Boundary line extraction device, boundary line extraction method, and program

Abstract

To provide a boundary line extraction device, a boundary line extraction method, and a program which can extract a boundary line of a geographic feature with higher accuracy.SOLUTION: The boundary line extraction device analyzes a microtopographical representation diagram to extract a boundary line of an area where a prescribed topographic feature appears. The boundary line extraction device comprises: storage means in which a trained model is stored, wherein the trained model is preliminarily trained so as to, in response to input of a microtopographical representation diagram, output a boundary area obtained by extending a line width of a boundary line of an area where a topographical feature appears in the microtopographical representation diagram; acquisition means which inputs a microtopographical representation diagram being an analysis object to the trained model to acquire a boundary area outputted from the trained model; and extraction means which thins the boundary area acquired by the acquisition means and extracts a boundary line of an area where a topographic feature appears in the microtopographical representation diagram being the analysis object.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a boundary line extraction device, a boundary line extraction method, and a program. [Background technology]

[0002] Information on landslide areas in mountainous regions is used for forestry conservation measures. Traditionally, landslide areas have been identified by engineers from map data that emphasizes the topography, such as microtopographical representations. Landslide areas usually occur in U-shaped areas, and their boundaries tend to appear on map data. Therefore, in the process of identifying the boundaries of the U-shape, the boundary line is identified (Non-Patent Document 1). This type of identification requires skill and a great deal of effort. Therefore, there is a need for support through automatic extraction. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] "Lecture 3: Extracting Disaster Factors through Topographic Interpretation," Road Disaster Prevention Inspection Techniques Seminar Related Materials, Japan Geological Survey Association, published May 27, 2022. https: / / www.zenchiren.or.jp / geocenter / lec-road / docs / lecture3.pdf Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the case of terrain features with narrow boundaries in an image, the correct region occupies a small proportion of the entire image. As a result, it is difficult to obtain appropriate learning results when training a machine learning model to automatically extract these features, and there is a problem in that the boundaries of the terrain features cannot be extracted with high accuracy.

[0005] An object of the present invention is to provide a boundary line extraction device, a boundary line extraction method, and a program that are capable of extracting the boundary lines of topographical features with higher accuracy. [Means for solving the problem]

[0006] In order to achieve the above object, the present invention provides A boundary line extraction device that analyzes a microtopographical representation map and extracts a boundary line of an area where a predetermined topographical feature appears, a storage means for storing a trained model that has been trained in advance to output a boundary area in which the line width of the boundary line of the area in which the topographical feature appears in the microtopographical representation map is expanded in response to the input of the microtopographical representation map; an acquisition means for inputting a microtopography representation map of an analysis target into the trained model and acquiring the boundary area output from the trained model; an extraction means for thinning the boundary area acquired by the acquisition means and extracting a boundary line of an area in which the topographical feature appears in the microtopographical representation map of the analysis target; The boundary extraction device is provided with: [Effects of the Invention]

[0007] According to the present invention, it is possible to extract the boundary lines of topographical features with higher accuracy. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing the functional configuration of an information processing device that is a boundary line extraction device according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram illustrating a procedure for detecting a collapsed area. [Figure 3] FIG. 10 is a diagram showing an example of a microtopography representation and the resulting boundary region. [Figure 4] FIG. 10 is a diagram showing an example of boundary line data obtained by thinning the obtained boundary region image. [Figure 5] 10 is a flowchart showing a control procedure for boundary detection processing. [Figure 6] 10 is a flowchart showing a control procedure for a wire connection process. [Figure 7] 10 is a flowchart showing a control procedure for a learning data generation process. [Figure 8] FIG. 10 is a diagram illustrating an example of generating correct answer data. [Figure 9] 10 is a flowchart showing a control procedure for a model learning process. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the functional configuration of an information processing device 1 which is a boundary line extraction device according to this embodiment.

[0010] The information processing device 1 includes a control unit 11, a storage unit 12 as a storage means, an input / output interface 13 (I / F), a display unit 14, an operation reception unit 15, and the like.

[0011] The control unit 11 controls the overall operation of the information processing device 1. The control unit 11 has a processor that performs arithmetic processing. The processor may be a single general-purpose CPU (Central Processing Unit), or may have multiple CPUs that perform arithmetic processing in parallel or independently depending on the application. The processor may include a processor specialized for specific arithmetic processing or image processing. The control unit 11 performs various control processing by reading and executing a program 120 from the storage unit 12.

