of the vertical width of tape 22 is known. Therefore, the calculation unit 122 can calculate the thickness of the tree by multiplying the ratio of the number of pixels in the horizontal width of tree 21 to the number of pixels in the vertical width of tape 22 by the actual length of the vertical width of tape 22.

[0053] Next, the above-mentioned display process will be described in detail with reference to FIG.

[0054] First, the discrimination unit 121 displays a horizontal line and a vertical line on the screen of the display unit 140 (step S61). As a result, a dashed horizontal line 51 corresponding to the position of the diameter at breast height and a dashed vertical line 52 passing through the center of the screen are displayed on the screen of the display unit 140 (see FIG. 7(b)). The position of the diameter at breast height may be, for example, several tens to several hundreds of pixels from the bottom edge of the screen. After displaying the horizontal line and the vertical line, the discrimination unit 121 displays marker images on the horizontal line at positions corresponding to tr_lft and tr_rgt (step S62). As a result, marker images M1 and M2 are displayed on the horizontal line 51 on the screen (see FIG. 7(b)).

[0055] After displaying these marker images, the discrimination unit 121 displays marker images at positions corresponding to tp_top and tp_btm on the vertical line (step S63). As a result, marker images M3 and M4 are displayed on the vertical line 52 on the screen (see FIG. 7(b)). After displaying these marker images, the discrimination unit 121 displays the thickness of the tree (step S64) and ends the process. As a result, for example, the thickness of the tree is displayed in a balloon image 53 arranged above the marker image M2 (see FIG. 7(b)).

[0056] As described above, the imaging device 100 according to this embodiment includes a discrimination unit 121 and a calculation unit 122. The discrimination unit 121 discriminates between a first region 41 of the tree 21 and a second region 42 of the tape 22 attached horizontally to the tree 21, each of which is contained within the image, based on a trained model that discriminates between the first region 41 and the second region 42. The calculation unit 122 calculates the thickness of the tree 21 based on the relationship between the width corresponding to the short side direction of the first region 41 and the vertical width of the second region 42. This makes it possible to calculate the thickness of the tree from the image, thereby reducing the effort and time required by the worker 10.

[0057] Although the preferred embodiment of the present invention has been described above in detail, it is not limited to the specific embodiment of the present invention, and various modifications and variations are possible within the scope of the gist of the present invention as set forth in the claims. For example, since the imaging device 100 has an RF circuit 100E, the recording information may be transmitted using wireless communication and stored in a server installed in a location different from the work location of the worker 10.

[0058] Furthermore, in the above-described embodiment, red and green are described as examples of colors, but the colors are not particularly limited to red and green, and may be, for example, blue and yellow, as long as they can distinguish the first region 41 and the second region 42. Furthermore, patterns may be used instead of colors as long as they can distinguish the first region 41 and the second region 42. In addition, by employing a trained model that has also trained on tree types, the type of tree identified from the image can be registered in the report information. [Explanation of symbols]

[0059] 100 Imaging device 110 Storage section 111 Model memory section 112 Record Storage Unit 120 Processing section 121 Discrimination part 122 Calculation Unit

Claims

1. A first region, which is the entire region of the tree, and a second region, which is the entire region of a rectangular tape attached horizontally to the tree, each of which fits within the image, are distinguished based on a trained model that distinguishes between the first region and the second region; Calculating the diameter at breast height of the tree trunk based on the relationship between the width corresponding to the short side direction of the first region and the vertical width of the second region. A calculation program that causes a computer to execute a process.

2. The calculation process calculates the diameter at breast height based on a ratio between the first number of pixels in the horizontal width and the second number of pixels in the vertical width.

2. The calculation program according to claim 1 .

3. the determining process includes, after determining the first area and the second area, displaying the first area and the second area in a distinguishable manner, and, when a first instruction to re-determine the first area and the second area is detected after the display, determining the first area and the second area again.

3. The calculation program according to claim 1 or 2.

4. the determining process includes determining the first area and the second area, and then displaying the first area and the second area in a distinguishable manner; The calculation process calculates the breast height diameter when a second instruction requesting calculation of the breast height diameter is detected after the display.

4. The calculation program according to claim 1, wherein the calculation program is a program for calculating a time period.

5. the determining process includes, after determining the first area and the second area, displaying an adjustment image for adjusting the width and the height based on independent operations; The calculation process calculates the breast height diameter based on the relationship between the width and the height after adjustment based on the operation on the adjustment image.

5. The calculation program according to claim 1, wherein the calculation program is a program for calculating a time period.

6. the determining process determines the first area and the second area, and then displays the first area and the second area by filling them in different colors without overlapping each other.

6. The calculation program according to claim 1, wherein the calculation program is a program for calculating a time period.

7. A first region, which is the entire region of the tree, and a second region, which is the entire region of a rectangular tape attached horizontally to the tree, each of which fits within the image, are distinguished based on a trained model that distinguishes between the first region and the second region; Calculating the diameter at breast height of the tree trunk based on the relationship between the width corresponding to the short side direction of the first region and the vertical width of the second region. A calculation method in which processing is performed by a computer.

Images