LEAK DETECTION SYSTEM, LEAK DETECTION METHOD, AND PROGRAM

The leak detection system uses an adsorption sheet and image analysis to detect fuel oil leaks in ship engines, addressing installation challenges and improving detection accuracy through real-time image analysis and learning mechanisms.

JP7588506B2Active Publication Date: 2024-11-22MITSUBISHI SHIPBUILDING CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2020212424
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-12-22
Publication Date
2024-11-22
Estimated Expiration
2040-12-22

AI Technical Summary

Technical Problem

Existing leak detection systems for fuel oil in ship engines are inadequate for early detection due to unpredictable leak locations and installation challenges, particularly in complex engine rooms with limited space and high costs.

Method used

A leak detection system using an adsorption sheet under pipes, a camera to photograph the sheet, and a detection device that analyzes images for unsteady events, issuing alarms based on color, illuminance, and size deviations, with learning mechanisms to prevent false detections.

Benefits of technology

Enables early detection of fuel oil leaks, reducing fire risks by reliably detecting leaks across complex spaces without redesigning the engine room, and improving detection accuracy through learning from actual and false detection patterns.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007588506000001
    Figure 0007588506000001
  • Figure 0007588506000002
    Figure 0007588506000002
  • Figure 0007588506000003
    Figure 0007588506000003
Patent Text Reader

Abstract

To provide a system that can detect a leakage of oil, for example.SOLUTION: The leakage detection system includes: a sheet laid on the floor of the lower part of a pipe; a camera for imaging the sheet; and a detector. The detector includes: an acquisition unit for acquiring an image taken by the camera; a detection unit for detecting the generation of an unsteady event that happens in the sheet on the basis of the image; and a notification unit for issuing an alarm when the detection unit detects the generation of an unsteady event.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a leak detection system, a leak detection method, and a program. [Background technology]

[0002] If fuel oil leaks from the main engine of a ship, there is a risk of a fire breaking out. Therefore, in ship maintenance, it is important to detect leakage of fuel oil or the like early. Currently, crew members patrol and visually check for small leakage of fuel oil or the like. However, crew members cannot monitor all the time by patrolling, so there is a possibility that leakage cannot be detected early. Patent Document 1 discloses a monitoring system that measures the condition of the ship's engine and lubricating oil with sensors, transmits the measured values ​​wirelessly, and monitors the condition of the ship from a remote location, but does not disclose any technology for early detection of leakage of lubricating oil or the like. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2011-214560 A Summary of the Invention [Problem to be solved by the invention]

[0004] Leaks of fuel oil and other liquids can occur anywhere in exposed piping, but the location of the leak is irregular. For leaks whose location cannot be predicted, a method can be considered in which trays are installed in a certain area and liquid that falls onto the tray is detected by a sensor. However, for example, the engine room where the main engine is installed is a complex web of equipment and piping, making it difficult to install trays in all necessary areas. In addition, engine room space is limited, and changing the design of the engine room to install trays is often unreasonable in terms of both cost and functionality.

[0005] The present disclosure provides a leak detection system, a leak detection method, and a program that can solve the above problems. [Means for solving the problem]

[0006] The leak detection system of the present disclosure includes a sheet, a camera that photographs the sheet, and a detection device. The detection device includes an image acquisition unit that acquires an image photographed by the camera, a detection unit that detects an unsteady event occurring on the sheet based on the image, and a notification unit that issues an alarm when the detection unit detects the occurrence of the unsteady event. The detection unit determines that the unsteady event may have occurred based on at least one of the color deviation, illuminance deviation, and size of the portion where any of the deviations has occurred between the image of the sheet in a steady state and the image acquired by the image acquisition unit, and determines whether the image in which it has been determined that the unsteady event may have occurred is similar to the image determined to be an erroneous detection based on erroneous detection prevention data including the color, illuminance, and size of oil droplet marks based on the image determined to be an erroneous detection, and if they are similar, determines that the unsteady event has not occurred, and if they are not similar, determines that the unsteady event may have occurred for a predetermined number of consecutive images acquired at a predetermined cycle, and if it is determined that they are not similar to the image determined to be an erroneous detection, determines that the unsteady event has been detected.

