Inundation height calculation system, inundation height calculation method and inundation height calculation program
Patent Information
- Application Number
- JP2022118506
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2022-07-26
- Publication Date
- 2024-11-20
- Estimated Expiration
- 2042-02-03
AI Technical Summary
Existing methods for determining flood height require manual measurement with a tape measure, which is time-consuming and inefficient during large-scale flood damage assessments.
An inundation height calculation system that detects a reference object from image data, specifies wall positions, calculates a unit height, and determines flood height based on pixel analysis, utilizing a camera and computational methods to automate the process.
Enables rapid and accurate measurement of flood height from image data, reducing the time and effort required for flood damage assessments.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to a technique for calculating the inundation height from image data.
Background Art
[0002] When a flood occurs, an insurance company dispatches an appraiser to the building of the insurance policyholder. Then, the appraiser determines whether the damage situation meets the requirements for paying insurance money.
[0003] The requirements for paying insurance money are, for example, to meet at least one of the following two conditions. Condition 1: There is flooding on the floor. Condition 2: There is flooding at a certain height or more (for example, 45 centimeters or more) from the ground surface. Regarding Condition 1, it can be easily determined by visually inspecting the inside of the building or checking the image data taken of the inside of the building. However, regarding Condition 2, the appraiser needs to measure the inundation height using a measuring tool such as a measuring tape, which is time-consuming.
[0004] Patent Document 1 describes specifying the degree of damage from the differences by comparing a photographed image of an accident vehicle with a registered image of the vehicle type of the accident vehicle.
Prior Art Documents
Patent Documents
[0005]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0006] When applying the technology described in Patent Document 1 to measuring flood height, the image data must contain information that allows for the identification of the height. Therefore, for example, it would be necessary to photograph a measuring tape or similar object along with the flooded building. Furthermore, human visual verification of the measuring tape would be required. In the event of a flood, there is a possibility of a large number of insurance claim applications being filed. Performing human visual verification would delay the processing of these applications. This disclosure aims to enable the easy measurement of flood level. [Means for solving the problem]
[0007] The flood level calculation system related to this disclosure is: A detection unit that detects reference objects from image data including water inundation lines on a wall surface, A wall position identification unit that identifies the wall bottom position, which is the position on the wall surface corresponding to the bottom of the reference object detected by the detection unit, and the wall top position, which is the position on the wall surface corresponding to the top of the reference object, A unit height calculation unit calculates the unit height, which is the height per unit pixel, from the number of pixels in the image data from the wall bottom position to the wall top position identified by the wall position identification unit and the actual height, which is the actual height of the reference object. A flood height calculation unit calculates the flood height, which is the distance from the floor surface to the flood line, based on the number of pixels from the floor surface to the flood line in the image data and the unit height calculated by the unit height calculation unit. It is equipped with.
[0008] The aforementioned reference object is an object with thickness, The wall position identification unit determines the position obtained by shifting the lower end of the reference object in the image data upward by the bottom image depth, which is the length of the bottom surface of the reference object in the image data. Identify it as the base of the wall.
[0009] The aforementioned reference material is a bottle, The wall position identification unit calculates the bottom image depth by multiplying the actual image ratio obtained by dividing the lid image depth, which is the length in the depth direction of the upper surface of the bottle lid portion in the image data, by the actual lid depth, which is the actual length in the depth direction of the upper surface of the bottle lid portion, by the actual bottom depth, which is the actual length in the depth direction of the bottom surface of the bottle.
[0010] The wall position identification unit determines the depth of the lid image by multiplying the vertical width of the lid portion identified from the image data by the top surface ratio.
[0011] The wall position identification unit identifies the position on the wall as the position obtained by shifting the position of the upper end of the bottle upward by an amount obtained by multiplying the length from the lower end to the upper end of the bottle in the image data by the displacement ratio.
[0012] The aforementioned flood level calculation system further, Image acquisition unit acquires the camera image obtained by the camera when the shooting conditions, which are the conditions for the camera position to take the picture, are met, as the image data. It is equipped with.
[0013] The aforementioned flood level calculation system further, Guide display unit that displays a guide superimposed on the camera image obtained by the aforementioned camera. Equipped with, The image acquisition unit acquires a camera image as image data in which the water inundation line on the wall surface is shown, while the reference object is following the guide displayed by the guide display unit.
[0014] The aforementioned guide indicates the upper and lower limits in the height direction. The image acquisition unit acquires a camera image showing the water inundation line on the wall surface as image data, with the reference object positioned between the upper limit position and the lower limit position.
[0015] The image acquisition unit acquires the image data when the angle of the camera relative to the wall surface is a specified angle.
[0016] The detection unit detects the reference object from an enlarged image obtained by enlarging the range indicated by the guide.
[0017] The flooding height calculation system further includes a type identification unit that identifies the type of the bottle from the size of the lid portion of the bottle detected by the detection unit and the size of the body portion of the bottle and the unit height calculation unit uses the actual height corresponding to the type identified by the type identification unit.
[0018] The type identification unit identifies the type of the bottle from the ratio of the lateral width of the lid portion to the lateral width of the body portion.
[0019] The flooding height calculation system further includes a correction unit that corrects the flooding height by a correction amount corresponding to the flooding height calculated by the flooding height calculation unit and
[0020] The flooding height calculation system further includes a reliability calculation unit that calculates the reliability of the flooding height, and a data transmission unit that transmits the reliability calculated by the reliability calculation unit together with the flooding height and [[ID=3s]]
[0021] The flooding height calculation method according to the present disclosure is a computer detects a reference object from image data including a flooding line on a wall surface, the computer identifies a wall bottom position that is a position on the wall surface corresponding to the bottom of the reference object and a wall top position that is a position on the wall surface corresponding to the top of the reference object, the computer calculates a unit height, which is the height per pixel, from the number of pixels in the image data from the wall bottom position to the wall top position and the actual height of the reference object, The computer calculates the flood height, which is the distance from the floor surface to the flood line, from the number of pixels in the image data from the floor surface to the flood line and the unit height.