[0012] The storage unit 12 has a RAM (Random Access Memory) and a non-volatile memory, and stores various data. The RAM provides a working memory space for the control unit 11 and stores temporary data. The non-volatile memory stores and holds the program 120, the machine learning model 122, setting data, and the like. The non-volatile memory is, for example, a flash memory or an HDD (Hard Disk Drive), but is not limited to these. The storage unit 12 may have a ROM (Read Only Memory). The ROM may store an initial control program, and the like. The program 120 includes control programs related to the boundary detection process, the learning data generation process, and the model learning process, which will be described later. The program 120 includes a trained model 121. The trained model 121 is a machine learning model 122 that has been trained. Once the trained model 121 is obtained, the machine learning model 122 does not need to be stored in the storage unit 12.

[0013] The input / output interface 13 inputs and outputs data to and from the outside of the information processing device 1 (including peripheral devices). The input / output interface 13 has a connection terminal 131 and a communication unit 132. The connection terminal 131 includes, for example, a USB (Universal Serial Bus) terminal and a LAN (Local Area Network) connector. The communication unit 132 controls communication in accordance with a communication protocol related to the LAN, such as TCP / IP.

[0014] The peripheral devices serving as the learning data storage means of this embodiment may include a database device 21, which is an auxiliary storage device, and an optical reader 22 that reads portable storage media (optical discs) such as CD-ROM, DVD, and Blu-ray (registered trademark). Also, the portable storage media may include a magnetic tape, and the peripheral devices may include a reader that reads this magnetic tape.

[0015] Data that the information processing device 1 can acquire from the outside via the input / output interface 13 includes microtopography representation map data 201, learning boundary line data 202, and learning boundary area data 203. The microtopography representation map data 201 stores one or more microtopography representation maps (learning microtopography representation maps) used for training the machine learning model 122, microtopography representation maps that are the target of analysis by the trained model 121 (analysis target microtopography representation maps), and the correspondence between each of the microtopography representation maps and their geographic coordinates. The microtopography representation map is an image that has been enhanced to make the relief of the terrain easier to read from a normal topographic map. Known examples of microtopography representation maps include CS (Curvature / Slope) 3D maps, red 3D maps, and the topographic map described in JP 2021-149059 A.

[0016] The microtopography representation is a pixmap (raster) image represented by a set of pixels arranged two-dimensionally in a grid pattern in the x and y directions. The image must be capable of expressing the range of colors required to represent the microtopography representation. The microtopography representation data 201 stores the XY coordinates of a plane Cartesian coordinate system (the center of gravity of the geographical area represented by each pixel) represented by the centers of the two diagonal pixels of the microtopography representation, in correspondence with each microtopography representation. This makes it possible to identify the correspondence between the position of each pixel of the microtopography representation and the geographical coordinates. The geographical coordinates may be latitude and longitude.

[0017] The training boundary data 202 is linked to the training microtopographic representation map and stores data on the boundary lines of collapsed areas (training boundary lines) that appear on the training microtopographic representation map. In other words, the training boundary lines indicate the boundary lines of areas where topographical features appear on the training microtopographic representation map. Each line may be multi-line data expressed as a vector using geographical coordinates. The training boundary data 202 is generated manually by an engineer or the like, referring to the microtopographic representation map data and, if necessary, aerial photographs, etc. Also, known collapsed area data may be used. Note that if a trained model with a certain degree of accuracy already exists, candidate extraction work may be performed as preprocessing using the trained model before the engineer's manual work.

[0018] The training boundary area data 203 stores training boundary areas generated by expanding the line width of the training boundary line, linked to the training microtopography representation map. Each area may be polygon data represented as a vector in a geographic coordinate system. The training boundary area is correct data of the boundary area for the training microtopography representation map, and the set of the training microtopography representation map and the training boundary area is training data for the machine learning model 122.

[0019] The display unit 14 displays on a display screen under the control of the control unit 11. The display screen is, for example, a liquid crystal display or an organic EL (Electro-Luminescent) display, but is not limited to these.

[0020] The operation reception unit 15 receives an input operation from outside and outputs an operation signal corresponding to the input operation to the control unit 11. The operation reception unit 15 includes, for example, a keyboard and a pointing device. The pointing device may be a mouse. The display unit 14 and / or the operation reception unit 15 may be peripheral devices of the information processing device 1. That is, they may be attached to the main body (computer) of the information processing device 1 including the control unit 11, the storage unit 12, and the input / output interface 13.