[0007] In the leak detection method disclosed herein, a sheet is photographed by a camera, the image photographed by the camera is acquired, a non-regular event occurring on the sheet is detected based on the image, and an alarm is issued when the occurrence of the non-regular event is detected. In the detection, it is determined that the unsteady event may have occurred based on at least one of the color deviation, illuminance deviation, and size of the part where any of the deviations occurs between the image of the sheet in a steady state and the acquired image, and it is determined whether the image in which it has been determined that the unsteady event may have occurred is similar to the image determined to be an erroneous detection based on erroneous detection prevention data including the color, illuminance, and size of oil droplet marks based on the image determined to be an erroneous detection, and if they are similar, it is determined that the unsteady event has not occurred, and if they are not similar, it is determined that the unsteady event may have occurred for a predetermined number of or more consecutive images acquired at a predetermined cycle, and if it is determined that they are not similar to the image determined to be an erroneous detection, it is determined that the unsteady event has been detected.

[0008] The program disclosed herein includes a process of acquiring an image of a sheet, detecting an unsteady event occurring on the sheet based on the image, and issuing an alarm when the occurrence of the unsteady event is detected. In the detection, the process includes determining that the unsteady event may have occurred based on at least one of a color deviation, an illuminance deviation, and a size of a portion where any of the deviations occurs between an image of the sheet in a steady state and the acquired image, determining whether the image in which the unsteady event may have occurred is similar to the image in which the unsteady event has been determined to be an erroneous detection based on erroneous detection prevention data including a color, an illuminance, and a size of an oil droplet trace based on the image in which the unsteady event has been determined to be an erroneous detection, and determining that the unsteady event has not occurred if the images are similar, and determining that the unsteady event may have occurred for a predetermined number of consecutive images acquired at a predetermined cycle if the images are not similar to the image in which the unsteady event has been determined to be an erroneous detection, and determining that the unsteady event has been detected if the images are not similar to the image in which the unsteady event has been determined to be an erroneous detection. This is a program that executes the above. Effect of the Invention

[0009] According to the above-mentioned leak detection system, leak detection method, and program, a leak of a detection target can be detected early. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of a leak detection system according to an embodiment. [Diagram 2] FIG. 2 is a top view of the suction sheet according to the embodiment. [Diagram 3] FIG. 2 is a diagram illustrating an example of a monitored area according to the embodiment. [Figure 4] 10 is a flowchart illustrating an example of a leak detection process according to the embodiment. [Diagram 5]FIG. 2 is a diagram illustrating an example of a hardware configuration of a detection device according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] <Embodiment> Hereinafter, a leak detection system according to an embodiment of the present disclosure will be described with reference to FIGS. (composition) FIG. 1 is a diagram illustrating an example of a leak detection system according to an embodiment. The leakage detection system 100 is a system for detecting oil leakage from tanks, main engines, auxiliary machinery, and piping (hereinafter collectively referred to as piping P1) in which fuel oil or the like to be detected is sealed in a ship. The leakage detection system 100 includes an adsorbent sheet SH, a camera 20 for photographing the surface of the adsorbent sheet SH, and a detection device 10 for analyzing the image photographed by the camera 20 to detect non-steady events on the surface of the adsorbent sheet SH. FIG. 1 shows a cross-sectional view of the piping P1. The adsorbent sheet SH is laid on the floor surface FL under the piping P1. For example, the adsorbent sheet SH is a white sheet made of cotton or the like, and selectively adsorbs oil. When oil O1 leaks from the piping P1 and drips onto the adsorbent sheet SH, the adsorbent sheet SH adsorbs the oil O1, and a black or brown stain-like adsorption mark is formed at the position where the oil O1 was dripped. FIG. 2 shows a top view of the adsorbent sheet SH. The oil droplet traces SP in FIG. 2 are traces of the oil that has leaked. The area under the pipe P1 may be narrow, but the adsorbent sheet SH can be laid in the narrow area. In addition, by laying the adsorbent sheet SH under the pipe P1, the oil droplet traces SP can be reliably left behind when oil leaks. The camera 20 is installed so as to be able to capture the entire surface of the adsorbent sheet SH, and continuously captures the surface of the adsorbent sheet SH at all times or at a predetermined time interval. When oil O1 leaks from the pipe P1, the image captured by the camera 20 captures the oil droplet traces SP. By monitoring the surface of the adsorbent sheet SH and detecting the occurrence of the oil droplet traces SP, the leakage of the oil O1 can be detected. The detection device 10 compares an image obtained by capturing the adsorbent sheet SH in a steady state in which no oil is dripping with an image in which the oil droplet traces SP are captured, and analyzes the changes. When the change in the image meets a predetermined criterion, the detection device 10 determines that an unsteady event has occurred in the adsorbent sheet SH, and issues an alarm.