[0022] The flood level calculation program related to this disclosure is: A detection process that detects reference objects from image data including water inundation lines on a wall surface, A wall position identification process that identifies the wall bottom position, which is the position on the wall surface corresponding to the bottom of the reference object detected by the detection process, and the wall top position, which is the position on the wall surface corresponding to the top of the reference object. A unit height calculation process calculates the unit height, which is the height per unit pixel, from the number of pixels in the image data from the wall bottom position to the wall top position identified by the wall position identification process and the actual height, which is the actual height of the reference object. A flood height calculation process calculates the flood height, which is the distance from the floor to the flood line, based on the number of pixels from the floor surface to the flood line in the image data and the unit height calculated by the unit height calculation process. The computer will function as a flood level calculation device. [Effects of the Invention]
[0023] In this disclosure, the flood level is calculated based on reference objects detected from image data. This makes it possible to easily measure the flood level. [Brief explanation of the drawing]
[0024] [Figure 1] Configuration diagram of the flood level calculation system 100 according to Embodiment 1. [Figure 2] Configuration diagram of user terminal 10 according to Embodiment 1. [Figure 3] Configuration diagram of the flood level calculation device 20 according to Embodiment 1. [Figure 4] A flowchart illustrating the overall processing of the flood level calculation system 100 according to Embodiment 1. [Figure 5] Flowchart of the imaging process according to Embodiment 1. [Figure 6] An explanatory diagram of the flood line designation process according to Embodiment 1. [Figure 7] An explanatory diagram of the unit height calculation process according to Embodiment 1. [Figure 8] Configuration diagram of user terminal 10 according to modified example 1. [Figure 9] Configuration diagram of the flood level calculation device 20 according to Modification Example 1. [Figure 10] Configuration diagram of the flood level calculation device 20 according to Embodiment 2. [Figure 11] An explanatory diagram showing how the reference object appears in Embodiment 2. [Figure 12] Flowchart of the unit height calculation process according to Embodiment 2. [Figure 13] An explanatory diagram of the wall bottom position identification process according to Embodiment 2. [Figure 14] An explanatory diagram of the wall bottom position identification process according to Embodiment 2. [Figure 15] An explanatory diagram of the wall bottom position identification process according to Embodiment 2. [Figure 16] An explanatory diagram of the wall-mounted position identification process according to Embodiment 2. [Figure 17] An explanatory diagram of the height calculation process according to Embodiment 2. [Figure 18] Configuration diagram of the user terminal 10 according to Embodiment 3. [Figure 19] Flowchart of the imaging process according to Embodiment 3. [Figure 20] An explanatory diagram of the guide selection process according to Embodiment 3. [Figure 21] An explanatory diagram of the image acquisition process according to Embodiment 3. [Figure 22] Diagram illustrating the detection process according to Embodiment 4. [Figure 23] Configuration diagram of the flood level calculation device 20 according to Embodiment 5. [Figure 24] An explanatory diagram of the types of reference materials according to Embodiment 5. [Figure 25] Flowchart of the unit height calculation process according to Embodiment 5. [Figure 26] Configuration diagram of the flood level calculation device 20 according to Embodiment 6. [Figure 27] Flowchart of the flood level calculation process according to Embodiment 6. [Figure 28] An explanatory diagram of the correction process according to Embodiment 6. [Figure 29] An explanatory diagram of the correction process according to Embodiment 6. [Figure 30] An explanatory diagram of the correction process according to Embodiment 6. [Figure 31] Configuration diagram of the flood level calculation device 20 according to Embodiment 7. [Figure 32] Flowchart of the flood level calculation process according to Embodiment 7. [Modes for carrying out the invention]
[0025] Embodiment 1. ***Explanation of the structure*** Referring to Figure 1, the configuration of the flood height calculation system 100 according to Embodiment 1 will be described. The flood level calculation system 100 comprises a user terminal 10 and a flood level calculation device 20. The user terminal 10 and the flood level calculation device 20 are connected via a transmission line 90. User terminal 10 is a computer, such as a smartphone or tablet, used by a user. User terminal 10 is a computer for taking pictures of the flooded area and acquiring image data. Flood height calculation device 20 is a computer, such as a server in a cloud system. Flood height calculation device 20 is a computer that calculates the flood height based on the image data acquired by user terminal 10. Transmission line 90 is, in specific examples, the internet. The flood level calculation device 20 is connected to the insurance company's server via a transmission line 91. A specific example of the transmission line 91 is the internet.
[0026] Embodiment 1 will be described assuming the configuration shown in Figure 1. However, the functions of the flood height calculation device 20 may also be provided by the user terminal 10. In this case, the flood height calculation system 100 will not include the flood height calculation device 20, but will only include the user terminal 10.
[0027] Referring to Figure 2, the configuration of the user terminal 10 according to Embodiment 1 will be described. The user terminal 10 comprises hardware including a processor 11, memory 12, storage 13, and a communication interface 14. The processor 11 is connected to and controls other hardware via signal lines. The user terminal 10 is connected to the camera 141 via the communication interface 14. Alternatively, the camera 141 is internally connected to and integrated with the user terminal 10.
[0028] The user terminal 10 comprises an image acquisition unit 111, a flood line setting unit 112, and an image transmission unit 113 as functional components. The functions of each functional component of the user terminal 10 are implemented by software. Storage 13 stores programs that implement the functions of each functional component of the user terminal 10. These programs are loaded into memory 12 by the processor 11, and then... This is executed by the processor 11. This enables the functionality of each functional component of the user terminal 10.
[0029] Referring to Figure 3, the configuration of the flood height calculation device 20 according to Embodiment 1 will be described. The flood level calculation device 20 comprises hardware including a processor 21, memory 22, storage 23, and a communication interface 24. The processor 21 is connected to and controls the other hardware via signal lines.
[0030] The flood level calculation device 20 comprises, as functional components, a detection unit 211, a unit height calculation unit 212, a flood level calculation unit 213, and a data transmission unit 214. The functions of each functional component of the flood level calculation device 20 are implemented by software. Storage 23 stores programs that implement the functions of each functional component of the flood height calculation device 20. These programs are loaded into memory 22 by the processor 21 and executed by the processor 21. This enables the implementation of the functions of each functional component of the flood height calculation device 20.
[0031] Processors 11 and 21 are integrated circuits (ICs) that perform processing. IC stands for Integrated Circuit. Specific examples of processors 11 and 21 include CPUs, DSPs, and GPUs. CPU stands for Central Processing Unit. DSP stands for Digital Signal Processor. GPU stands for Graphics Processing Unit.
[0032] Memory 12 and 22 are memory devices that temporarily store data. Specific examples of memory 12 and 22 are SRAM and DRAM. SRAM stands for Static Random Access Memory. DRAM stands for Dynamic Random Access Memory.
[0033] Storage 13 and 23 are storage devices for storing data. A concrete example of storage 13 and 23 is an HDD. HDD stands for Hard Disk Drive. Alternatively, storage 13 and 23 may be portable recording media such as SD® memory cards, CompactFlash®, NAND flash, flexible disks, optical disks, compact disks, Blu-ray® discs, and DVDs. SD stands for Secure Digital. DVD stands for Digital Versatile Disk.
[0034] Communication interfaces 14 and 24 are interfaces for communicating with external devices. Specific examples of communication interfaces 14 and 24 include Ethernet®, USB, and HDMI® ports. USB stands for Universal Serial Bus. HDMI stands for High-Definition Multimedia Interface.
[0035] In Figure 2, only one processor 11 was shown. However, there may be multiple processors 11, and multiple processors 11 may work together to execute programs that implement each function. Similarly, in Figure 3, only one processor 21 was shown. However, there may be multiple processors 21, and multiple processors 21 may work together to execute programs that implement each function.