[0021] Next, as a boundary line extraction method according to this embodiment, an operation for detecting a collapsed area as an area where a predetermined topographical feature appears will be described. A collapse refers to the collapse of layers of rock en masse, typically on steep slopes, and a collapsed area is the area where such a collapse occurs. A collapsed area typically has a clear boundary, revealing a linear discontinuity. Collapses often occur on the uppermost slopes of a valley, resulting in a U-shaped boundary surrounding the valley. Meanwhile, rocks and sediment from the collapsed slope tend to spread and deposit on gentler or nearly flat slopes, filling the valley. In other words, a collapsed area is typically defined as a combination of a steep, U-shaped discontinuity surrounded by this and a relatively gentler area extending downstream. Vegetation may grow on the sediments long after the collapse. In such cases, aerial photographs often reveal traces of the collapse within the collapsed area, compared to the boundary where the steep slope remains clearly visible.

[0022] FIG. 2 is a diagram illustrating the procedure for detecting collapsed areas. The trained model 121 is stored in advance in the storage unit 12 (storage step). As shown in FIG. 2(a), in this embodiment, the trained model 121 extracts relatively clear boundaries of collapsed areas from the microtopography representation map to be analyzed. This trained model 121 outputs a boundary region having an area, as will be described later (P1; acquisition step). The boundary region acquired as the output of the trained model 121 is changed to a boundary line by thinning processing (P2; extraction step), and the final result is obtained.

[0023] The trained model 121 and the machine learning model 122 can be neural network models having a network structure related to image segmentation. For example, they may be modeled as Res U-Net. The trained model 121 and the machine learning model 122 can obtain the probability of each pixel being a boundary region. By binarizing the probability value using a predetermined reference value, an image indicating whether or not a pixel is a boundary region can be obtained.

[0024] FIG. 3 shows an example of a microtopography representation and the resulting boundary region. Figure 3(a) shows a grayscale CS 3D map, a representation of microtopography. This CS 3D map is a grayscale version of the Shizuoka Prefecture CS 3D map (created by Shizuoka Prefecture, Shizuoka Prefecture Forest Planning Division, 2017). The CS 3D map is a map in which elevation, slope, and curvature are overlaid with different colors and other gradations. This map makes it easy to identify collapsed areas from slopes to valleys. Here, the dark areas correspond to valleys, and the U-shaped areas that suddenly cut deep near the source of the valley generally correspond to collapsed areas. This CS 3D map was input into the trained model 121, and an example of a boundary region image obtained by binarizing the resulting probability distribution is shown in Figure 3(b). As mentioned above, many collapsed areas have U-shaped boundaries. In some cases, multiple collapsed areas may be located adjacent to each other.

[0025] FIG. 4 is a diagram showing an example of boundary line data obtained by thinning the obtained boundary region image. As shown in Figure 4(a), a boundary region is a curve with a certain width, often exhibiting a characteristic U-shape. In the thinning process, this region is thinned, i.e., a line image with its width reduced in a direction perpendicular to the extension direction is obtained, and then this is represented as a multipolygon, which is a series of multiple polygons. The thinning process may be performed using any conventionally known method. As shown in Figure 4(b), the multipolygon is represented by a combination of only vertical and horizontal lines of pixels corresponding to the image resolution. In other words, each polygon may be a combination of shapes such as a rectangle, a hook, or an L-shape. By connecting representative positions within each polygon, such as the center of gravity, a polygonal boundary line, as shown in Figure 4(c), is obtained.

[0026] It is not easy to manually determine the correct data for such a boundary area, and so boundary lines are usually given as correct data. However, learning using only lines makes it difficult to obtain effective learning results, as the area occupied by the "lines" in the image is too small. Therefore, as shown in Figure 2(b), the line width of the given boundary line (learning boundary line) is expanded and thickened, and then converted into an image of a boundary area (learning boundary area) with width, thereby obtaining an image of the correct area (P11; expansion means). The correct area image is then associated with the original microtopography representation to obtain learning data.

[0027] The machine learning model 122 is trained using the obtained training data (P12; training means), thereby obtaining the trained model 121.

[0028] 5 is a flowchart showing the control procedure of the boundary detection process executed by the information processing device 1. This boundary detection process is triggered by the operation receiving unit 15 or the communication unit 132 receiving, for example, a predetermined input operation by the user together with the designation of the microtopographic representation map to be analyzed.