[0012] 1, only one camera 20 is connected to the detection device 10, but multiple cameras 20 may be connected to the detection device 10. In this case, the multiple cameras 20 may be provided so as to capture images of the adsorption sheets SH laid at different positions.

[0013] FIG. 3 shows the state of the monitored area (engine room) inside the ship. In FIG. 3, Q1 is the main engine, and C1 is a cover connected to the main engine. The cover C1 is installed on the upper and side of the fuel pipe P1 connected to the main engine Q1 in the X direction, and the bottom surface is open. An internal C1 fuel pipe (not shown) is provided under the cover C1. In this example, an adsorption sheet SH is laid on the bottom of the internal C1 fuel pipe, and the camera 20 is placed in a position where the adsorption sheet SH can be photographed. This allows the surface of the adsorption sheet SH to be constantly monitored. As shown in the figure, the bottom of the internal C1 fuel pipe is a narrow space, but if the adsorption sheet SH is used, it can be installed even in a narrow area, and there is no need to change the design of the engine room. The internal C1 fuel pipe is an example of the piping P1.

[0014] The detection device 10 includes an image acquisition unit 11 , an input reception unit 12 , a learning unit 13 , a determination unit 14 , a notification unit 15 , and a memory unit 16 . The image acquisition unit 11 acquires images captured by the camera 20 and stores them in the storage unit 16. The captured images may be moving images or still images. In the case of moving images, the image acquisition unit 11 stores the images for each frame. The input receiving unit 12 receives various inputs such as instructions to start and end leakage monitoring. For example, when the detection device 10 detects an unsteady event and outputs an alarm, the crew checks the image in response to the alarm, confirms whether or not an oil leak has occurred, and inputs the confirmation result to the detection device 10. The input receiving unit 12 acquires the confirmation result input by the crew.

[0015] The learning unit 13 learns an image of the suction sheet SH in a steady state (a state where no oil is dripping). For example, the learning unit 13 may calculate the range of the brightness value of each pixel of the suction sheet SH in a state where no oil droplets are dripping from a large number of images taken in a state where no oil droplets are dripping on the surface of the suction sheet SH, and set this range as the threshold value, or may calculate an average value for each pixel at each position of the image in the steady state and set this value as the threshold value. If the brightness value of each pixel of the image taken of the surface of the suction sheet SH differs depending on the time of day or weather, the learning unit 13 may set the threshold value of the steady state for each time of day or weather. In addition, the learning unit 13 learns the characteristics of the image that causes the false detection based on the image that was the source of the false detection when the confirmation result of the false detection was input by the crew, and creates data for preventing false detection (b). The data for preventing false detection in (b) includes, for example, the color, illuminance, size, number, distribution pattern, etc. of the oil droplet traces SP based on the image determined to be a false detection. Furthermore, when a large number of cases in which oil leakage has been accurately detected have been accumulated through operation of the leakage detection system 100, the learning unit 13 may learn the pattern of oil droplet traces SP that are created when an oil leakage occurs based on the images of accurate detection (c). The pattern of the oil droplet traces SP is, for example, the color, size, number, distribution pattern, etc. of the oil droplet traces SP.

[0016] The determination unit 14 detects an unsteady event that has occurred on the suction sheet SH. For example, the determination unit 14 analyzes the image acquired by the image acquisition unit 11, compares it with the threshold value of the image indicating the steady state set by the learning unit 13, and checks (A) the color deviation, (B) the deviation of the lighting reflectance (deviation of illuminance), (C) the size (area) of the part where the deviation of the color or illuminance is greater than or equal to a predetermined value, and the like. If there is a deviation of greater than or equal to a predetermined threshold value for at least one of (A) to (C), the determination unit 14 determines whether or not the deviation indicates an unsteady event that has occurred on the surface of the suction sheet SH. For example, if the deviation (A) is confirmed for the image acquired by the image acquisition unit 11, the determination unit 14 determines that the image acquired by the image acquisition unit 11 indicates that an unsteady event may have occurred on the suction sheet SH, unless the deviation matches the characteristics indicated by the erroneous detection prevention data. When the state in which it is determined that an unsteady event may have occurred continues for a predetermined time or more (when the state in which it is determined that an unsteady event may have occurred continues for successively acquired images), the determination unit 14 determines that an unsteady event has occurred in the suction sheet SH.

