Gas camera system, thumb nail generation method, and program

Abstract

To provide a gas camera system capable of generating a thumbnail suitable for a gas inspection purpose.SOLUTION: The gas camera system includes the inspection information setting unit that sets the inspection purpose of the gas, the thumbnail generation unit that generates the thumbnail suitable for the inspection purpose from the time-series images obtained by photographing the gas cloud 70, and the gas flow rate estimation unit that estimates the flow rate of the gas in each frame, and in a case where the inspection purpose is the measurement of the flow rate of the gas, the thumbnail generation unit generates the thumbnail 65 by selecting the frame from the time-series images based on the value obtained by statistically processing the flow rate.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a gas camera system, and a method and program for generating thumbnails of gas clouds in a gas camera system. [Background technology]

[0002] A gas camera system has been developed that can inspect and monitor gas leaks using images captured by an infrared camera, taking advantage of the fact that gas absorbs infrared light of specific wavelengths. When using a gas camera system, users may need to select and check desired test data from a large number of test data sets. Since the test data is video data, displaying thumbnails, which are still images that clearly show the data, allows users to grasp the status at a glance and makes it easier to select the desired test data. In this case, generating thumbnails that clearly show the status is an important issue that significantly affects user convenience. For example, Patent Document 1 proposes generating thumbnails by focusing on gas regions that appear in the inspection data. In Patent Document 1, when the inspection data includes a gas region, an image of a frame in which the area of ​​the gas region is the largest or the average brightness value of the gas region is the largest is selected as the thumbnail. Note that the gas region is the region in which the pixels that make up the visualized image of a gas cloud are distributed. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2019 / 058864 Summary of the Invention [Problem to be solved by the invention]

[0004] These thumbnails are easy to understand when inspecting for gas leaks, but they are not necessarily the best thumbnails for viewing gas flow rates, because gas flow rates are not necessarily correlated with either the area or brightness of the gas region alone, and selecting frames based on area or brightness values ​​does not adequately reflect the gas flow rate. Furthermore, since the information that users want to check in thumbnails varies depending on the purpose of the inspection, if thumbnails are generated using a fixed, predetermined method, as in Patent Document 1, there is a possibility that the thumbnails generated will make it difficult for users to grasp the information they want to check, resulting in a loss of convenience for users. The present invention has been made to solve the above problems, and has an object to provide a gas camera system, a thumbnail generation method, and a program that can generate thumbnails suitable for gas inspection purposes. [Means for solving the problem]

[0005] In order to solve the above problems, the gas camera system, the gas cloud thumbnail generation method, and the program according to the present invention have the following configurations (1) to (19). (1) A gas camera system comprising an inspection information setting unit that sets the inspection purpose of a gas, and a thumbnail generating unit that generates thumbnails suitable for the inspection purpose from time-series images of a gas cloud. (2) The gas camera system described in (1) further comprises a gas flow rate estimation unit that estimates the gas flow rate in each frame from time-series images of a gas cloud, and when the inspection purpose is to measure the gas flow rate, the thumbnail generation unit selects frames from the time-series images based on values ​​obtained by statistically processing the flow rate and generates thumbnails. (3) The gas camera system according to (2), wherein the thumbnail generating unit selects a frame from the time-series images when the flow rate is at a maximum value and generates a thumbnail.

[0006] (4) The gas camera system according to (2), wherein the thumbnail generating unit selects one of the frames from the time-series images when the flow rate is at an average value, and generates a thumbnail. (5) The gas camera system according to (2), wherein the thumbnail generating unit selects a frame when the flow rate is a median value from the time-series images and generates a thumbnail. (6) The gas camera system according to (3), wherein the thumbnail generating unit selects a frame from the time series images at the timing when the gas with the maximum flow rate is ejected and generates a thumbnail.

[0007] (7) The gas camera system described in (2), wherein the thumbnail generation unit calculates an interval average as the average value for each predetermined interval from the time series images, and generates a thumbnail by selecting the maximum frame in the interval where the interval average is maximum. (8) The gas camera system described in (2), wherein the thumbnail generation unit calculates an interval average as the average value for each predetermined interval from the time series images, and selects the first frame of the interval in which the interval average is maximum to generate a thumbnail. (9) The gas camera system described in (2), wherein the thumbnail generation unit calculates an interval average as the average value for each predetermined interval from the time series images, and generates a thumbnail by selecting the middle frame of the interval in which the interval average is maximum.

[0008] (10) The gas camera system described in (2), wherein the thumbnail generation unit calculates an interval average as the average value for each predetermined interval from the time series images, and generates a thumbnail by selecting the median frame of the interval in which the interval average is maximum. (11) The gas camera system described in (2), wherein the thumbnail generation unit performs a type of statistical processing specified externally from the time-series images, selects frames based on the results, and generates thumbnails. (12) In the gas camera system described in (1), when the inspection purpose is a gas leak inspection, the thumbnail generation unit selects a frame from the time series images at a timing when the source of gas leakage is easily identified and generates a thumbnail.

[0009] (13) The gas camera system described in (12), wherein the thumbnail generation unit selects a frame from the time series images in which a shape of a gas cloud diffusing from a leak source is captured, and generates a thumbnail. (14) The gas camera system described in (1), wherein if the thumbnail generation unit is unable to select a frame suitable for the inspection purpose specified by the user, it selects a frame suitable for another inspection purpose and generates a thumbnail. (15) The gas cloud is photographed as time-series images in a plurality of formats, and the thumbnail generation unit selects a time-series image format according to the inspection purpose and generates a thumbnail suitable for the inspection purpose from the time-series image. This gas camera system is described in (1).