[0029] As shown in FIG. 5(a), the control unit 11, which serves as an acquisition means, acquires the microtopography representation map to be analyzed from the microtopography representation map data 201 and inputs it to the trained model 121 (S11). The size of the image that can be input to the trained model 121 (input size) is predetermined. This size is smaller than the microtopography representation map to be analyzed. The acquisition means sequentially inputs divided images obtained by dividing the microtopography representation map to be analyzed into images of the input size to the trained model 121. Note that if the size of the microtopography representation map to be analyzed is not a multiple of the input size, some of the divided images may be set to have overlapping ranges.

[0030] The acquisition means operates the trained model 121 and acquires a probability distribution map output from the trained model 121 (S12). In response to the input of the segmented image in process S11, the trained model 121 outputs a probability distribution map of the same size as the segmented image. The acquisition means acquires a probability distribution map of the same size as the analysis target microtopography representation map by arranging the output probability distribution map in the same position as the corresponding segmented image. In this case, the value of a pixel that overlaps between multiple segmented images may be set to the maximum value of the pixel. Alternatively, the pixel value may be set to the average value of the values ​​obtained from the multiple segmented images. The value of each pixel in the analysis target microtopography representation map represents the probability that it is a boundary area of ​​a collapsed area. The boundary area is an area obtained by expanding the width of the boundary line of the collapsed area. The acquisition means binarizes the probability distribution map using a predetermined reference value and identifies the boundary area of ​​the collapsed area (S13). The trained model 121 may include binarization processing and output a binarized image. The control unit 11, which serves as an extraction means, executes boundary line identification processing based on the binarized image (S14).

[0031] As shown in Fig. 5(b), in the boundary line identification process, the extraction means thins the boundary area (S41). For example, the extraction means reduces the width of the boundary area perpendicular to the extension direction to a minimum reference value. The minimum reference value may be, for example, one pixel.

[0032] The extraction means converts the line image obtained by thinning the boundary area into polygons (S42). That is, as described above, the extraction means generates a polygon representing a geographical area corresponding to each group of pixels connected in the x direction and / or y direction in the line image. Rectangular polygons are generated in parts of the line image where pixels are connected only in the x direction or y direction. L-shaped or hook-shaped polygons are generated in parts of the line image where a group of pixels in the x direction and a group of pixels in the y direction are connected at a common pixel.

[0033] The extraction means counts the number of polygons for each group of polygons adjacent to each other in a chain. The extraction means determines that a boundary area with less than three polygons is a detection noise that is not U-shaped, and excludes it from subsequent processing (S43). If there is no group with three or more polygons, the extraction means terminates the boundary line identification process without performing the processing from S44 onwards.

[0034] The extraction means extracts a representative position, here a center of gravity position, within each polygon that was not excluded in step S43 (S44). The coordinate system of the representative position is a geographic coordinate system. The extraction means executes a connection process to connect the extracted center of gravity positions (S45). Then, the control unit 11 ends the boundary line identification process and returns the process to the boundary detection process.

[0035] FIG. 6 is a flowchart showing the control procedure for the wire connection process. The extraction means counts the number of other polygons adjacent to each polygon (S401). The extraction means determines whether there is one polygon with an adjacent number of "1" (S402). A polygon with an adjacent number of "1" is a polygon at the end of a boundary line. In a normal boundary line, there are two polygons with an adjacent number of "1". If it is determined that there is one polygon with an adjacent number of "1" (S402; Y), the extraction means sets the polygon with an adjacent number of "1" as the reference polygon (S403). Then, the processing of the extraction means proceeds to step S405.

[0036] If it is determined that there is more than one polygon with an adjacent number of "1" (S402; N), the polygon with an adjacent number of "1" that is closest to a predetermined position, for example, the upper left corner of the analysis target microtopography representation map, is set as the reference polygon (S404). Then, the processing of the extraction means proceeds to step S405.

[0037] When the process proceeds to step S405, the extraction means adds the representative position (center of gravity position) of the reference polygon to a point list that lists points on the boundary line (S405). From the center of gravity position of the reference polygon, the extraction means identifies the polygon that has the closest center of gravity position among the polygons that have not yet been set as the reference polygon (S406). At this time, if there are no unset polygons left, the extraction means proceeds to step S411.