[0017] The notification unit 15 outputs an alarm when the determination unit 14 determines that an unsteady event has occurred in the suction sheet SH. For example, the notification unit 15 may display a message notifying that oil may have leaked on a display device connected to the detection device 10, or send a message or e-mail containing alarm information to a mobile terminal or computer used by the crew. The notification unit 15 may also output a signal indicating that an unsteady event has occurred to an alarm connected to the detection device 10. The memory unit 16 stores the images acquired by the image acquisition unit 11, the threshold values ​​and data for preventing erroneous detection set by the learning unit 13, the patterns of oil droplet traces SP, and the like.

[0018] (operation) Next, the operation of the leak detection system 100 will be described. FIG. 4 is a flowchart illustrating an example of the leakage detection process according to the embodiment. While the leak detection system 100 is performing the monitoring process, the leak detection system 100 repeatedly performs the following process at a predetermined control period. As a premise, the memory unit 16 stores at least (a) a threshold value indicating a steady state set by the learning unit 13. For example, the memory unit 16 stores a steady-state luminance value for each pixel on the surface of the suction sheet SH as viewed from the shooting position of the camera 20. The camera 20 photographs the suction sheet SH (step S1). The camera 20 outputs the photographed image to the detection device 10. In the detection device 10, the image acquisition unit 11 acquires the image (step S2). The image acquisition unit 11 stores the acquired image in the storage unit 16 and outputs it to the judgment unit 14. Next, the judgment unit 14 analyzes the image (step S3). For example, the judgment unit 14 reads out information such as a threshold value indicating a steady state from the storage unit 16. The threshold value includes a luminance value for each pixel when the suction sheet SH is photographed in a state in which there are no oil droplets. If the luminance value changes depending on the time of day, the judgment unit 14 reads out image information for the current time period. Next, the judgment unit 14 compares the luminance value of each pixel of the image indicating the steady state with the luminance value of each pixel of the image currently acquired by the image acquisition unit 11, and judges whether or not there is a deviation in color or illuminance for each pixel. For example, if the luminance values ​​of corresponding pixels (pixels at the same position) deviate by a predetermined amount or more, the determination unit 14 determines that there is a deviation in color or illuminance between the latest image and the image in the steady state. The determination unit 14 also calculates the size of the portion in which the deviation has occurred based on the number of pixels determined to have a deviation in color or illuminance. For example, if the deviation in color or illuminance is equal to or greater than a predetermined size, the determination unit 14 determines that the image has changed from the steady state. If (b) is stored in the storage unit 16, the determination unit 14 may compare the portion determined to have changed from the steady state with (b) to determine whether the detected change is close to the deviation in color or illuminance of an image confirmed to have been erroneously detected in the past, the size of the portion determined to have a deviation, and determine that the change detected this time is not due to oil leakage if the difference between the change detected this time and the change indicated by (b) is within a predetermined range. The determination unit 14 determines that there is a possibility of oil leakage if the change detected this time and the change shown in (b) deviate by a predetermined amount or more. If it has determined that there is a possibility of oil leakage, and if it has determined that there is a possibility of oil leakage continuously for a predetermined period of time or more (for a predetermined number of images or more), the determination unit 14 determines that an unsteady event has been detected in the images of the adsorbent sheet SH (Step S4; Yes), and if the determination that there is a possibility of oil leakage has not continued for the predetermined period of time, the determination is withheld (Step S4; No).This makes it possible to prevent a change in the image that occurs when a person or object temporarily crosses the line of sight of the camera 20 from being erroneously detected as the occurrence of an unsteady event. When the determination unit 14 determines that an unsteady event has been detected in the image, it outputs the determination result to the notification unit 15. Based on the determination result of the determination unit 14, the notification unit 15 outputs information on the location where the unsteady event was detected together with an alarm that an unsteady event has been detected on the suction sheet SH (step S5).