[0036] ***Explanation of operation*** Referring to Figures 4 to 7, the operation of the flood level calculation system 100 according to Embodiment 1 will be explained. The operation procedure of the flood height calculation system 100 according to Embodiment 1 corresponds to the flood height calculation method according to Embodiment 1. Furthermore, the program that implements the operation of the flood height calculation system 100 according to Embodiment 1 corresponds to the flood height calculation program according to Embodiment 1.
[0037] Referring to Figure 4, the overall processing flow of the flood height calculation system 100 according to Embodiment 1 will be explained. (Step S1: Image capture process) The image acquisition unit 111 of the user terminal 10 acquires image data including flood lines on the wall surface. Flood lines are traces of water damage left on the wall surface. Flood lines indicate the upper limit of the flooded area. In this process, the image acquisition unit 111 acquires image data so that it includes a reference object. The reference object is a predetermined object whose size can be identified. It is desirable that the reference object is easily obtainable and has a defined size standard. Specific examples of reference objects include bottles such as PET bottles or beer bottles, paper cartons such as toilet paper or milk cartons, and aluminum cans of 350ml or similar sizes. Alternatively, the reference object may be a box or similar item sent to the policyholder in advance by the insurance company.
[0038] The imaging process according to Embodiment 1 will be described with reference to Figure 5. As a prerequisite, an application program for calculating the flood level is executed. This activates the functions of the image acquisition unit 111, the flood line setting unit 112, and the image transmission unit 113. In addition, the camera 141 on the user terminal 10 is activated via the application program.
[0039] (Step S11: Image acquisition process) The image acquisition unit 111 acquires image data by taking a picture with the camera 141 that captures the flood line and the reference object simultaneously. In this case, the reference object is assumed to be placed on the floor.
[0040] (Step S12: Floodline designation process) The flood line setting unit 112 sets the position of the flood line in the image data acquired in step S11. Specifically, as shown in Figure 6, the flood line setting unit 112 sets the position of the specified flood line according to the user's operation. For example, the position of the flood line is specified by dragging and moving the line for setting the flood line on the image data acquired in step S11.
[0041] (Step S13: Data transmission process) The image transmission unit 113 transmits the image data acquired in step S11 and the location information of the flood line set in step S12 to the flood level calculation device 20. At this time, in step S13, the image transmission unit 113 also transmits information that can identify the insurance contract, such as the insurance policyholder's identification information, along with the image data.
[0042] (Step S2: Detection process) The detection unit 211 of the flood level calculation device 20 detects reference objects from image data. Specifically, the detection unit 211 inputs the image data transmitted in step S13 into an object detection model to detect reference objects. The object detection model is a model constructed using a neural network or the like.
[0043] (Step S3: Unit height calculation process) The unit height calculation unit 212 of the flood height calculation device 20 identifies the number of pixels in the height direction in the image data of the reference object detected in step S2. As shown in Figure 7, the number of pixels in the height direction of the reference object is the number of pixels from the bottom end to the top end of the reference object. For example, in step S2, If a detection frame is obtained that surrounds a known reference object, the number of pixels in the height direction of the reference object is the number of pixels from the bottom edge to the top edge of the detection frame. The unit height calculation unit 212 calculates the unit height, which is the height per unit pixel, from the number of pixels in the height direction and the actual height, which is the actual height of the reference object. Since the reference object is an object whose size can be determined, the actual height of the reference object is known. In Embodiment 1, the unit height calculation unit 212 calculates the unit height, which is the height per pixel, by dividing the actual height by the number of pixels, as shown in Equation 1. <Expression 1> Unit height (centimeters) = Actual height (centimeters) / Number of pixels
[0044] (Step S4: Flood level calculation process) The flood level calculation unit 213 of the flood level calculation device 20 identifies the number of pixels from the floor surface to the flood line in the image data. In Embodiment 1, as shown in Figure 7, the position of the floor surface is the lower end of the reference object. The flood level calculation unit 213 calculates the flood level, which is the distance from the floor surface to the flood line, from the identified number of pixels and the unit height calculated in step S3. Specifically, the flood level calculation unit 213 calculates the flood level by multiplying the number of pixels from the floor surface to the flood line by a unit height, as shown in Equation 2. <Expression 2> Flood height (centimeters) = number of pixels × unit height (centimeters)
[0045] (Step S5: Data transmission process) The data transmission unit 214 of the flood height calculation device 20 transmits the flood height calculated in step S4 and the insurance policyholder identification information received in step S13 to the insurance company's server.
[0046] Multiple types of reference objects may exist. In this case, in step S1, the user specifies which reference object to use. Alternatively, in step S2, it is determined which reference object was detected. Then, in step S3, the actual height corresponding to the type of reference object used is used. The user terminal 10 may be configured to connect directly to the insurance company's server via a transmission line 90, and when the application program is started, it may send the insurance policyholder's identification information, etc., to the insurance company's server, and in step S5, it may be configured to send only the flood level from the flood level calculation device 20 to the insurance company's server. In this case, the insurance company's server performs a comparison of the flood level received in step S5 with the insurance policyholder's identification information, etc., which has been received in advance. In step S13, the user terminal 10 may be configured to send image data and other information, as well as the insurance policyholder's identification information, to the insurance company's server, and the flood depth calculation device 20 may be configured to receive image data and other information from the insurance company's server. The flood depth calculation device 20 then executes the processes from step S2 onward as described above.
[0047] ***Effects of Embodiment 1*** As described above, the flood level calculation system 100 according to Embodiment 1 acquires image data including reference objects along with the flood line. The flood level calculation system 100 identifies a unit height based on the reference object and calculates the flood level from the unit height. This makes it possible to easily measure the flood level.
[0048] ***Other configurations*** <Example 1> In Embodiment 1, each functional component was implemented in software. However, in Modification 1, each functional component may be implemented in hardware. The differences between this Modification 1 and Embodiment 1 will be explained below.
[0049] Referring to Figure 8, the configuration of the user terminal 10 according to Modification Example 1 will be explained. When each functional component is implemented in hardware, the user terminal 10 includes an electronic circuit 15 instead of a processor 11, memory 12, and storage 13. The electronic circuit 15 is a dedicated circuit that implements the functions of each functional component, as well as the functions of the memory 12 and storage 13.
[0050] Referring to Figure 9, the configuration of the flood height calculation device 20 according to Modification 1 will be explained. When each functional component is implemented in hardware, the flood level calculation device 20 includes an electronic circuit 25 instead of a processor 21, memory 22, and storage 23. The electronic circuit 25 is a dedicated circuit that implements the functions of each functional component, as well as the functions of the memory 22 and storage 23.