[0038] The extraction means determines whether the distance between the center of gravity of the reference polygon and the center of gravity of the identified polygon is equal to or greater than a reference value (S407). If it is determined that the distance is not equal to or greater than the reference value (S408; N), the extraction means determines that the identified polygon is located on the same boundary line as the reference polygon, and sets the identified polygon as the next reference polygon (S408). Then, the processing of the extraction means returns to step S405.

[0039] If it is determined that the distance is equal to or greater than the reference value (S408; Y), the extraction means determines in step S413 whether the distance to the endpoint (usually the starting point) of another connected boundary line is equal to or greater than the reference value (S411). If it is determined that there is a connected boundary line whose distance is not equal to or greater than the reference value (S411; N), the extraction means considers the connected boundary line to be part of the boundary line represented by the point list being processed, and adds and merges the points of the connected boundary line in order to the point list (S412). The extraction means deletes the added and merged connected boundary line from the connected data. Then, the processing of the extraction means proceeds to step S413. If it is determined that the distance to each endpoint of another connected boundary line is equal to or greater than the reference value (S411; Y), the point list being processed is considered to be that of a new boundary line, and the processing of the extraction means proceeds to step S413. Incidentally, even if there is no data of other connected boundary lines, the processing of the extraction means proceeds to step S413.

[0040] When the process proceeds to step S413, the extraction means connects the points stored in the point list in the set order to identify the boundary line (S413). That is, the boundary line is obtained as vector data connecting coordinates. The extraction means stores the data of the point list in the memory unit 12 as data of the connected boundary line, and then erases the contents of the point list.

[0041] The extraction means determines whether there are any remaining polygons that have not yet been set as reference polygons (S414). If it is determined that there are any remaining polygons (S414; Y), the processing by the extraction means returns to step S402. If it is determined that there are no remaining polygons (S414; N), the extraction means ends the line connection processing and returns the processing to the boundary line identification processing. Through the above processing, the boundary line of the collapsed area is extracted from the microtopography representation map.

[0042] Fig. 7 is a flowchart showing a control procedure for a learning data generation process for training the machine learning model 122. Fig. 8 is a diagram showing an example of generation of correct answer data.

[0043] The control unit 11 selects a learning microtopography representation map to be used as learning data from the microtopography representation map data 201 (S31). The control unit 11 sets a learning boundary line using line data (S32). As shown in FIG. 8(a), the boundary line may be a broken line. Alternatively, the boundary line may be drawn freehand as a free curve. The boundary line may be set using a pointing device of the operation reception unit 15 on an image displayed on the display screen of the display unit 14. The control unit 11 may also acquire data of a learning boundary line that has been set externally via the communication unit 132. The line data set here may be data expressed as a vector in geographical coordinates.

[0044] The control unit 11, which functions as an expansion means, thickens the set learning boundary line (S33). The thickening of the vector data can be achieved simply by changing the line width information. The expansion width is set in advance so that the thickness in the microtopography representation map is two pixels or more and the boundary area is sufficiently thinner than the collapsed area. The expansion width may be set in actual size, or the width in the image data to be expanded may be determined according to the scale of the learning microtopography representation map so that the width of the boundary line is one meter. As shown in Figure 8(b), a boundary line that is simply thicker than that in Figure 8(a) is obtained.

[0045] The expansion means identifies the position of the thick line and converts the vector data into a bitmap (S34). That is, as shown in FIG. 8(c), the control unit 11 identifies pixel positions where the thick line overlaps in a bitmap image with the same resolution as the training microtopography representation, and sets a binary value according to whether or not there is overlap with the thick line. The control unit 11 combines the selected training microtopography representation with the generated bitmap data to generate training data (S35). The generated bitmap data is linked to the training microtopography representation and registered in the training boundary area data 203.

[0046] The control unit 11 determines whether or not the selection of the learning microtopography representation map has been completed (S36). If it is determined that the selection of the learning microtopography representation map has not been completed (S36; N), the control unit 11 returns to step S31. If it is determined that the selection of the learning microtopography representation map has been completed (S36; Y), the control unit 11 ends the learning data generation process.

[0047] 9 is a flowchart showing the control procedure of the model learning process executed by the information processing device 1. The control unit 11 executes the model learning process by specifying the machine learning model 122 to be learned and the learning data to be used for learning.