[0019] When an alarm is output, the crew checks the image in which the non-routine event was detected. The crew inputs the check result to the detection device 10. For example, if an oil leak has actually occurred, the crew inputs to the detection device 10 that it has been accurately detected, and if no oil leak has actually occurred, the crew inputs to the detection device 10 that it was a false detection. The input receiving unit 12 acquires the check result input by the crew (step S6). If no leakage occurs, the crew will stop the alarm. If leakage occurs, the crew will check the site and take measures such as repairs.

[0020] Next, the learning unit 13 performs learning based on the confirmation result (step S7). For example, if the input in step S6 is a false positive, the learning unit 13 analyzes the color, size, and distribution pattern of the part that was falsely detected as an oil droplet trace SP based on the image group that caused the false positive, and registers the analysis result in the storage unit 16 as data for preventing false positive. Also, for example, if the input in step S6 is accurate detection, the learning unit 13 analyzes the color, oil droplet size, number of oil droplets, distribution pattern of oil droplets, etc. based on the original image group in which the accurate detection of oil leakage was performed, and registers the analysis result in the storage unit 16 as data for determination by pattern recognition. When a large amount of data for determination is accumulated, the determination unit 14 may perform detection of the non-stationary event in steps S3 to S4 based on the data for determination. For example, if the image of the part suspected of being an oil droplet included in the image acquired by the image acquisition unit 11 is close to the pattern indicated by the data for determination, the determination unit 14 determines that they are oil droplets. Also, for example, clustering may be performed using the data for determination to create a cluster indicating the characteristics of the image when oil leakage occurs, and the presence or absence of oil leakage may be determined by comparing the characteristics of the image to be determined with the cluster.

[0021] (effect) As described above, according to this embodiment, by laying the suction sheet SH under the pipe P1, traces of oil leakage from the pipe P1 can be left in a detectable state. In addition, the camera 20 constantly photographs the suction sheet SH and analyzes changes in the image in real time, making it possible to immediately detect any unsteady events occurring in the image of the suction sheet SH. Furthermore, when an unsteady event is detected, an alarm is issued, making it possible to detect the oil leakage at an early stage. Early detection of the oil leakage can reduce the risk of fire.

[0022] There are various patterns of oil droplets when oil leakage occurs. Furthermore, if oil leakage is not detected early, the risk of fire and the like increases. Therefore, in this embodiment, the emphasis is placed on quickly outputting an alarm if there is a possibility of oil leakage, rather than precisely determining whether the deviation from the steady state detected in the image is due to oil leakage. This makes it possible to detect oil leakage early without missing anything. On the other hand, when an alarm is output, the image features can be learned from the image group that caused accurate detection or erroneous detection based on whether oil leakage actually occurred, and can be used for subsequent judgments, thereby improving the detection accuracy. In addition, when it is determined that oil leakage has not occurred based on the erroneous detection prevention data, the notification unit 15 may output to a display device or the like that an alarm was not issued to prevent erroneous detection, just in case.

[0023] Also, although it is impossible to predict where in the pipe P1 an oil leak will occur, according to this embodiment, by laying the suction sheet SH over the entire area of ​​the pipe P1, it is possible to reliably detect the oil leak no matter where the leak occurs. Also, the suction sheet SH can be laid relatively freely, regardless of the spatial constraints around the pipe P1, and the area subject to detection of oil leaks can be expanded.

[0024] FIG. 5 is a diagram illustrating an example of a hardware configuration of the detection device according to the embodiment. The computer 900 includes a CPU 901 , a main memory device 902 , an auxiliary memory device 903 , an input / output interface 904 , and a communication interface 905 . The above-described detection device 10 is implemented in a computer 900. Each of the above-described functions is stored in the auxiliary storage device 903 in the form of a program. The CPU 901 reads the program from the auxiliary storage device 903, loads it in the main storage device 902, and executes the above-described processing in accordance with the program. The CPU 901 also reserves a storage area in the main storage device 902 in accordance with the program. The CPU 901 also reserves a storage area in the auxiliary storage device 903 for storing data being processed in accordance with the program.