[0010] (16) The gas camera system according to (1), wherein the thumbnail generating unit adds predetermined information to the generated thumbnail. (17) A gas camera system in which the thumbnail generation unit further has a thumbnail storage unit that stores thumbnails generated by the gas camera system described in any one of (1) to (16), and the thumbnail storage unit selects and displays thumbnails that correspond to the user's search criteria from the stored thumbnails. (18) A method for generating thumbnails of gas clouds, comprising the steps of: an inspection information setting unit setting an inspection purpose for the gas; and a thumbnail generating unit generating thumbnails suitable for the inspection purpose from time-series images of the gas cloud. (19) A program for causing a computer to execute a procedure for setting a gas inspection purpose and a procedure for generating thumbnails suitable for the inspection purpose from time-series images of a gas cloud. [Effects of the Invention]

[0011] According to the present invention, thumbnails suitable for gas inspection purposes can be generated. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a schematic diagram illustrating a gas camera system according to an embodiment; [Figure 2] 1 is a block diagram illustrating a hardware configuration of a gas camera system according to an embodiment. [Figure 3] FIG. 1 is a block diagram illustrating a functional configuration of a gas camera system according to an embodiment. [Figure 4] FIG. 10 is a diagram illustrating a typical example of a screen displayed when shooting with a gas camera. [Figure 5A] FIG. 10 is a diagram illustrating a typical example of a screen on which an extracted gas region is displayed. [Figure 5B] FIG. 2 is a diagram illustrating a pixel. [Figure 6] 10A and 10B are diagrams illustrating thumbnails of a gas leak inspection in the gas camera system according to the embodiment. [Figure 7] 10A and 10B are diagrams illustrating thumbnails of gas flow rate measurements in a gas camera system according to an embodiment. [Figure 8A] 1 is an example of an infrared image. [Figure 8B] 1 is an example of a visible light image. [Figure 8C] 1 is an example of a thermography image. [Figure 9A] FIG. 10 is a schematic diagram illustrating the timing of calculating the gas flow rate. [Figure 9B] FIG. 10 is a schematic diagram illustrating the timing of gas ejection. [Figure 10] 10 is a graph illustrating an example of a section in which a gas flow rate is calculated using an interval average. [Figure 11] FIG. 1 is a diagram showing the shape of a gas cloud. [Figure 12] 10 is a flowchart of a process for gas inspection. [Figure 13] 10 is a flowchart illustrating a process in the gas camera system according to the embodiment. [Figure 14] 10 is a flowchart illustrating a process in the gas camera system according to the embodiment. [Figure 15] 10 is a flowchart illustrating a process in the gas camera system according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] [Gas Camera System] A gas camera system 1 according to an embodiment will be described with reference to FIGS. 1 to 11. The gas camera system 1 is a system that can inspect and monitor gas leaks using images captured by an infrared camera. Inspection and monitoring using images is performed by utilizing the fact that gases absorb infrared rays of specific wavelengths. Gases that can be inspected are hydrocarbon gases such as methane, ethane, and propane. As shown in FIG. 1, gas camera system 1 includes inspection information setting section 40 and thumbnail generating section 30, and here also includes gas flow rate estimating section 10. Gas camera system 1 also includes infrared camera 22, visible light camera 28, and image processing section 26. The layout of each of these components is not particularly limited, as long as they are included in gas camera system 1 as a whole. Here, some of these components are arranged in overlapping relation to measuring device 20 and information processing device 80. Measuring device 20 and information processing device 80 are connected via a network. Note that monitor screen 58 of measuring device 20 is folded. Below, the hardware configuration and functional configuration of gas camera system 1 will be described in that order.

[0014] 2 is a block diagram illustrating an example of the hardware configuration of the measurement device 20. The measurement device 20 has an infrared camera 22, a visible light camera 28, a control unit 50, an auxiliary storage device 53, a communication interface 54, an input device 55, and an output device 56, which are interconnected by a bus 57. The information processing device 80 has a hardware configuration similar to that of the measurement device 20, excluding the infrared camera 22 and the visible light camera 28. Here, the image processing unit 26, the inspection information setting unit 40, the thumbnail generation unit 30, and the gas flow rate estimation unit 10 are incorporated as functions of the control unit 50, but may also be independent hardware. The control unit 50 is a device that controls the measuring device 20 and the information processing device 80. As illustrated in FIG. 2, the control unit 50 has a central processing unit (CPU) 51 and a main memory device 52, and performs information processing related to the control of the measuring device 20 and the information processing device 80. The main memory device 52 is a ROM or RAM. The auxiliary memory device 53 is a hard disk or the like. The communication interface 54 is a connection device to a local area network (LAN) or the Internet. The input device 55 is a touch panel, a keyboard, or the like. The output device 56 is a display or the like.

[0015] (infrared camera) The infrared camera 22 is a device that captures infrared images that contain infrared information. The infrared camera 22 can capture images in a time series. As shown in FIG. 2, the infrared camera 22 includes an optical system 23, a filter 24, and an image sensor 25. The optical system 23 is a component that refracts light and forms an image of the subject on the image sensor 25. The filter 24 is a component that passes infrared light of a specific wavelength. The filter 24 is disposed between the optical system 23 and the image sensor 25 and passes only infrared light of a specific wavelength from the light that passes through the optical system 23. The wavelength that is passed varies depending on the type of gas. For example, if methane is the target gas to be detected, the filter 24 is a bandpass filter that passes light with a wavelength between 3.2 μm and 3.4 μm. The image sensor 25 can be, for example, an indium antimonide (InSb) image sensor. The visible light camera 28 is a device that captures visible light images, which are images of visible light. Similar to the infrared camera 22, the visible light camera 28 can capture images in time series.