[0048] The control unit 11, which serves as a learning means, selects one training microtopography representation map included in the microtopography representation map data 201, cuts out one image of the above-mentioned input size from the training microtopography representation map, and inputs this to the machine learning model 122 (S21). The size of the image that can be input to the machine learning model 122 is predetermined to the above-mentioned input size, and this size is smaller than the training microtopography representation map. The learning means, for example, randomly selects a training microtopography representation map based on a random number, cuts out an image of the input size from the training microtopography representation map at a random position within the training microtopography representation map based on the random number, and inputs the cut-out image to the machine learning model 122. The learning means acquires as an output result (S22) the probability distribution map output from the machine learning model 122. In response to the input of the cut-out image in process S21, the machine learning model 122 outputs a probability distribution map of the same size as the cut-out image.

[0049] The learning means calculates the error of the probability distribution map relative to the correct answer data (S23). First, the learning means reads out the learning boundary area corresponding to the learning microtopography representation map selected in step S21 from the learning boundary area data 203 as the correct answer data. Next, the learning means calculates the error (intra-area error) in the boundary area (inside the learning boundary area) in the correct answer data and the error (out-area error) in the area that is not a boundary area (outside the learning boundary area) in the correct answer data. The learning means calculates the error, for example, by using cross entropy loss as the loss function. In this case, the probability at pixel x indicated by the probability distribution map, i.e., the probability that pixel x is a pixel in the boundary area, is defined as q(x), and the probability of a pixel in the learning boundary area is defined as x pоs , pixels outside the learning boundary area are denoted by x neg Then, the error in the region is -Σln{q(x pоs )} and the out-of-domain error is -Σln{q(x neg )}.

[0050] The learning means divides the errors into errors within the region and errors outside the region, and weights the errors (S24). The learning means assigns different weights to the errors between these regions. For example, the learning means may weight the number of pixels N within the learning boundary region. pos The ratio R pos and the number of pixels outside the learning boundary area N neg The ratio R neg The learning method is weighted according to (1-R pos )=R neg =N neg / (N pos +N neg ) multiplied by the intra-region error, and (1-R neg )=R pos =N pos / (N pos +N neg The error may be calculated by multiplying the out-of-area error by the number of pixels N pos and the number of pixels N neg Alternatively, the learning means may calculate the number of pixels N by averaging the calculated values ​​of the correct answer data of the learning micro-topography representation map included in the micro-topography representation map data 201. pos and the number of pixels Nneg Also, the number of pixels N pos and the number of pixels N neg may be set in advance based on a microtopography representation map that is not included in the microtopography representation map data 201. As described above, since the number of pixels within the learning boundary region is significantly smaller than the number of pixels outside the learning boundary region, the intra-region error may be weighted more heavily and the respective errors may be added up.

[0051] The learning means provides feedback to the parameters of the machine learning model 122 so as to minimize the weighted error, and updates the machine learning model 122 (S25). The method of error feedback may be a conventionally known method, and for example, the backpropagation method may be used.

[0052] The learning means determines whether the selection and input of the learning microtopography representation map has been completed (S26). The learning means determines that the input of step S21 has been completed when the input has been performed a predetermined number of times, or when the error calculated in step S24 is less than a predetermined reference value. If it is determined that the input has not been completed (S26; N), the processing of the learning means returns to step S21. If it is determined that the input has been completed (S26; Y), the learning means determines that the trained machine learning model 122 is the trained model 121 (S27). Then, the learning means terminates the model learning processing.

[0053] The present invention is not limited to the above-described embodiment, and various modifications are possible. For example, the architecture of the machine learning model 122 may be related to image segmentation other than Res U-Net, such as PSPNet, DeepLabV3+, TransUnet, and SegFormer.

[0054] In the above example, the cross-entropy (CE) function is used as the loss function to separately weight the error, but this is not limiting. Focal loss, which takes weighting into account from the beginning, may also be used. Furthermore, Dice loss may also be used in addition to the CE function.

[0055] In the above description, the boundary line is obtained by connecting the centroid positions of each polygon, but this is not limiting. For example, depending on the shape of the polygon, the centroid position may be located outside the polygon. Taking such cases into consideration, or limiting such cases, a representative position (interior guarantee point) located inside the polygon may be appropriately determined. Furthermore, when thinning the boundary area and determining the position on the boundary line, it is not necessary to use a multi-polygon.

[0056] Although the above description has been given with reference to an example in which the boundary line of a collapsed area is extracted, the boundary line is not limited to this and may be any other topographical feature such as a zero-order valley.