[0051] Electronic circuits 15 can include single circuits, complex circuits, programmed processors, parallel programmed processors, logic ICs, GAs, ASICs, and FPGAs. GA stands for Gate Array. ASIC stands for Application Specific Integrated Circuit. FPGA stands for Field-Programmable Gate Array. Each functional component may be implemented in a single electronic circuit 15, or each functional component may be implemented by distributing them across multiple electronic circuits 15.
[0052] <Modification 2> As a second variation, some of the functional components may be implemented in hardware, while others may be implemented in software.
[0053] The processor 11, memory 12, storage 13, and electronic circuit 15 are collectively referred to as the processing circuit. In other words, the function of each functional component is realized by the processing circuit.
[0054] Embodiment 2. Embodiment 2 differs from Embodiment 1 in that it calculates the unit height by identifying the position of the reference object on the wall surface. Embodiment 2 explains this difference, while omitting explanations of the same points.
[0055] ***Explanation of the structure*** Referring to Figure 10, the configuration of the flood height calculation device 20 according to Embodiment 2 will be described. The flood level calculation device 20 differs from the flood level calculation device 20 shown in Figure 3 in that it includes a wall location identification unit 215 as a functional component. The function of the wall location identification unit 215 is implemented by software or hardware, similar to other functional components.
[0056] ***Explanation of operation*** Referring to Figures 11 to 17, the operation of the flood level calculation system 100 according to Embodiment 2 will be described. The operation procedure of the flood height calculation system 100 according to Embodiment 2 corresponds to the flood height calculation method according to Embodiment 2. Furthermore, the program that implements the operation of the flood height calculation system 100 according to Embodiment 2 corresponds to the flood height calculation program according to Embodiment 2.
[0057] In Embodiment 2, the reference object is described as a bottle. In Embodiment 2, the reference object is described as a PET bottle of a specific size. In Embodiment 2, the PET bottle is placed on the floor with the bottle in contact with the wall. Since the reference object is a plastic bottle, the reference object has thickness in the depth direction. Therefore, as shown in Figure 11, in the image data obtained by photographing the reference object from diagonally above, the lower end of the reference object is The distance from the bottom to the top is affected by the thickness of the reference object. In other words, the distance from the bottom to the top of the reference object does not accurately represent the height of the reference object. Therefore, in Embodiment 2, the flood height calculation system 100 identifies the bottom wall position 31, which is the position on the wall surface corresponding to the bottom of the reference object, and the top wall position 32, which is the position on the wall surface corresponding to the top of the reference object. The flood height calculation system 100 then calculates the unit height from the bottom and top positions on the wall surface.
[0058] In Embodiment 2, the approximate shooting angle of the reference object is assumed to be predetermined.
[0059] Referring to Figure 12, the unit height calculation process according to Embodiment 2 (step S3 in Figure 4) will be explained. (Step S31: Wall bottom position identification process) The wall position identification unit 215 of the flood level calculation device 20 identifies the wall bottom position 31. There are two methods for determining the wall base position 31: Method 1 and Method 2.
[0060] <Method 1> Method 1 is a method based on geometric estimation. The reference object, a plastic bottle, is placed on the floor in contact with the wall. Therefore, as shown in Figure 13, the upper end position 33 of the bottom surface of the plastic bottle becomes the wall bottom position 31. However, the upper end position 33 of the bottom surface of the plastic bottle is not captured in the image data. Therefore, the wall position identification unit 215 estimates the upper end position 33 of the bottom surface of the plastic bottle geometrically and identifies the wall bottom position 31. Specifically, the wall position identification unit 215 identifies the wall bottom position 31 as the position obtained by shifting the lower end 35 of the plastic bottle in the image data upward by the bottom image depth 34, which is the length of the bottom surface of the plastic bottle in the image data.
[0061] As shown in Figure 14, the image data can be considered to be roughly a parallel projection image. Therefore, as described above, the image ratio, which is the ratio of the actual size of the lid portion to the size in the image data, is roughly equal to the ratio of the actual size of the bottom portion to the size in the image data. The wall position identification unit 215 then uses this relationship to calculate the bottom image depth 34 and identify the wall bottom position 31.
[0062] Specifically, as shown in Figure 14, the wall position identification unit 215 calculates the actual image ratio by dividing the lid image depth 36 by the actual lid depth 37. The lid image depth 36 is the length in the depth direction of the upper surface of the lid portion in the image data. The method for determining the lid image depth 36 will be described later. The actual lid depth 37 is the actual length in the depth direction of the upper surface of the lid portion of the PET bottle. Since the reference object is an object whose size can be determined, the actual lid depth 37 is known. The wall position identification unit 215 calculates the bottom image depth 34 by multiplying the actual bottom depth 38, which is the actual length of the bottom surface of the bottle in the depth direction, by the actual image ratio. In other words, the wall position identification unit 215 calculates the bottom image depth 34 as shown in Equation 3. <Expression 3> Bottom image depth 34 = Actual bottom depth 38 × Lid image depth 36 / Actual lid depth 37 The wall position identification unit 215 then identifies the wall bottom position 31 as the position obtained by shifting the lower end 35 of the PET bottle in the image data upward by the depth 34 of the bottom image.
[0063] This explains how to determine the depth 36 of the lid image. It is possible to detect the lid portion from image data by using an object detection model, etc. However, it can be difficult to identify only the top surface of the lid portion. Therefore, as shown in Figure 15, the wall position identification unit 215 calculates the lid image depth 36 by multiplying the lid height 39, which is the vertical width of the entire lid portion, by the top surface ratio, which represents the proportion of the top surface portion to the entire lid portion. In other words, the wall position identification unit 215 calculates the lid image depth 36 as shown in Equation 4. <Expression 4> Lid image depth 36 = Lid height 39 × Top surface ratio
[0064] The wall position identification unit 215 calculates the top surface ratio using a regression line based on the ratio of the lid's vertical width 39 to the lid's horizontal width 40. Specifically, the wall position identification unit 215 calculates the top surface ratio as shown in Equation 5. Here, the slope and intercept are set in advance through simulations or the like. <Formula 5> Top surface ratio = Slope × (Lid height 39 / Lid width 40) + Intersection
[0065] Furthermore, the wall position identification unit 215 may have its top surface ratio set in advance. If the shooting angle of the reference object is roughly determined, the top surface ratio is also roughly determined. Therefore, it is possible to determine and set the top surface ratio in advance.
[0066] <Method 2> Method 2 is a statistical estimation method. If the shooting angle of the reference object is generally fixed, then the depth ratio, which is the ratio of the height in the image data of the PET bottle to the depth 34 of the bottom image, is also generally fixed. The height in the image data of the PET bottle is the length from the bottom end to the top end of the PET bottle. In other words, the height in the image data of the PET bottle is the number of pixels in the height direction of the reference object as described in Embodiment 1. Therefore, the wall position identification unit 215 pre-sets the depth ratio, assuming that the shooting angle of the reference object is roughly determined. The wall position identification unit 215 calculates the bottom image depth 34 by multiplying the height in the image data of the PET bottle by the depth ratio. Then, the wall position identification unit 215 identifies the wall bottom position 31 as the position obtained by shifting the lower end 35 of the PET bottle in the image data upward by the bottom image depth 34.