[0016] Next, the functional configuration of gas camera system 1 will be described with reference to Fig. 3. Fig. 3 is a block diagram illustrating the functional configuration of gas camera system 1. Below, a case where a gas leak test or a flow rate test is performed based on a user selection will be described. In gas leak inspections, a monitoring image that visualizes gas clouds is displayed on a screen, allowing the location and shape of the gas clouds and the relative concentration distribution to be observed on the screen. In gas flow rate inspections, the volume of gas flowing per minute, for example, can be obtained as a numerical value.

[0017] (Test information setting section) The test information setting unit 40 is a means for setting information associated with gas testing. The information associated with testing includes information on test conditions necessary for processing in the gas flow rate estimation unit 10, information specifying the test such as date, time, and location, as well as information on the purpose of the test, i.e., what test will be performed. The test information setting unit 40 can set the purpose of gas testing. This information is input by, for example, a user. The inspection information setting unit 40 sends information on inspection conditions such as the type of gas, the distance from the infrared camera 22 to the gas cloud, and the temperature to the gas flow rate estimation unit 10. This enables the gas flow rate estimation unit 10 to calculate the gas flow rate.

[0018] The test information setting unit 40 sends information about the purpose of the test to the thumbnail generation unit 30. That is, the test information setting unit 40 sets in the thumbnail generation unit 30 whether a gas leak test or a gas flow rate test will be performed. This allows the thumbnail generation unit 30 to generate thumbnails suitable for the purpose of the test, whether for a leak test or a flow rate test, using different methods depending on the purpose of the test. The information on the purpose of the test may be obtained by requesting the user to input whether to perform a gas leak test or a flow rate test, or may be determined by the test information setting unit 40 from the test that has been performed.

[0019] (Gas flow rate estimation unit) The gas flow rate estimation unit 10 is a means for estimating the gas flow rate from time-series infrared images of a gas cloud. The gas flow rate estimation unit 10 estimates the gas flow rate for each frame from time-series images of a gas cloud. The gas flow rate (L / min) is, for example, the volume (L) of gas passing through a cross section perpendicular to the screen per unit time (min) at 0°C and 1 atmosphere. Note that the time-series images of a gas cloud include time-series images of various formats, such as infrared images, high-sensitivity images, visible light images, and thermography images. Furthermore, the gas flow rate estimation unit 10 generates a monitoring image and extracts a gas region during its processing. The monitoring image is an image in which the gas cloud is visualized. The gas region is the region in which the pixels constituting the visualized gas cloud image are distributed. 3, the gas flow rate estimation unit 10 includes a monitoring image generation unit 11A, a gas region extraction unit 11B, a vector calculation unit 12, a transit time calculation unit 13, a concentration thickness product calculation unit 14, a gas amount calculation unit 15, and a flow rate calculation unit 16. Next, each component of the gas flow rate estimation unit 10 will be described. The transit time calculation unit 13 is also referred to as a first calculation unit, the gas amount calculation unit 15 as a second calculation unit, and the flow rate calculation unit 16 as a third calculation unit.

[0020] The monitoring image generating unit 11A is a means for visualizing gas clouds from time-series infrared images and generating monitoring images that facilitate monitoring of gas leaks. The monitoring image generating unit 11A receives time-series infrared images and outputs monitoring images. The monitoring image generating unit 11A performs, for example, the following processing. Temperature changes can be observed from time-series infrared images. The frequency of temperature changes where a gas cloud is present is higher than where a gas cloud is not present. The monitoring image generating unit 11A utilizes this to perform processing to remove low-frequency components. The infrared image is time-series image data, for example, 30 frames per second. The monitoring image generating unit 11A calculates a moving average of the brightness of each pixel over a predetermined number of frames, such as 21 frames, and calculates the difference from the infrared image. This processing enables gas cloud detection even when there is a large temperature change in the background. The monitoring image generating unit 11A can also perform processing to remove high-frequency components resulting from noise, etc. Figure 4 is a schematic diagram of a screen 60 that displays a gas cloud in a monitoring image superimposed on a visible light image. The processing by the monitoring image generating unit 11A is disclosed, for example, in Patent Document 1. Note that the technology disclosed therein is not necessarily required.

[0021] The gas region extraction unit 11B is a means for extracting a gas region. As an example, the gas region extraction unit 11B receives a time-series monitoring image as an input and outputs a gas region. The gas region extraction unit 11B performs the following process, for example. The gas region extraction unit 11B extracts the maximum brightness value of each pixel for each predetermined number of frames, such as 30 frames, of the time-series monitoring image. The gas region extraction unit 11B then generates an image in which the extracted maximum value is set for each pixel. The gas region extraction unit 11B further performs noise removal and binarization processing on the monitoring image to identify a gas region 72, as shown in FIG. 5A. The gas region 72 is time-series image data and can be displayed superimposed on an infrared image or the like.

[0022] The vector calculation unit 12 is a means for calculating the average motion vector of pixels. A motion vector is a vector that represents the movement of an object or the like on the screen by the direction and magnitude at the position of each pixel. Here, as an example, the vector calculation unit 12 receives the monitored image and the gas region as input, and outputs the average motion vector of each pixel in the gas region. The vector calculation unit 12 obtains a motion vector for each pixel corresponding to the gas region 72 in the monitoring image and calculates the average. The motion vector can be calculated using a known method such as template matching. Here, as shown in FIG. 5B, the average motion vector 91 of the gas region 72 is a vector in the direction in which the gas region 72 expands. Note that FIG. 5B exaggerates the pixels 62. The actual number of pixels in the gas region 72 is, for example, 5032 pixels, and the average motion vector is, for example, 25.5 pixels / sec. The average motion vector represents the degree to which gas is moving within the gas region 72.