[0025] In addition, a program for realizing all or part of the functions of the detection device 10 may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read into a computer system and executed to perform processing by each functional unit. The term "computer system" as used herein includes hardware such as an OS and peripheral devices. In addition, if a WWW system is used, the term "computer system" also includes a homepage providing environment (or display environment). In addition, the term "computer-readable recording medium" refers to a portable medium such as a CD, DVD, or USB, and a storage device such as a hard disk built into a computer system. In addition, if the program is distributed to the computer 900 through a communication line, the computer 900 that receives the program may deploy the program in the main storage device 902 and execute the above processing. In addition, the above program may be for realizing part of the above-mentioned functions, and may further be capable of realizing the above-mentioned functions in combination with a program already recorded in the computer system.

[0026] As described above, several embodiments according to the present disclosure have been described, but all of these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. These embodiments and their modifications are included in the scope of the invention and its equivalents as described in the claims, as well as in the scope and gist of the invention. For example, in the embodiment, an adsorbent sheet is used that selectively adsorbs oil derived from cotton, but sheets made from other materials that can leave detectable traces of oil droplets may also be used.

[0027] <Additional Notes> The leak detection system, the leak detection method, and the program described in each embodiment can be understood, for example, as follows.

[0028] (1) A leak detection system 100 according to a first embodiment comprises a sheet (adhesive sheet SH), a camera 20 that photographs the sheet, and a detection device 10. The detection device 10 comprises an image acquisition unit 11 that acquires an image captured by the camera 20, a detection unit (determination unit 14) that detects an unsteady event occurring on the sheet (adhesive sheet SH) based on the image, and a notification unit 15 that issues an alarm when the detection unit detects the occurrence of an unsteady event. By using the adsorbent sheet SH, oil droplets can be left in a detectable state, and the detection target area can be set to cover the entire area where oil leakage may occur. By photographing the sheet SH with the camera 20 and monitoring the image, oil leakage can be detected early.

[0029] (2) A leak detection system 100 according to a second aspect is the leak detection system 100 of (1), wherein the detection unit (determination unit 14) determines that the non-steady event has occurred based on at least one of the color deviation, the illuminance deviation, and the size of the portion where any of the deviations has occurred between the steady-state image of the sheet and the image acquired by the image acquisition unit. This allows oil leakage to be detected early.

[0030] (3) The leak detection system 100 according to the third aspect is the leak detection system 100 of (1), wherein the detection unit (determination unit 14) determines that the non-steady event has occurred based on a learned model (a cluster indicating the characteristics of judgment data or an image when an oil leak occurs) that stores the pattern of the non-steady event occurring on the seat. This allows oil leakage to be detected accurately.

[0031] (4) The leak detection system 100 according to a fourth aspect is the leak detection system 100 of (1) to (3), further comprising an input receiving unit 12 that receives an input of a confirmation result for the image when the detection unit (determination unit 14) detects the non-regular event, and the detection unit (determination unit 14) determines whether the image acquired by the image acquisition unit 11 is similar to the image when the confirmation result is a false detection, and if similar, determines that the non-regular event has not occurred. This improves the detection accuracy. For example, if a situation frequently occurs in which an object with a similar color and size to oil droplets falls on the suction sheet SH placed under the pipe P1, it is possible to prevent an alarm from being output in such a situation.

[0032] (5) The leak detection system 100 according to a fifth aspect is the leak detection system 100 according to any one of (1) to (4), in which the sheet is a sheet that adsorbs the substance to be detected. This allows the oil droplets to remain detectable.

[0033] (6) The leak detection system 100 according to a sixth aspect is the leak detection system 100 of (1) to (5), in which the sheet is provided on the floor surface of an area where the substance to be detected may leak. This makes it possible to detect oil leakage without fail, regardless of where the leakage originates.

[0034] (7) A leak detection method according to a seventh aspect includes photographing a sheet with a camera, acquiring the image captured by the camera, detecting the occurrence of a non-regular event occurring on the sheet based on the image, and issuing an alarm when the occurrence of the non-regular event is detected.