[0067] (Step S32: Wall-top position identification process) The wall position identification unit 215 of the flood level calculation device 20 identifies the wall position 32. Specifically, the wall position identification unit 215 identifies the wall position 32 by statistical estimation, similar to method 2 in step S31. As shown in Figure 16, if the shooting angle of the reference object is roughly determined, the displacement ratio, which is the ratio of the height in the image data of the PET bottle to the displacement amount 41, is also roughly determined. The displacement amount 41 is the distance between the upper end 42 of the PET bottle and the wall position 32. Therefore, the wall position identification unit 215 pre-sets the displacement ratio, assuming that the shooting angle of the reference object is roughly determined. The wall position identification unit 215 calculates the displacement amount 41 by multiplying the height in the image data of the PET bottle by the displacement ratio. Then, the wall position identification unit 215 identifies the position where the upper end 42 of the PET bottle in the image data is shifted upward by the displacement amount 41 as the wall top position 32.
[0068] (Step S33: Height calculation process) As shown in Figure 17, the unit height calculation unit 212 of the flood level calculation device 20 determines the number of pixels in the height direction in the image data from the wall bottom position 31 determined in step S31 to the wall top position 32 determined in step S32. The unit height calculation unit 212 calculates the unit height, which is the height per unit pixel, from the number of pixels in the height direction and the actual height, which is the actual height of the reference object.
[0069] In the flood level calculation process (step S4 in Figure 4), the flood level calculation unit 213 of the flood level calculation device 20 uses the wall bottom position 31 identified in step S31 as the floor position and identifies the number of pixels from the floor to the flood line. Then, the flood level calculation unit 213 calculates the flood level, which is the distance from the floor to the flood line, from the identified number of pixels and the unit height calculated in step S3.
[0070] ***Effects of Embodiment 2*** As described above, the flood height calculation system 100 according to Embodiment 2 calculates the unit height by identifying the position on the wall surface of a reference object. This allows for accurate calculation of the unit height. As a result, the flood height can be calculated with high accuracy.
[0071] Furthermore, the flood level calculation system 100 according to Embodiment 2 accurately identifies the position of the bottom surface. This allows for accurate calculation of the flood level.
[0072] ***Other configurations*** <Variation 3> In Embodiment 2, the wall bottom position 31 was determined by calculation. However, the wall bottom position 31 may be specified by the user. In this case, the wall bottom position 31 is set along with the flood line in step S1 of Figure 4. The wall bottom position 31 is then transmitted to the flood height calculation device 20 along with the image data, etc.
[0073] Similarly, in Embodiment 2, the wall-top position 32 was determined by calculation. However, the wall-top position 32 may be specified by the user. In this case, the wall-top position 32 is set along with the flood line in step S1 of Figure 4. The wall-top position 32 is then transmitted to the flood height calculation device 20 along with the image data, etc.
[0074] The flood level calculation device 20 may determine whether the wall bottom position 31 set by the user is valid. Specifically, the wall position identification unit 215 identifies the wall bottom position 31 using the method described in Embodiment 2. The wall position identification unit 215 then determines whether the difference between the wall bottom position 31 set by the user and the identified wall bottom position 31 is within a threshold. If the difference is within the threshold, the wall position identification unit 215 determines that the wall bottom position 31 set by the user is valid. On the other hand, if the difference exceeds the threshold, the wall position identification unit 215 determines that the wall bottom position 31 set by the user is not valid. If the flood level calculation device 20 determines that the wall bottom position 31 is not valid, it may request the user to reset the wall bottom position 31. The flood level calculation device 20 may similarly determine whether the wall-top position 32 set by the user is valid or not. If the flood level calculation device 20 determines that it is not valid, it may request the user to reset the wall-top position 32.
[0075] Embodiment 3. Embodiment 3 differs from Embodiments 1 and 2 in that it acquires camera images as image data obtained by camera 141 when the shooting conditions are met. The shooting conditions are the conditions for the camera position during shooting. Embodiment 3 will explain these differences, and will omit explanations of the same points. Embodiment 3 describes a case in which a modification has been made to Embodiment 1. However, it is also possible to modify Embodiment 2.
[0076] ***Explanation of the structure*** Referring to Figure 18, the configuration of the user terminal 10 according to Embodiment 3 will be described. User terminal 10 differs from user terminal 10 shown in Figure 2 in that it includes a guide display unit 114 as a functional component. The function of the guide display unit 114 is implemented by software or hardware, as with other functional components.
[0077] ***Explanation of operation*** Referring to Figures 19 to 21, the operation of the flood height calculation system 100 according to Embodiment 3 will be described. The operation procedure of the flood height calculation system 100 according to Embodiment 3 corresponds to the flood height calculation method according to Embodiment 3. Furthermore, the program that implements the operation of the flood height calculation system 100 according to Embodiment 3 corresponds to the flood height calculation program according to Embodiment 3.
[0078] Referring to Figure 19, the imaging process according to Embodiment 3 (step S1 in Figure 4) will be described. Similar to Embodiment 1, an application program for calculating the flood level is executed as a prerequisite.
[0079] (Step S11A: Guide selection process) As shown in Figure 20, the guide display unit 114 displays the guide 51 superimposed on the camera image obtained by the camera 141. In Embodiment 3, the guide 51 indicates the upper limit position 52 and the lower limit position 53 in the height direction. The guide display unit 114 prepares multiple guides 51 with varying distances between the upper limit position 52 and the lower limit position 53. The guide display unit 114 then allows the user to select the guide 51 to use. Specifically, the user selects the guide 51 to use according to the height of the flood line, etc. The guide display unit 114 displays the selected guide 51 superimposed on the camera image. For example, if the flood line is at a high position, a reference object placed on the floor may only appear small in the image.
[0080] (Step S12A: Image acquisition process) The image acquisition unit 111 acquires a camera image showing the water inundation line on the wall surface as image data when the reference object is aligned with the guide 51. Specifically, when the reference object is aligned with the guide 51, the image acquisition unit 111 automatically acquires the camera image as image data. Alternatively, when the reference object is aligned with the guide 51, the image acquisition unit 111 makes the shutter button pressable, and when the shutter button is pressed, it acquires the camera image as image data.
[0081] In Embodiment 3, the guide 51 indicates an upper limit position 52 and a lower limit position 53 in the height direction. Therefore, the image acquisition unit 111 acquires a camera image showing the water inundation line on the wall surface as image data, with a reference object placed between the upper limit position 52 and the lower limit position 53. In this case, it is desirable that the upper end of the reference object is close to the upper limit position 52 and the lower end of the reference object is close to the lower limit position 53. By doing so, image data in the predetermined state is acquired. Therefore, it is necessary to select an appropriate guide 51 in step S11A.