[0023] The transit time calculation unit 13 is a first calculation unit, and is a means for estimating the time it takes for gas to pass through the gas region 72. The transit time calculation unit 13 receives the average motion vector and the gas region as input, and outputs the transit time. The transit time calculation unit 13 sets, for example, a rectangle on the screen that circumscribes the gas region 72 and has sides parallel to the average motion vector. Then, it calculates the transit time of the gas from the number of pixels corresponding to the length of the side of the rectangle and the number of pixels of the average motion vector. For example, if the number of pixels on the side of the rectangle is 172 pix and the average motion vector is 25.5 pix / sec, the transit time is 6.75 seconds.

[0024] The concentration thickness product calculation unit 14 is a means for calculating the concentration thickness product of a gas. The concentration thickness product of a gas is the product of the gas concentration and the gas thickness. Here, as an example, the concentration thickness product calculation unit 14 receives the monitored image and the gas region as input, and outputs the average value of the concentration thickness products of the pixels in the gas region. The calculation of the concentration-thickness product is disclosed in, for example, Japanese Patent No. 6344533. However, it is not necessary to rely on the technique disclosed therein.

[0025] Because gas clouds fluctuate due to wind and other factors, the intensity of infrared light detected by each pixel changes even when the infrared camera 22 is fixed. This makes it possible to capture changes in infrared intensity when gas is present and when it is not. For example, by determining in advance the correspondence between the proportion of infrared light absorbed by gas and the concentration-thickness product, the concentration-thickness product can be calculated from changes in infrared intensity. The concentration thickness product calculation unit 14 calculates the concentration thickness product of each pixel in the gas region and calculates the average value. The average value of the concentration thickness product is, for example, 0.285% LELm. The gas concentration can be expressed as a ratio to the LEL (Lower Explosion Limit) concentration, which is the lower explosion limit. The volume concentration of the lower explosion limit is expressed as 100% LEL. The concentration thickness product is multiplied by the depth direction as seen from the infrared camera 22, and expressed in units of % LELm. The lower explosion limit is the lowest concentration at which a flammable gas mixed with air will explode upon ignition. For example, in the case of methane, a volume concentration of 5% is 100% LEL, and 0.285% LELm is equivalent to 0.01425% m.

[0026] The gas amount calculation unit 15 is a second calculation unit and is a means for calculating the volume of gas. The gas amount calculation unit 15 receives the concentration-thickness product and the gas region as input, and outputs the gas amount of the gas region. The gas amount calculation unit 15 obtains the gas amount by multiplying the concentration-thickness product by the area of ​​the gas region. The area of ​​the gas region can be calculated by counting the number of pixels in the gas region 72 as shown in FIG. 5B and multiplying it by the corresponding area. The length corresponding to the width α of one pixel at the location of the gas cloud varies depending on the distance between the infrared camera 22 and the gas cloud. This distance can be determined using information input from the inspection information setting unit 40. The length corresponding to the pixel width α is, for example, a value such as 0.09219 m, and one pixel is 0.008499 m 2 The area of ​​the gas region with 5032 pixels is 42.77 m 2 If the concentration-thickness product is 0.0285%LELm (0.01425%m), the gas volume is 6.095L (0.006095m 3 ) is calculated as follows.

[0027] The flow rate calculation unit 16 is a third calculation unit and is a means for calculating the gas flow rate. The flow rate calculation unit 16 receives the gas volume and the gas transit time as input, and outputs the gas volume divided by the gas transit time as the gas flow rate (L / min). For example, if the gas volume is 6.095 L and the transit time is 6.75 seconds, the gas flow rate is calculated to be 54.2 L / min. The process from generating a monitoring image to calculating a gas flow rate is disclosed, for example, in Japanese Patent No. 6693609. However, it is not necessary to rely on the technology disclosed therein.

[0028] (Thumbnail generation section) The thumbnail generation unit 30 is a means for generating thumbnails to be added to the inspection data. The thumbnail generation unit 30 generates thumbnails suitable for the purpose of the inspection from time-series images of a gas cloud. The thumbnail generation unit 30 generates still images from the inspection data, which is video data. As shown in FIG. 3, the thumbnail generation unit 30 includes a representative value extraction unit 31, a frame selection unit 32, and an image selection unit 33. Here, a case where an information addition unit 34 is further included will be described. When the purpose of the inspection is to measure the gas flow rate, the thumbnail generation unit 30 generates thumbnails by selecting frames from time-series images of a gas cloud based on statistically processed values ​​of the gas flow rate. When the purpose of the inspection is to inspect for gas leaks, the thumbnail generation unit 30 generates thumbnails by selecting frames from time-series images of a gas cloud based on statistically processed values ​​of the number of pixels and average brightness of the gas cloud. For example, the thumbnail generation unit 30 may generate thumbnails by selecting frames from time-series images of a gas cloud in which the number of pixels and average brightness of the gas cloud are at their maximum, average, or median values. The thumbnail generation unit 30 may calculate an interval average as the average value for each predetermined interval from the time-series images, and select the maximum frame in the interval where this interval average is at its maximum value to generate thumbnails. The thumbnail generation unit 30 may calculate an interval average as the average value for each predetermined interval from the time-series images, and select the first frame in the interval where this interval average is at its maximum value to generate thumbnails.