[0057] In the above description, the control unit 11 executes all processing within the information processing device 1, but this is not limited to this. Processing may be divided among multiple information processing devices. Furthermore, learning of the machine learning model 122 may be performed outside the information processing device 1. Only the trained model 121 may be copied to the information processing device 1 and used.

[0058] In the above description, the storage unit 12, which is composed of a nonvolatile memory such as an HDD or flash memory, has been used as an example of a computer-readable medium for storing the program 120 for controlling boundary extraction and machine learning model learning of the present invention, but is not limited to this. Other computer-readable media that can be used include other nonvolatile memories such as MRAM and portable recording media such as CD-ROMs and DVD discs. Furthermore, a carrier wave can also be used as a medium for providing program data according to the present invention via a communication line. In addition, the specific configurations, contents and procedures of the processing operations, etc. shown in the above embodiments can be modified as appropriate without departing from the spirit of the present invention. The scope of the present invention includes the scope of the invention described in the claims and its equivalents.

[0059] As described above, the information processing device 1 of this embodiment analyzes a microtopography representation and extracts a predetermined topographical feature, for example, the boundary line of an area where a collapsed area appears. The information processing device 1 includes a memory unit 12 that stores a trained model 121 that has been trained in advance to output a boundary area in which the line width of the boundary line of an area where a topographical feature appears in the microtopography representation in response to an input of the microtopography representation, and a control unit 11. The control unit 11, as an acquisition unit, inputs the microtopography representation to be analyzed to the trained model 121 and acquires the boundary area output from the trained model 121. The control unit 11, as an extraction unit, thins the acquired boundary area and extracts the boundary line of an area where a topographical feature appears in the microtopography representation to be analyzed. In this way, by performing a two-stage process of extracting a boundary area using the trained model 121 and then thinning the boundary area to extract a boundary line, the information processing device 1 can extract the boundary line of a topographical feature with higher accuracy.

[0060] The extraction means may also extract a vector-format boundary line along the line image obtained by thinning the boundary area acquired by the acquisition means. By extracting the boundary line as vector-format data, boundary line data that is independent of scale and resolution can be obtained.

[0061] The information processing device 1 may also be connected to or connectable to a database device 21 serving as training data storage means for storing multiple training microtopographic representations used for the training and training boundary lines indicating the boundaries of areas where the topographical features appear in each of the training microtopographic representations, or an optical reader 22 for reading portable recording media. The control unit 11 may, as expansion means, generate a training boundary area by performing a process for expanding the line width of each training boundary line. The control unit 11 may, as learning means, update the machine learning model 122 so as to minimize the error between the output result when each training microtopographic representation is input to the machine learning model 122 and the training boundary area corresponding to the training microtopographic representation, thereby generating the trained model 121. The update may be performed, for example, by updating parameters using an error backpropagation algorithm using gradient descent. In this way, the information processing device 1 may train the machine learning model 122 on its own device to obtain the trained model 121. In this case, contrary to when it is in use, the boundary lines are made thicker to be used as correct answer data for training, thereby allowing the machine learning model 122 to be trained appropriately.

[0062] The learning means may also weight an in-region error, which indicates that an output result corresponding to a pixel within the learning boundary region is not a pixel of the boundary region, more heavily than an out-region error, which indicates that an output result corresponding to a pixel outside the learning boundary region is a pixel of the boundary region, and calculate the error by adding the in-region error and the out-region error. Because the boundary region still occupies a significantly small proportion of the area in the image even after being bolded, evaluating the in-region error and the out-region error with the same weighting is likely to underestimate the in-region error. Therefore, by weighting the in-region error more heavily than the out-region error, the machine learning model 122 is trained to extract the boundary region more efficiently and accurately.

[0063] Furthermore, the boundary of the area where topographical features appear may be a U-shaped boundary representing the boundary of a collapsed area. Compared to the boundary on the mountain side of a collapsed area, which tends to appear clearly on a microtopographic representation map, the boundary on the valley side is unclear, making it difficult to identify the entire area. Furthermore, for disaster prevention and other measures, the upstream boundary is often particularly important. Therefore, being able to more easily identify the U-shaped boundary of a collapsed area is highly useful.