[0082] In this case, the image acquisition unit 111 acquires image data while also considering the shooting angle. Specifically, the image acquisition unit 111 acquires image data when the angle of the camera 141 relative to the wall is a specified angle. The specified angle is a pre-set angle. The specified angles can be set for pitch, roll, and yaw. The specified angle may have a certain range, such as from 43 to 47 degrees. In other words, when the reference object follows the guide 51 and the angle of the camera 141 relative to the wall is a specified angle, the image acquisition unit 111 automatically acquires the camera image as image data. Alternatively, when the above conditions are met, the image acquisition unit 111 makes the shutter button pressable and acquires the camera image as image data when the shutter button is pressed. As shown in Figure 20, the guide display unit 114 may display a spirit level 54 to indicate the angle of the camera 141 and the specified angle. This makes it easier for the user to adjust the angle of the camera 141 to the specified angle.
[0083] Furthermore, as shown in Figure 21, the image acquisition unit 111 may acquire image data using only a certain area on the central side of the shooting range of the camera 141. In other words, the image acquisition unit 111 may acquire only a certain area on the central side of the shooting range of the camera image as image data. This is because distortion becomes greater further out of the camera 141's field of view, making it difficult to accurately calculate the flood level.
[0084] The process from step S13A to step S14A is the same as the process from step S12 to step S13 in Figure 5.
[0085] ***Effects of Embodiment 3*** As described above, in the flood height calculation system 100 according to Embodiment 3, camera images obtained by the camera 141 when the shooting conditions are met are acquired as image data. This provides image data in which reference objects are photographed at a generally defined shooting angle. As a result, the flood height can be calculated with high accuracy.
[0086] Image data acquisition is performed by the user, who is the insurance policyholder. Therefore, there is a risk that image data may not be acquired properly, leading to inaccurate calculations of flood levels.
[0087] In particular, by combining the method described in Embodiment 3 with the method described in Embodiment 2, which involves identifying the position of a reference object on the wall surface and calculating the unit height, the flood level can be calculated with high accuracy.
[0088] ***Other configurations*** <Modification 4> In Embodiment 3, the guide 51 indicated an upper limit position 52 and a lower limit position 53. The guide 51 is not limited to this; it may also indicate an upper and lower limit with respect to the position of the upper end of the reference object, and an upper and lower limit with respect to the lower end of the reference object. The reference object may be considered to be following the guide 51 when the upper end of the reference object is between the upper and lower limits with respect to the position of the upper end, and the lower end of the reference object is between the upper and lower limits with respect to the lower end.
[0089] Furthermore, the guide 51 may indicate a range not only in the vertical direction but also in the horizontal direction.
[0090] Embodiment 4. Embodiment 4 differs from Embodiment 3 in that it detects the reference object by expanding the range indicated by the guide 51. Embodiment 4 will explain this difference, and the same points will not be explained.
[0091] ***Explanation of operation*** Referring to Figure 22, the operation of the flood level calculation system 100 according to Embodiment 4 will be described. The operation procedure of the flood height calculation system 100 according to Embodiment 4 corresponds to the flood height calculation method according to Embodiment 4. Furthermore, the program that implements the operation of the flood height calculation system 100 according to Embodiment 4 corresponds to the flood height calculation program according to Embodiment 4.
[0092] In step S2 of Figure 4, the detection unit 211 of the flood height calculation device 20 generates an enlarged image 55, which is an enlarged version of the image data showing the range indicated by the guide 51, as shown in Figure 22. The detection unit 211 then detects a reference object from the enlarged image 55. Specifically, the detection unit 211 inputs the enlarged image 55 into an object detection model to detect the reference object.
[0093] The detection unit 211 may also generate an enlarged image 55 that expands the range indicated by the guide 51 to a slightly wider range.
[0094] ***Effects of Embodiment 4*** As described above, the flood height calculation system 100 according to Embodiment 4 is shown in Guide 51. The range is expanded to detect the reference object. In object detection using image recognition, detection accuracy tends to decrease when the size of the reference object in the image data is small compared to the size of the entire image data. Furthermore, a smaller reference object size results in a larger area of the image data where the reference object is not visible, increasing the likelihood of false detection of objects other than the reference object. However, in the flood level calculation system 100 according to Embodiment 4, the reference object is detected from an enlarged image 55 in which the area containing the reference object is expanded. Therefore, a decrease in detection accuracy and false detection can be prevented.
[0095] ***Other configurations*** <Modification 5> When using the method described in Embodiment 2, it is necessary to detect the cap portion of the PET bottle. In this case, the detection unit 211 cuts out and enlarges the central portion in the left-right direction at the upper end of the area where the reference object, the PET bottle, was detected in step S2 of Figure 4. The detection unit 211 may then detect the cap portion from the enlarged image data. It is assumed that the cap portion is located in the central portion in the left-right direction at the upper end.
[0096] Embodiment 5. Embodiment 5 differs from Embodiment 2 in that there are multiple types of bottles used as the reference material. Embodiment 5 explains these differences, while omitting explanations of the similarities.
[0097] ***Explanation of the structure*** Referring to Figure 23, the configuration of the flood height calculation device 20 according to Embodiment 5 will be described. The flood height calculation device 20 differs from the flood height calculation device 20 shown in Figure 10 in that it includes a type identification unit 216 as a functional component. The function of the type identification unit 216 is implemented by software or hardware, similar to other functional components.
[0098] ***Explanation of operation*** The operation of the flood level calculation system 100 according to Embodiment 5 will be described with reference to Figures 24 and 25. The operation procedure of the flood height calculation system 100 according to Embodiment 5 corresponds to the flood height calculation method according to Embodiment 5. Furthermore, the program that implements the operation of the flood height calculation system 100 according to Embodiment 5 corresponds to the flood height calculation program according to Embodiment 5.
[0099] In Embodiment 5, as in Embodiment 2, the reference material will be described as a PET bottle. As shown in Figure 24, there are several types of PET bottles, such as those with a capacity of 500 mL (milliliters) and those with a capacity of 2 L (liters). Each type has a different size. Therefore, to calculate the flood level, it is necessary to identify the type of PET bottle used as the reference object. Here, we will explain assuming that either a 500 mL PET bottle or a 2 L PET bottle is used as the reference object.
[0100] Referring to Figure 25, the unit height calculation process according to Embodiment 5 will be explained. (Step S31A: Type identification process) The type identification unit 216 of the flood height calculation device 20 identifies the type of reference object. In Embodiment 5, the type identification unit 216 identifies the type of PET bottle based on the size of the lid portion and the size of the body portion of the PET bottle detected in step S2 of Figure 4.