[0029] The representative value extraction unit 31 is a means for extracting a representative value. The representative value may be, for example, a maximum value, an average value, or a median value. When the purpose of the test is to measure the flow rate of a gas, the representative value extraction unit 31 receives the gas flow rate from the flow rate calculation unit 16 as an input and outputs a representative value of the gas flow rate. In the case where the purpose of the inspection is a gas leak inspection, as an example, the representative value extraction unit 31 receives the monitoring image from the monitoring image generation unit 11A and the gas region from the gas region extraction unit 11B as inputs, and outputs the number of pixels of the gas cloud and a representative value of the average brightness value. The number of pixels of the gas cloud can be the number of pixels in the monitoring image whose brightness value is equal to or greater than a predetermined value. It may also be the number of pixels in the gas region. The average brightness value of the gas cloud can be the average brightness value of the pixels corresponding to the gas region in the monitoring image.

[0030] The frame selection unit 32 is a means for selecting a frame when a representative value is obtained. The frame when a representative value is obtained is the frame at the time when the representative value is obtained. The frame selection unit 32 receives the representative value as input and outputs information for identifying the frame. The information for identifying the frame is the time when the frame was captured and the frame number, which is the serial number of the frame. If there is no frame at the time when the representative value is obtained, the frame selection unit 32 can select a frame that has a value closest to the representative value.

[0031] The image selection unit 33 is a means for selecting an image to be combined with a gas cloud. Time-series images in a plurality of formats are sent to the image selection unit 33 from the image processing unit 26, which will be described later. The image selection unit 33 selects a time-series image in a format appropriate for the inspection purpose from these time-series images and combines it with the image of the gas cloud. For example, the image selection unit 33 can select a high-sensitivity image if the inspection purpose is to measure the gas flow rate, or a visible light image if the inspection purpose is to inspect for gas leaks. The image selection unit 33 receives input of time-series images in multiple formats and information specifying frames, and outputs selected images. The output of the image selection unit 33 becomes a thumbnail image. The thumbnail generation unit 30 can output the output of the image selection unit 33 as a thumbnail, and can also add information to the thumbnail image.

[0032] The information addition unit 34 is a means for adding predetermined information to thumbnail images. Information is added by processing the image so that the information is displayed on the image. The information addition unit 34 receives the thumbnail image and the information to be added as input, and outputs the thumbnail with the added information. The predetermined information is, for example, a representative value, and if the purpose is to measure the gas flow rate, it is possible to further add information such as inspection information used as a parameter for calculating the flow rate and information on the area on the screen that is the target of the calculation. Furthermore, if the purpose is to inspect for gas leaks, it is possible to display, for example, the location of the gas leak source.

[0033] As shown in Figure 6, a thumbnail 64 for gas leak detection combines a visible light image with an image of a gas cloud with the maximum area (number of pixels). A plus sign is added to indicate the location 74 of the gas leak source. This allows the user to easily identify the location of a gas leak in an actual gas facility. 7, when the purpose is to measure the gas flow rate, the thumbnail 65 is a high-sensitivity image when the gas flow rate is at its maximum, and includes a calculation region 76 for calculating the gas flow rate, and numerical values ​​78 for the gas concentration-thickness product and flow rate. This allows the user to easily grasp the position and shape of the gas cloud while checking the calculated numerical values.

[0034] The gas camera system 1 includes an image processing unit 26. The image processing unit 26 processes the captured images. The image processing unit 26 processes the captured infrared and visible light images to generate multiple types of time-series images, including high-sensitivity images and thermography images. Image 66 in FIG. 8A is an example of an infrared image, and image 67 in FIG. 8B is an example of a visible light image. A high-sensitivity image is, for example, the image of thumbnail 65 in FIG. 7, which is an image obtained by increasing the sensitivity of the monitoring image. A thermography image is, for example, image 68 in FIG. 8C, from which information corresponding to the temperature distribution can be obtained. Note that in FIG. 8C, areas with low brightness correspond to high temperatures. The image processing unit 26 combines these images with an image of a gas cloud extracted from the monitoring image to generate multiple types of time-series images.

[0035] The gas camera system 1 having the above configuration is equipped with an inspection information setting section that sets the inspection purpose of the gas, and is therefore able to generate thumbnails in a manner suited to each of the different inspection purposes of the gas. The gas camera system 1 can measure gas flow rates and perform gas leak inspections, and can generate different thumbnails for each piece of inspection data. By selecting an image that represents a representative value, such as the maximum value, average value, or median, and adding additional information, the contents of the inspection data can be easily understood at a glance, improving user convenience. The gas camera system 1 can capture a gas cloud as time-series images in a plurality of formats, select a time-series image format according to the inspection purpose, and generate thumbnails suitable for the inspection purpose from the time-series images.

[0036] The thumbnail generator 30 can perform externally specified statistical processing on the time-series images of the gas cloud, select frames based on the results, and generate thumbnails. The specified statistical processing type can be, for example, extracting a maximum value, an average value, or a median value. When such a statistical processing type is specified in the gas camera system 1, the representative value extractor 31 performs the processing specified, and the frame selector 32 can select a frame corresponding to the specified representative value. Furthermore, if the thumbnail generation unit 30 is unable to select a frame suitable for the inspection purpose specified by the user, it can select a frame suitable for another inspection purpose and generate a thumbnail. For example, there may be cases where data suitable for processing corresponding to the inspection purpose does not exist due to a malfunction of the equipment at the time of shooting. In such cases, the thumbnail generation unit 30 generates thumbnails using available data. For example, if an inspection is performed to measure gas flow rate but the calculation is not performed correctly, the gas camera system 1 can generate a thumbnail for a gas leak inspection. The gas region extraction unit, vector calculation unit, concentration thickness product calculation unit, and representative value extraction unit have been described as examples in which a monitoring image is used as input, but the data used does not have to be a monitoring image; for example, data containing information on the infrared intensity at each pixel and the brightness of the gas cloud can be used. The gas camera system 1 may be configured with a plurality of measuring devices 20 and a plurality of information processing devices 80.