[0064] Furthermore, the boundary line extraction method by the control unit 11 of this embodiment includes the following steps: (1) a storage step of storing a trained model 121 that has been trained in advance to output a boundary line area in which the line width of the boundary line of an area where a topographical feature appears in a microtopographical representation map is expanded in response to an input of the microtopographical representation map; (2) an acquisition step of inputting the microtopographical representation map to be analyzed into the trained model 121 and acquiring the boundary line area output from the trained model 121; and (3) an extraction step of thinning the boundary line area acquired in the acquisition step and extracting the boundary line of an area where a topographical feature appears in the microtopographical representation map to be analyzed. According to this boundary line extraction method, the control unit 11 can utilize the capabilities of the trained model 121 to more accurately extract the desired boundary line.

[0065] Furthermore, by installing and executing the program 120 relating to the above-mentioned boundary line extraction method on a computer, it is possible to obtain the boundary lines of topographical features more accurately using the trained model 121 without requiring special hardware. [Explanation of symbols]

[0066] 1. Information processing equipment 11 Control section 12 Storage section 120 Programs 121 trained models 122 Machine Learning Models 13 Input / Output Interface 131 Connection terminal 132 Communications Department 14 Display section 15 Operation reception section 21 Database device 22 Optical reading device 201 Microtopographical Representation Map Data 202 Boundary line data for learning 203 Boundary Region Data for Learning

Claims

1. A boundary line extraction device that analyzes a microtopographical representation map and extracts a boundary line of an area where a predetermined topographical feature appears, a storage means for storing a trained model that has been trained in advance to output a boundary area in which the line width of the boundary line of the area in which the topographical feature appears in the microtopographical representation map is expanded in response to the input of the microtopographical representation map; an acquisition means for inputting a microtopography representation map of an analysis target into the trained model and acquiring the boundary area output from the trained model; an extraction means for thinning the boundary area acquired by the acquisition means and extracting a boundary line of an area in which the topographical feature appears in the microtopographical representation map of the analysis target; A boundary extraction device comprising:

2. the extraction means extracts the boundary line in a vector format along a line image obtained by thinning the boundary area acquired by the acquisition means; 2. The boundary line extraction device according to claim 1.

3. a learning data storage means for storing a plurality of learning microtopographic representation maps used in the learning and learning boundary lines indicating boundary lines of areas in which the topographical features appear in each of the learning microtopographic representation maps; an expansion means for expanding the line width of each of the learning boundaries to generate a learning boundary area; a learning means for updating the machine learning model so as to minimize an error between an output result when the training microtopography representation map is input into the machine learning model and the training boundary area corresponding to the training microtopography representation map, thereby generating the trained model; The boundary extraction device according to claim 1 , comprising:

4. the learning means weights an in-area error, which indicates that the output result corresponding to a pixel within the learning boundary area is not a pixel of the boundary area, more heavily than an out-of-area error, which indicates that the output result corresponding to a pixel outside the learning boundary area is a pixel of the boundary area, and calculates the error by adding the in-area error and the out-of-area error.

4. The boundary line extraction device according to claim 3.

5. 5. The boundary extraction device according to claim 1, wherein the boundary of the area where the topographical feature appears is a U-shaped boundary representing a boundary of a collapsed area.

6. A boundary line extraction method in which a control unit analyzes a microtopographic representation map and extracts a boundary line of an area in which a predetermined topographical feature appears, a storage step in which the control unit stores a trained model that has been trained in advance to output a boundary area in which the line width of the boundary line of the area in which the topographical feature appears in the microtopographical representation map is expanded in response to the input of the microtopographical representation map; an acquisition step in which the control unit inputs a microtopography representation map to be analyzed into the trained model and acquires the boundary area output from the trained model; an extraction step in which the control unit thins the boundary area acquired in the acquisition step and extracts a boundary line of an area in which the topographical feature appears in the microtopographical representation map of the analysis target; A boundary extraction method including:

7. A computer that analyzes the microtopographic representation map and extracts the boundary lines of areas where predetermined topographical features appear, a storage means for storing a trained model that has been trained in advance to output a boundary area in which the line width of the boundary line of the area in which the topographical feature appears in the microtopographical representation map is expanded in response to the input of the microtopographical representation map; an acquisition means for inputting a microtopography representation map of an analysis target into the trained model and acquiring the boundary area output from the trained model; an extraction means for thinning the boundary area acquired by the acquisition means and extracting a boundary line of an area in which the topographical feature appears in the microtopographical representation map of the analysis target; A program that functions as a

Images