[0101] Specifically, the type identification unit 216 identifies the type of PET bottle based on the ratio of the width of the lid to the width of the body. For example, the type identification unit 216 determines that a bottle is a 500 mL PET bottle if the ratio of the width of the lid to the width of the body is greater than threshold A. On the other hand, the type identification unit 216 determines that a bottle is a 2 L PET bottle if the ratio is less than or equal to threshold A.
[0102] The type identification unit 216 may interrupt processing as an error if the ratio value is greater than threshold B or less than threshold C. Here, threshold B is greater than threshold A, and threshold C is less than threshold A. This makes it possible to interrupt processing as an error if a 1L PET bottle or the like is used.
[0103] The processing from steps S32A to S34A is the same as the processing from steps S31 to S33 in Figure 12. However, the actual size of the PET bottle is the size corresponding to the type identified in step S31A. Specifically, the actual depth of the lid 37, the actual depth of the base 38, and the actual height are the sizes corresponding to the type identified in step S31A. Also, the inclination and intercept, or top surface ratio, used when calculating the lid image depth 36 are the values corresponding to the type identified in step S31A.
[0104] ***Effects of Embodiment 5*** As described above, the flood level calculation system 100 according to Embodiment 5 identifies the type of reference object. This allows the flood level to be calculated using any of the multiple types of reference objects. In the event of a flood, it may be difficult to obtain the specified reference object. Therefore, being able to calculate the flood level using any of the multiple types of reference objects greatly enhances convenience.
[0105] Embodiment 6. Embodiment 6 differs from Embodiments 1 to 5 in that it corrects the calculated flood level. Embodiment 6 explains this difference, while omitting explanations of the same points. Embodiment 6 describes a case where a modification has been made to Embodiment 1. However, it is also possible to modify Embodiments 2 to 5.
[0106] ***Explanation of the structure*** Referring to Figure 26, the configuration of the flood height calculation device 20 according to Embodiment 6 will be described. The flood height calculation device 20 differs from the flood height calculation device 20 shown in Figure 3 in that it includes a correction unit 217 as a functional component. The function of the correction unit 217 is implemented by software or hardware, similar to other functional components.
[0107] ***Explanation of operation*** Referring to Figures 27 to 30, the operation of the flood height calculation system 100 according to Embodiment 6 will be described. The operation procedure of the flood height calculation system 100 according to Embodiment 6 corresponds to the flood height calculation method according to Embodiment 6. Furthermore, the program that implements the operation of the flood height calculation system 100 according to Embodiment 6 corresponds to the flood height calculation program according to Embodiment 6.
[0108] Referring to Figure 27, the flood height calculation process according to Embodiment 6 (step S4 in Figure 4) will be explained. (Step S41: Height calculation process) The flood level calculation unit 213 calculates the flood level as described in Embodiment 1.
[0109] (Step S42: Correction process) The correction unit 217 of the flood height calculation device 20 adjusts the flood height according to the flood height calculated in step S41. The flood level calculated in step S41 is corrected by the correct amount. As shown in Figure 28, in the image data obtained by camera 141, the central part appears larger. In other words, the closer the area is to camera 141, the larger the image appears. Therefore, even at the same distance, the central part of the image data has a higher number of pixels. To address this, as shown in Figure 29, the approximate deviation for each flood level is statistically recorded in advance as a correction amount. In Figure 29, the correction amount is recorded every 10 centimeters. The correction unit 217 reads out a correction amount corresponding to the flood height calculated in step S41. Then, it corrects the flood height calculated in step S41 by the amount of the correction that was read out.
[0110] In Figure 29, the correction amount was a specific distance. However, as shown in Figure 30, the correction amount may also be a percentage of the flood level. This allows for more appropriate correction.
[0111] ***Effects of Embodiment 6*** As described above, the flood height calculation system 100 according to Embodiment 6 corrects the calculated flood height by a correction amount corresponding to the calculated flood height. This makes it possible to calculate the flood height with high accuracy.
[0112] Embodiment 7. Embodiment 7 differs from Embodiments 1 to 6 in that it calculates the reliability of the calculated flood height. Embodiment 7 explains this difference, while omitting explanations of the same points. Embodiment 7 describes a case where a modification has been made to Embodiment 1. However, it is also possible to modify Embodiments 2 to 6.
[0113] ***Explanation of the structure*** Referring to Figure 31, the configuration of the flood height calculation device 20 according to Embodiment 7 will be described. The flood height calculation device 20 differs from the flood height calculation device 20 shown in Figure 3 in that it includes a reliability calculation unit 218 as a functional component. The function of the correction unit 217 is implemented by software or hardware, similar to the other functional components.
[0114] ***Explanation of operation*** Referring to Figure 32, the operation of the flood height calculation system 100 according to Embodiment 7 will be described. The operation procedure of the flood height calculation system 100 according to Embodiment 7 corresponds to the flood height calculation method according to Embodiment 7. Furthermore, the program that implements the operation of the flood height calculation system 100 according to Embodiment 7 corresponds to the flood height calculation program according to Embodiment 7.
[0115] Referring to Figure 32, the flood height calculation process according to Embodiment 7 (step S4 in Figure 4) will be explained. (Step S41A: Height calculation process) The flood level calculation unit 213 calculates the flood level as described in Embodiment 1.
[0116] (Step S42A: Confidence calculation process) The reliability calculation unit 218 of the flood height calculation device 20 calculates the reliability of the flood height calculated in step S41. Specifically, the reliability calculation unit 218 calculates the reliability based on information regarding the calculation accuracy of the unit height or flood level, such as the shooting angle of the image data and the position of the reference object. In this case, the reliability calculation unit 218 may also calculate the error range of the flood level according to the reliability. The error range becomes larger as the reliability decreases.
[0117] This section explains a specific example of how confidence levels are calculated. As described in Embodiment 3, assume that image data has been acquired by displaying the guide 51. In this case, the reliability calculation unit 218 calculates the reliability according to the degree to which the reference object followed the guide 51. For example, the reliability calculation unit 218 calculates the reliability according to the distance between the upper end of the reference object and the upper limit position 52, and the distance between the lower end of the reference object and the lower limit position 53. The reliability calculation unit 218 increases the reliability as the sum of these distances decreases.
[0118] As explained in Modification Example 3, at least one of the wall bottom position 31 and the wall top position 32 is specified. In this case, the reliability calculation unit 218 calculates the reliability according to the difference between the specified wall bottom position 31 and the wall bottom position 31 calculated by the method described in Embodiment 2. The reliability calculation unit 218 also calculates the reliability according to the difference between the specified wall top position 32 and the wall top position 32 calculated by the method described in Embodiment 2. The reliability calculation unit 218 increases the reliability as the difference decreases. Furthermore, the reliability calculation unit 218 may significantly reduce the reliability if the wall bottom position 31 is not between the lower and upper ends of the reference object. This is because the wall bottom position 31 should be between the lower and upper ends of the reference object.