[0037] (Variation) Next, some modified examples of the gas camera system 1 will be described. In the first modification, the thumbnail generating unit 30 generates a thumbnail by selecting a frame at the timing when the gas erupts and the gas flow rate reaches its maximum from the time-series images of the gas cloud.

[0038] There is a time lag between the timing at which the gas flow rate is calculated and the timing at which the gas erupts. As illustrated in FIG. 9A, the time at which the gas flow rate reaches its maximum is often the time immediately after a gas cloud 70 passes through a flow rate calculation area 76. If this time is used as a thumbnail, the gas cloud will be far away from the leak source position 74, making it difficult to see the state of the gas leak, especially when the gas erupts intermittently. To address this issue, the thumbnail generation unit 30 calculates backwards to the time at which the gas with the maximum flow rate recorded erupted from the leak source, and uses the image at that time as the thumbnail. Specifically, the gas flow velocity is calculated from the mean motion vector, and the time it takes for the gas to actually flow from the leak source is calculated by dividing the distance from the leak source to the outer frame of the calculation domain 76 by the flow velocity using that flow velocity and the leak source position 74. The image at the time calculated by subtracting the calculated time from the time the maximum flow rate was detected is then used as the thumbnail. As a result, the generated thumbnail is an image of the time when gas cloud 70 erupted from the leak source indicated by position 74, as shown in FIG. 9B.

[0039] In the second modification, the thumbnail generator 30 calculates an interval average as the average value for each predetermined interval from time-series images of a gas cloud, and selects the frame with the maximum value in the interval where the interval average is maximum to generate a thumbnail. Note that the second modification may be used for either gas flow rate or gas leak, but the case of gas flow rate will be described here. In gas camera systems, flow rates are calculated by dividing the time into certain intervals to prevent the influence of noise, etc. In such cases, selecting one flow rate value only selects the interval and does not determine the time, so it is necessary to further determine the time to create a thumbnail. The predetermined section can be defined by time or number of frames. As illustrated in FIG. 10, section averages β and γ are calculated for predetermined sections 93 and 94. For example, if the section average γ for section 94 is the maximum value, one frame from this section 94 may be selected. The frame selection unit 32 may select, for example, the maximum frame or the median frame from this section 94. Alternatively, the frame may be the first frame, the middle frame, or the last frame of section 94. Note that the predetermined sections 93 and 94 may be sections shorter than 5 seconds, or may be sections equivalent to 5 seconds.

[0040] In a third modification, when the purpose of the inspection is a gas leak inspection, the thumbnail generating unit 30 generates thumbnails by selecting frames from time-series images of a gas cloud at timings when the gas leakage source is easily identifiable. Preferably, the thumbnail generating unit 30 generates thumbnails by selecting frames in which the shape of a gas cloud diffusing from the leakage source is captured. The location of the gas leak source is important information from the results of a gas leak inspection, and it is effective to leave an image as a thumbnail that clearly shows the location of the leak source. Gas is affected by wind and diffuses. When the wind is constant, it diffuses from the leak source in a constant direction at a constant rate, and the shape of the gas cloud becomes an isosceles triangle 75, as shown in screen 69 in Figure 11. On the other hand, when the wind is variable, the shape becomes complex, making it difficult to identify the leak source from the screen. Therefore, whether the shape of the gas cloud is close to an isosceles triangle is calculated using similarity, and the timing with the highest score is selected as the thumbnail. Specifically, for example, machine learning can be used to learn patterns that make it easy to identify the location of the leak source and select the thumbnail.

[0041] In the fourth modified example, the thumbnail generation unit 30 further includes a thumbnail storage unit 35. The thumbnail storage unit 35 is a means for storing generated thumbnails. In the fourth modified example, the thumbnail generation unit 30 generates thumbnails for the same time-series image data, for example, for both gas flow rate measurement and gas leak inspection. In addition, multiple thumbnails generated by the gas camera system already described are stored in the thumbnail storage unit 35. The thumbnail storage unit 35 then selects and displays a thumbnail that corresponds to the user's search criteria from the stored thumbnails. The thumbnail generation unit 30 may perform all of the thumbnail generation already described. The thumbnail storage unit 35 displays an appropriate thumbnail from among multiple thumbnails according to the criteria selected by the user when displaying the data. When operating the gas camera system, the user may search for desired test data from a large amount of test data by sorting, filtering, etc. In this case, the thumbnail corresponding to the item used by the user for sorting, filtering, etc. is displayed. For example, if the user sorts or filters by flow rate value, thumbnails generated from the flow rate value may be displayed, and if the user sorts or filters by the presence or absence of a gas leak, thumbnails generated from the number of pixels or brightness value of the gas cloud may be displayed.

[0042] [Thumbnail generation method] Next, several examples of a method for generating thumbnails of gas clouds according to an embodiment will be described with reference to FIGS. 12 to 15. The method for generating thumbnails of gas clouds includes step S30 in which the inspection information setting unit 40 sets the purpose of gas inspection, and steps S40, S40A, S50, S50A, and S50B in which the thumbnail generating unit generates thumbnails suitable for the inspection purpose from time-series images of the gas cloud. Note that the description here also includes the processing of the entire gas camera system. Steps S1 to S3 are the processing of the entire gas camera system.