[0035] (8) A program according to an eighth aspect causes a computer 900 to execute a process of acquiring an image of a sheet, detecting the occurrence of a non-regular event on the sheet based on the image, and notifying an alarm when the occurrence of the non-regular event is detected. [Explanation of symbols]

[0036] 100 Leak Detection System 10. Detection device 11. Image acquisition section 12 Input reception section 13. Learning Department 14... Judgment section 15...Notification section 16...Storage section 20. Camera SH···Sheet SP...Oil, etc. 900...Computer 901··CPU 902...Main memory 903...Auxiliary storage device 904 Input / Output Interface 905 Communication Interface

Claims

1. A seat, A camera for photographing the sheet; A detection device; Equipped with The detection device includes: an image acquisition unit that acquires an image captured by the camera; A detection unit that detects an unsteady event occurring on the sheet based on the image; a notification unit that issues an alarm when the detection unit detects the occurrence of the non-regular event; Equipped with The detection unit determines that the unsteady event may have occurred based on at least one of the color deviation, illuminance deviation, and size of the portion where any of the deviations has occurred between the image of the sheet in a steady state and the image acquired by the image acquisition unit, and determines whether the image in which it has been determined that the unsteady event may have occurred is similar to the image determined to be an erroneous detection based on erroneous detection prevention data including the color, illuminance, and size of the oil droplet traces based on the image determined to be an erroneous detection, and if they are similar, determines that the unsteady event has not occurred, and if they are not similar, determines that the unsteady event may have occurred for a predetermined number of consecutive images acquired at a predetermined cycle, and if it is determined that the image is not similar to the image determined to be an erroneous detection, determines that the unsteady event has been detected. Leak detection systems.

2. The sheet is installed directly on a floor surface of a space having a narrow portion without installing a tray on the floor surface. The leak detection system of claim 1 .

3. The sheet is installed directly on the floor of the engine room of the ship, The raw material of the sheet is cotton. The leak detection system according to claim 1 or 2.

4. an input receiving unit that receives an input of a confirmation result for the image acquired by the image acquiring unit when the detection unit detects the unsteady event; Further comprising: The leak detection system according to any one of claims 1 to 3.

5. The sheet is a sheet that adsorbs a substance to be detected. The leak detection system according to any one of claims 1 to 4.

6. The sheet is placed on a floor surface in an area where the substance to be detected may leak. The leak detection system according to any one of claims 1 to 5.

7. The sheet is photographed with a camera, Acquire an image captured by the camera; Detecting an unsteady event occurring on the sheet based on the image; notifying an alarm when the occurrence of the non-stationary event is detected; In the detection, it is determined that the unsteady event may have occurred based on at least one of the color deviation, illuminance deviation, and size of the portion where any of the deviations occurs between the image of the sheet in a steady state and the acquired image, and it is determined whether the image in which it has been determined that the unsteady event may have occurred is similar to the image determined to be an erroneous detection based on erroneous detection prevention data including the color, illuminance, and size of the oil droplet traces based on the image determined to be an erroneous detection, and if they are similar, it is determined that the unsteady event has not occurred, and if they are not similar, it is determined that the unsteady event may have occurred for a predetermined number of consecutive images acquired at a predetermined cycle, and if it is determined that the image is not similar to the image determined to be an erroneous detection, it is determined that the unsteady event has been detected. Leak detection methods.

8. On the computer, Acquire an image of the sheet, Detecting an unsteady event occurring on the sheet based on the image; A process of notifying an alarm when the occurrence of the non-steady event is detected, In the detection, a process of determining that the unsteady event may have occurred based on at least one of a color deviation, an illuminance deviation, and a size of a portion where any of the deviations has occurred between an image of the sheet in a steady state and the acquired image, determining whether the image in which it has been determined that the unsteady event may have occurred is similar to the image determined to be an erroneous detection based on erroneous detection prevention data including a color, an illuminance, and a size of an oil droplet trace based on the image determined to be an erroneous detection, and if they are similar, determining that the unsteady event has not occurred, and if they are not similar, determining that the unsteady event may have occurred for a predetermined number of consecutive images acquired at a predetermined cycle, and determining that the unsteady event has been detected if they are not similar to the image determined to be an erroneous detection; A program that executes the following.

Citation Information

Patent Citations

  • Transformer oil oil leak monitoring devices

    CN207248448U

  • Method and device for detecting abnormality

    JP1994119454A

  • Method and equipment for detecting leakage by image

    JP1996122197A

  • Method for detecting leakage of liquid

    JP1996201212A

  • Method of monitoring oil leakage, and oil leakage monitoring device

    JP2011075425A