[0119] As described in Embodiment 5, the type of reference object is determined. In this case, the reliability calculation unit 218 calculates the reliability according to the determination accuracy. The reliability calculation unit 218 increases the reliability as the determination accuracy increases. For example, suppose we use the ratio obtained by dividing the width of the lid by the width of the body to determine whether it is a 500mL or 2L PET bottle. In this case, we set the accuracy of the determination result for each ratio value. This allows us to specify the accuracy of the determination result.
[0120] In step S5 of Figure 4, the data transmission unit 214 transmits the reliability along with the flood height to the insurance company's server. If an error range is calculated, the data transmission unit 214 also transmits the error range to the insurance company's server.
[0121] ***Effects of Embodiment 7*** As described above, the flood height calculation system 100 according to Embodiment 7 calculates the reliability of the flood height. This makes it easier for insurance companies to determine whether the calculated flood height is reliable or not. In flood damage insurance claims, if a user uses the system in an unintended manner, the insurance payout may be calculated to be higher than it actually is. Furthermore, malicious users may tamper with the standard criteria to claim unfairly high insurance payouts. An example of tampering with a standard criterion would be using a plastic bottle with an unusual shape. Therefore, if the reliability of the criterion falls below the standard, insurance company staff could conduct a visual inspection.
[0122] Furthermore, the flood level calculation system 100 according to Embodiment 7 calculates the error range. Insurance payouts for flood damage are sometimes determined based on the level of flooding. Therefore, insurance companies may take the following actions: If the insurance payout does not change even after considering the margin of error, they will use the calculated flooding level. If considering the margin of error may change the insurance payout, they will perform a visual inspection.
[0123] In addition, the term "part" in the above explanation may be replaced with "circuit," "process," "procedure," "processing," or "processing circuit."
[0124] The embodiments and variations of this disclosure have been described above. Some of these embodiments and variations may be implemented in combination. Alternatively, some or all of them may be implemented in part. However, this disclosure is not limited to the embodiments and variations described above, and various modifications are possible as needed. [Explanation of symbols]
[0125] 100 Flood height calculation system, 10 User terminal, 11 Processor, 12 Memory, 13 Storage, 14 Communication interface, 15 Electronic circuit, 111 Image acquisition unit, 112 Flood line setting unit, 113 Image transmission unit, 114 Guide display unit, 141 Camera, 20 Flood height calculation device, 21 Processor, 22 Memory, 23 Storage, 24 Communication interface, 25 Electronic circuit, 211 Detection unit, 212 Unit height calculation unit, 213 1. Flood height calculation unit, 214. Data transmission unit, 215. Wall position identification unit, 216. Type identification unit, 217. Correction unit, 218. Reliability calculation unit, 31. Wall bottom position, 32. Wall top position, 33. Top end position, 34. Bottom image depth, 35. Bottom end, 36. Cover image depth, 37. Cover actual depth, 38. Bottom actual depth, 39. Cover vertical width, 40. Cover horizontal width, 41. Amount of displacement, 42. Top end, 51. Guide, 52. Upper limit position, 53. Lower limit position, 54. Level indicator, 55. Magnified image, 90. Transmission line.
Claims
1. A detection unit that detects a reference object from image data including a water inundation line on a wall surface; a unit height calculation unit that calculates a unit height, which is a height per unit pixel, from the number of pixels in the image data of the reference object detected by the detection unit and an actual height, which is an actual height of the reference object; a flood height calculation unit that calculates a flood height, which is the distance from the floor surface to the flood line, based on the number of pixels from the floor surface to the flood line in the image data and the unit height calculated by the unit height calculation unit; a reliability calculation unit that calculates the reliability of the inundation height calculated by the inundation height calculation unit; A flood height calculation system comprising:
2. The flood height calculation system further comprises: an image acquisition unit that acquires, as the image data, a camera image captured by a camera in a state where a shooting condition, which is a condition of a camera position for shooting, is satisfied; a guide display unit that displays guides indicating upper and lower limit positions in a height direction by superimposing the guides on a camera image obtained by the camera; Equipped with The reliability calculation unit calculates the reliability according to a distance between an upper end of the reference object and the upper limit position and a distance between a lower end of the reference object and the lower limit position. The flood height calculation system according to claim 1.
3. The reliability calculation unit calculates the reliability so that the reliability is higher as the sum of the distance between the upper end of the reference object and the upper limit position and the distance between the lower end of the reference object and the lower limit position is shorter. The flood height calculation system according to claim 2.
4. The reference object is an object having a thickness, The flood height calculation system further comprises: a wall position specifying unit that specifies a position obtained by shifting a lower end of the reference object in the image data upward by a bottom image depth that is a length in a depth direction of the bottom surface of the reference object in the image data as a wall bottom position that is a position on the wall surface corresponding to the bottom of the reference object. Equipped with The reliability calculation unit calculates the reliability according to a difference between a wall bottom position specified by a user and a wall bottom position identified by the wall position identification unit. The flood height calculation system according to claim 1.
5. The reliability calculation unit calculates the reliability so that the smaller the difference between the wall bottom position specified by the user and the wall bottom position identified by the wall position identification unit, the higher the reliability. The flood height calculation system according to claim 4.
6. The reference object is a bottle, The reliability calculation unit calculates the reliability based on a ratio value obtained by dividing the width of the lid portion of the bottle detected by the detection unit by the width of the body portion of the bottle. The flood height calculation system according to claim 1.
7. The flood height calculation system further comprises: a data transmission unit that transmits the reliability calculated by the reliability calculation unit together with the flood height to a server of an insurance company; The flood height calculation system according to any one of claims 1 to 6, comprising:
8. The reliability calculation unit calculates the error range such that the lower the calculated reliability, the larger the error range of the flood height. The flood height calculation system according to claim 1.
9. The flood height calculation system further comprises: a data transmission unit that transmits the reliability and the error range calculated by the reliability calculation unit together with the flood height to a server of an insurance company; The flood height calculation system according to claim 8 .
10. A computer detects a reference object from image data including a water inundation line on a wall surface; a computer calculates a unit height, which is a height per unit pixel, from the number of pixels in the image data of the reference object and an actual height, which is an actual height of the reference object; A computer calculates a flood height, which is the distance from the floor surface to the flood line, from the number of pixels from the floor surface to the flood line in the image data and the unit height, A flood height calculation method in which a computer calculates the reliability of the flood height.
11. A detection process for detecting a reference object from image data including a water inundation line on a wall surface; a unit height calculation process for calculating a unit height, which is a height per unit pixel, from the number of pixels in the image data of the reference object detected by the detection process and an actual height, which is an actual height of the reference object; a flood height calculation process for calculating a flood height, which is the distance from the floor surface to the flood line, based on the number of pixels from the floor surface to the flood line in the image data and the unit height calculated by the unit height calculation process; a reliability calculation process for calculating the reliability of the inundation height calculated by the inundation height calculation process; A flood height calculation program that enables a computer to function as a flood height calculation device.