[0043] 12, first, inspection information including the inspection purpose input by the user is input (step S10), and the gas camera system 1 executes the inspection (step S20). Then, the inspection information setting unit 40 sets the inspection purpose of the gas (step S30), and determines whether the inspection purpose is gas flow rate measurement (step S35). In step S35, if the purpose of the inspection is to measure the gas flow rate (Yes in step S35), the representative value extraction unit 31 extracts a representative value of the flow rate (step S41). Next, the frame selection unit 32 selects a frame that will be the representative value (step S43). Note that the frame is identified by time here. The image selection unit 33 then selects the frame of the high-sensitivity image at that time as a thumbnail image (step S45), and the information addition unit 34 adds information such as the flow rate value and the gas flow rate calculation region to the image and stores it together with the inspection data (step S47), at which point processing S1 ends.

[0044] In step S35, if the inspection purpose is a gas leak inspection (No in step S35), the representative value extraction unit 31 extracts a representative value of the area (number of pixels) or brightness of the gas cloud (step S51). Next, the frame selection unit 32 selects a frame that will be the representative value (step S53). Then, the image selection unit 33 selects the frame of the monitoring image at that time as a thumbnail image (step S55), and the information addition unit 34 adds information such as the gas leak location to the image and stores it together with the inspection data (step S57), and processing S1 ends. Note that step S40, which is the process from step S41 to step S47 in process S1, and step S50, which is the process from step S51 to step S57, are processes for generating thumbnails suitable for the purpose of inspection according to the embodiment.

[0045] As shown in FIG. 13, in step S20 where an inspection is performed, the following steps S21 to S28 are performed. First, the measurement device 20 captures time-series images of the gas cloud (step S21). The monitoring image generation unit 11A generates a monitoring image from the time-series images of the gas cloud (step S22). The gas region extraction unit 11B extracts the gas region (step S23). Next, the vector calculation unit 12 calculates the average motion vector of pixels based on the gas region extracted by the gas region extraction unit 11B (step S24). The concentration thickness product calculation unit 14 calculates the average value of the gas concentration thickness product in the gas region (step S25). The transit time calculation unit 13, which is a first calculation unit, calculates the gas velocity in the gas region and the gas transit time required for the gas to pass through the gas region using time-series images of the gas cloud captured by the infrared camera (step S27). The gas amount calculation unit 15, which is a second calculation unit, calculates the area at the actual position of the gas cloud corresponding to the number of pixels in the gas region, and calculates the amount of gas in the gas region using the area and the average value of the gas concentration thickness product (step S26). Then, when the flow rate calculation unit 16, which is the third calculation unit, calculates the estimated gas flow rate value using the gas transit time and the gas amount (step S28), the processing in Fig. 13 ends. Note that the processing in step S27 for calculating the gas transit time can be performed after the processing in step S24 for calculating the mean movement vector and before step S28 for calculating the estimated gas flow rate value.

[0046] In process S2 of the gas camera system illustrated in Fig. 14, step S40 is changed to step S40A, and step S50 is changed to step S50A. Step S40A is a case where, in measuring the flow rate, the time when gas with the maximum flow rate was ejected is calculated. Step S50A is a case where, in gas leak testing, a representative value of the interval average is extracted and the first frame of the interval is selected. Note that steps S10 to S35 are the same as those in process S1 described above, and therefore their explanation will be omitted. In step S40A, the representative value extraction unit 31 extracts the maximum value of the flow rate (step S41). Next, the frame selection unit 32 calculates the time when the gas with the maximum flow rate erupted (step S44). The image selection unit 33 then selects the frame of the high-sensitivity image at that time as a thumbnail image (step S45). The information addition unit 34 adds information such as the flow rate value and the calculation region of the gas flow rate to the image and stores it together with the inspection data (step S47), and processing S2 ends.

[0047] Also, in step S50A, the representative value extraction unit 31 calculates the interval average for the area or brightness of the gas cloud and extracts the interval where the maximum value is obtained (step S52). Next, the frame selection unit 32 calculates the time of the first frame in the interval where the maximum value is obtained (step S54). The image selection unit 33 then selects the frame of the monitoring image at that time as a thumbnail image (step S55), and the information addition unit 34 adds information such as the gas leak location to the image and stores it together with the inspection data (step S57), at which point processing S2 ends. The process of extracting the representative value of the section average in step S50A can be similarly carried out in measuring the flow rate of gas.

[0048] In the process S3 of the Cass camera system shown in Figure 15, step S50A is further changed to step S50B. Step S50B is a case where a frame is selected at a timing that makes it easy to identify the leak source from the shape of the gas cloud during gas leak testing. Note that the explanation of step S40A is also omitted. In step S50B, frame selection unit 32 selects a frame at a time when the gas leak source can be easily identified (step S58). Image selection unit 33 selects the frame at that time of the monitoring image as a thumbnail image (step S55). Information addition unit 34 adds information such as the gas leak location to the image and stores it together with the inspection data (step S57), and processing S3 ends. There are many other variations of the thumbnail generation method besides the one described above.

[0049] [Thumbnail generator] The program according to the embodiment is a program that is loaded into a computer to cause the computer to execute a thumbnail generation process. The thumbnail generation program can be obtained via a telecommunications line and can be recorded on a computer-readable recording medium.

[0050] It should be noted that the present invention is not limited to these embodiments and modifications.

[0051] Effect of this embodiment The gas camera system 1 of this embodiment can generate thumbnails suitable for gas inspection purposes. Furthermore, the gas camera system 1 can generate thumbnails using criteria suitable for flow rate estimation, allowing users to confirm the highest flow rate and the average flow rate when there is a lot of noise. The gas camera system 1 may select and display a representative value from a time series of flow rate values. In this case, by selecting frames using the same criteria, users can confirm thumbnails that match the representative value.

[0052] When calculating the flow rate as an average value over a certain period, the gas camera system 1 can confirm the state of the gas cloud at the moment when the gas with the maximum flow rate erupts from the leak source. The gas camera system 1 can easily identify the source of a gas leak during a gas inspection, and can handle cases where data suitable for the purpose is not stored. The gas camera system 1 can also generate thumbnail images that make it easier to understand the situation. This gas camera system 1 can present suitable thumbnails for the same inspection data depending on the purpose selected by the user when displaying the data. [Explanation of symbols]

[0053] 1 Gas Camera System 10 Gas flow rate estimation unit 11A Surveillance image generation section 11B Gas region extraction section 12 Vector calculation section 13 Passage time calculation section 14 Concentration thickness product calculation unit 15 Gas volume calculation unit 16 Flow rate calculation section 20 Measuring equipment 22 Infrared camera 23 Optical system 24 filters 25 Image sensor 26 Image processing section 28 Visible Light Camera 30 Thumbnail generation section 31 Representative value extraction section 32 Frame selection section 33 Image selection section 34 Information Addition Section 40 Examination information setting section

Claims

1. an inspection information setting unit for setting the inspection purpose of the gas; a thumbnail generation unit that generates thumbnails suitable for the inspection purpose from time-series images of the gas cloud; A gas camera system comprising:

2. a gas flow rate estimation unit that estimates a gas flow rate in each frame from time-series images of the gas cloud; 2. The gas camera system according to claim 1, wherein when the inspection purpose is measurement of a gas flow rate, the thumbnail generation unit selects frames from the time-series images based on values ​​obtained by statistically processing the flow rate and generates thumbnails.

3. 3. The gas camera system according to claim 2, wherein said thumbnail generating section selects a frame when said flow rate is at a maximum value from said time-series images and generates a thumbnail.

4. 3. The gas camera system according to claim 2, wherein said thumbnail generating section selects, from said time-series images, any of frames when said flow rate is at an average value, and generates a thumbnail.

5. 3. The gas camera system according to claim 2, wherein said thumbnail generating section selects a frame when said flow rate is a median value from said time-series images and generates a thumbnail.

6. 4. The gas camera system according to claim 3, wherein said thumbnail generating section selects a frame from said time series images at a timing when gas is ejected at a maximum flow rate, and generates a thumbnail.

7. 3. The gas camera system according to claim 2, wherein the thumbnail generation unit calculates an interval average as an average value for each predetermined interval from the time series images, and selects a frame with a maximum value in an interval in which the interval average is maximum to generate a thumbnail.

8. 3. The gas camera system according to claim 2, wherein the thumbnail generation unit calculates an interval average as an average value for each predetermined interval from the time-series images, and selects the first frame of the interval in which the interval average is maximum to generate a thumbnail.

9. 3. The gas camera system according to claim 2, wherein the thumbnail generation unit calculates an interval average as an average value for each predetermined interval from the time-series images, and selects an intermediate frame in the interval in which the interval average is maximum to generate a thumbnail.

10. 3. The gas camera system according to claim 2, wherein the thumbnail generation unit calculates an interval average as an average value for each predetermined interval from the time-series images, and selects a median frame of the interval in which the interval average is maximum to generate a thumbnail.

11. 3. The gas camera system according to claim 2, wherein said thumbnail generating section performs an externally designated type of statistical processing on said time-series images, selects frames based on the results, and generates thumbnails.

12. 2. The gas camera system according to claim 1, wherein when the inspection purpose is a gas leak inspection, the thumbnail generating unit generates thumbnails by selecting frames from the time series images at timings at which a gas leak source is easily identified.

13. 13. The gas camera system according to claim 12, wherein said thumbnail generating section generates a thumbnail by selecting a frame from said time series images in which a shape of a gas cloud diffusing from a leakage source is captured.

14. 2. The gas camera system according to claim 1, wherein said thumbnail generating section, if unable to select a frame suitable for an inspection purpose designated by a user, selects a frame suitable for another inspection purpose and generates a thumbnail.

15. the gas cloud is captured as a time series of images in multiple formats; 2. The gas camera system according to claim 1, wherein said thumbnail generating section selects time-series images in a format appropriate for said inspection purpose, and generates thumbnails suitable for said inspection purpose from said time-series images.

16. 2. The gas camera system according to claim 1, wherein said thumbnail generating section adds predetermined information to the generated thumbnail.

17. the thumbnail generating unit further comprises a thumbnail storage unit for storing thumbnails generated by the gas camera system according to any one of claims 1 to 16, The thumbnail storage unit selects and displays a thumbnail that corresponds to a user's search criteria from among the stored thumbnails.

18. an inspection information setting unit setting an inspection purpose for the gas; a step of generating thumbnails suitable for the inspection purpose from time-series images of the gas cloud by a thumbnail generating unit; A method for generating a thumbnail of a gas cloud comprising:

19. On the computer, Procedures for setting gas inspection objectives; a step of generating thumbnails suitable for the inspection purpose from time-series images of the gas cloud; A program to execute.

Images